| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ENOLAB – Laboratori de Microbiologia Enològica, Departament de Microbiologia i Ecologia, Facultat de Ciències Biològiques, Universitat de València, Dr Moliner 50, 46100 Burjassot, València, Spain
Correspondence
I. Pardo
Isabel.Pardo{at}uv.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The GenBank/EMBL/DDBJ accession numbers for the 16S rRNA sequences of strains 8, 59b, 68, 71, 88, 116, 154, 166, 203 and 459 determined in this study are AY681126–AY681135, respectively.
Additional dendrograms, examples of RAPD profiles and the 16S rRNA gene sequence similarity matrix are available as supplementary material in IJSEM Online.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
). Lactobacilli are found in a great variety of habitats (Hammes et al., 1991). The genus Lactobacillus consists of more than 110 species with validly published names (Euzéby, 1997). They are genetically quite diverse; their G+C content ranges from 32 to 54 mol%. This shows that Lactobacillus is not a well-defined genus (Vandamme et al., 1996), as has been demonstrated by the transfer of some Lactobacillus species to newly created genera: Carnobacterium (Collins et al., 1987), Atopobium (Collins & Wallbanks, 1992), Weissella (Collins et al., 1993) and Paralactobacillus (Leisner et al., 2000). These reorganizations are mainly supported by DNA–DNA reassociation studies and 16S rRNA gene sequence data (Collins et al., 1991; Schleifer & Ludwig, 1995).
The classical division of the lactobacilli is based on their fermentative characteristics: (i) obligately homofermentative, (ii) facultatively heterofermentative and (iii) obligately heterofermentative. Identification to the species level by simple phenotypic tests may sometimes be difficult because most of the bacteria have very similar nutritional requirements and grow under similar environmental conditions (Vandamme et al., 1996).
The development of molecular biological methods such as cellular fatty acids analysis (Rizzo et al., 1987), whole-cell protein analysis (Dicks & van Vuuren, 1988; Patarata et al., 1994), DNA–DNA hybridization (Schillinger et al., 1989; Lonvaud-Funel et al., 1991; Sohier & Lonvaud-Funel, 1998), electrophoretic patterns of peptidoglycan hydrolases (Lortal et al., 1997) and ribotyping (Zhong et al., 1998; Motoyama et al., 2000) have proved useful for identification of some species of lactobacilli. Among PCR-based methods, species-specific primers (Sohier et al., 1999; Chagnaud et al., 2001), internal rrn spacer region primers (Berthier & Ehrlich, 1998; Nour, 1998; Dubernet et al., 2002), random amplification of polymorphic DNA (RAPD) (Cocconcelli et al., 1995; Daud Khaled et al., 1997; Nigatu et al., 2001), analysis of repetitive bacterial DNA elements (rep-PCR) (Gevers et al., 2001), tDNA-PCR (Baele et al., 2002) and amplified 16S rDNA restriction analysis (16S-ARDRA) (Ventura et al., 2000; Rodas et al., 2003) have been employed.
In the present work, we have applied a set of molecular and phenotypic techniques to characterize wine lactobacilli, comparing them with reference strains in a polyphasic approach. The species found in wine are Lactobacillus brevis, Lactobacillus buchneri, Lactobacillus casei, Lactobacillus collinoides, Lactobacillus coryniformis subsp. coryniformis, Lactobacillus fermentum (=Lactobacillus cellobiosus; Dellaglio et al., 2004), Lactobacillus curvatus, Lactobacillus delbrueckii, Lactobacillus fructivorans (=Lactobacillus trichodes; Weiss et al., 1983; Vandamme et al., 1996), Lactobacillus hilgardii, Lactobacillus kunkeei, Lactobacillus mali, Lactobacillus nagelii, Lactobacillus plantarum and Lactobacillus paracasei subsp. paracasei (Edwards et al., 1998, 2000; Rodas et al., 2003). Reference strains of these species and other closely related species (Schleifer & Ludwig, 1995) were included in the analysis. Taxonomic dendrograms were deduced from the comparisons and taxonomic relationships were established. We also evaluated the discriminatory power of each individual method (RFLP-PFGE, RAPD, 16S-ARDRA, ribotyping) for identification and typing of 178 lactobacilli isolated from grape must and wine.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
. All strains were grown in MRS broth (Scharlab) supplemented with L-cysteine hydrochloride (0·5 g l–1) at 28 °C. Cultures were maintained frozen at –20 °C in 20 % v/v glycerol and lyophilized.
|
Wine strains were purified on MRS agar and characterized as strains of Lactobacillus on the basis of their Gram-positive character, absence of catalase activity and production of lactic acid from hexose fermentation (Cappuccino & Sherman, 1987). Analysis of the final products of sugar catabolism was done by HPLC as described by Frayne (1986). Each strain was examined microscopically for cellular morphology and arrangement and for motility. We determined the type of glucose fermentation following the procedure described by Zúñiga et al. (1993).
Phenotypic identification was performed according to Hammes et al. (1991), who propose an range of different tests for identifying homofermentative or heterofermentative lactobacilli. The phenotypic traits studied were: (i) growth at 15 and 45 °C in MRS broth, (ii) production of acid from different carbohydrates in MRS fermentation broth (Scharlab) and (iii) the optical nature of the isomer of lactate formed, determined by HPLC and L-lactic kit (cat. no. 139084; Roche).
RFLP-PFGE.
To perform this technique, DNA must be extracted in an agarose matrix to prevent mechanical DNA breakage. Genomic DNA in agarose plugs was prepared by a modification of the method of Matushek et al. (1996). Bacterial strains were grown in MRS to an OD600 of 0·5, cells were harvested from 2·5 ml culture by centrifugation and resuspended in 0·5 ml 2x lysis buffer [12 mM Tris/HCl (pH 7·6), 2 M NaCl, 200 mM EDTA (pH 8·0), 1 % w/v Brij 58, 0·4 % w/v deoxycholate, 1 % w/v sodium lauroyl sarcosine] containing 1 mg lysozyme ml–1 (Sigma), 60 µg RNase A ml–1 (DNase-free; Sigma) and 40 U mutanolysin ml–1 (Sigma). Cell suspensions were mixed with 0·5 ml of 1·5 % w/v low-melting-point agarose (SeaPlaque GTG agarose; FMC) and pipetted into a plug mould (syringe of 1 ml). Solidified plugs were cut with a sharp blade forming slices of 1 mm thickness, which were incubated for 12 h at 37 °C in 6 ml of 1x lysis buffer, containing 5 mg lysozyme ml–1, 30 µg RNase A ml–1 and 10 U mutanolysin ml–1. After incubation, the lysis solution was replaced with 6 ml ES [0·5 M EDTA (pH 9·0), 1 % w/v sodium lauroyl sarcosine] containing 100 µg proteinase K ml–1 (Boehringer Mannheim) and 1 % SDS, and incubated 6 h at 50 °C. The slices were washed three times with 10 ml TE 10/0·1 buffer [10 mM Tris/HCl (pH 8·0), 0·1 mM EDTA (pH 8·0)] and incubated with 10 ml TE 10/0·1 buffer for 1 h at 50 °C. The solution was replaced with 10 ml fresh buffer and then incubated at room temperature for 1 h. Finally, the slices were immersed in 10 ml TE 10/100 buffer [10 mM Tris/HCl (pH 8·0), 100 mM EDTA (pH 8·0)] for storage at 4 °C. Under these conditions, discs were stable for several months, without DNA degradation or apparent changes in quality. Before digestion, the slices were washed twice in TE 10/0·1 buffer for 1 h and equilibrated for 1 h in the appropriate restriction enzyme buffer. Single digestions with SfiI, NotI and SmaI were performed for 27 h at 50, 37 and 25 °C, respectively. Electrophoresis was carried out in a CHEF DR II apparatus (Bio-Rad) in 1 % w/v SeaKem GTG agarose (FMC) with 0·5x TBE buffer (45 mM Tris/HCl, 45 mM boric acid and 1 mM EDTA, pH 8·0) at 15 °C. A constant voltage (200 V) was applied to the system and fragment separation was performed using a two-phase program. To separate the SfiI and NotI fragments of obligately heterofermentative lactobacilli, a pulse time was applied from 0·5 to 20 s for 12 h and then another from 30 to 60 s for 8 h. For homofermentative and facultatively heterofermentative lactobacilli digested with NotI, a pulse time was used from 0·5 to 5 s for 12 h and then another from 5 to 10 s for 8 h. Electrophoretic conditions for separating the SmaI fragments in all lactobacilli were a pulse time from 0·5 to 5 s for 10 h and then another from 0·5 to 10 s for 4 h. MidRange PFG marker I or LowRange PFG marker (New England BioLabs) were used as molecular size standards. Agarose gels were stained with ethidium bromide (0·5 µg ml–1) and images were digitized with a GelPrinter Plus system (TDI).
16S-ARDRA.
This technique was applied following the protocol described by Rodas et al. (2003). Template DNA was prepared from cell suspensions in water; primers pA and pH were used to amplify 16S rRNA genes as already described (Edwards et al., 1989). Amplified fragments were quantified and then separately digested with three enzymes (BfaI, MseI and AluI). Digestion products were separated by conventional gel electrophoresis using a 1 Kb plus ladder (Invitrogen, BRL) as a molecular size standard. Electrophoresis conditions were previously described by Rodas et al. (2003). Staining and digitization of images were done as described in the RFLP-PFGE method.
RAPD.
DNAs used for random amplifications were obtained from agarose slices for RFLP-PFGE purposes. One slice was cut into four pieces with a sterile sharp blade, a quarter was suspended in 20 µl of distilled water and then heated at 94 °C for 5 min to melt the agarose and 1 µl of this solution was used in the PCR mixture.
Three random primers were used separately in this study, 16R and 17R described by Tompkins et al. (1996), and COC described by Cocconcelli et al. (1995). DNA amplification was carried out in a 50 µl PCR mixture containing 200 µM dNTPs, 1 µM selected random primer, 10 mM Tris/HCl (pH 8·8), 50 mM KCl, 0·1 % Triton X-100, 2 mM MgCl2, 1·5 U DyNazyme II DNA polymerase (Finnzymes) and 1 µl template DNA prepared as described above. PCR was performed in a Techne Thermal cycler (Techne). The amplification conditions for all primers used were those described by Tompkins et al. (1996) but we included a ramp rate of 20 °C min–1.
RAPD products were resolved by electrophoresis in 1·2 % (w/v) SeaKem LE agarose (FMC) in 0·5x TBE gels and a 1 Kb plus ladder was used as a molecular size standard. Staining and digitization of images were done as described in the RFLP-PFGE method.
Ribotyping.
Ribotyping was performed using the RiboPrinter microbial characterization system (Qualicon). DNA was obtained from colonies grown on MRS agar at 28 °C for 2–4 days. Ribotypes were obtained as reported in the manufacturer's protocol and EcoRI was used to restrict the DNA.
Data analysis.
Digitized images of all molecular techniques and phenotypic traits were introduced on BioNumerics software version 2.5 (Applied Maths), which carried out conversion, normalization and further analysis of the patterns. Phenotypic profiles were analysed using the simple matching similarity coefficient and UPGMA. PFGE patterns with SfiI and NotI and ARDRA patterns were grouped with the UPMGA clustering method using the band-based Dice similarity coefficient. SmaI PFGE patterns, RAPD and ribotypes were analysed using Pearson's product moment correlation coefficient and the UPMGA clustering method. The level of reproducibility of each experiment was calculated from duplicate experimental profiles of 10 randomly selected strains. The overall data for each strain obtained from each technique were combined on the BioNumerics software, maintaining the same coefficients used for single pattern analysis and clustered by using the UPMGA method.
Diversity indexes were calculated using the DIVERS online informatic program (F. J. Pérez-López and F. M. Sola-Fernández; available at http://perso.wanadoo.es/jp-l/descargas.htm).
Amplification and sequencing of the complete 16S rRNA gene, and sequence data analysis.
Complete 16S rRNA genes were amplified as described for the 16S-ARDRA technique. The primers used for complete sequencing of the 16S rRNA gene were: pA, pH (Edwards et al., 1989), I2B (5'-CCCTACTGCTGCCTCCCGTAGGAGT-3'), S1i (5'-GGACTACCAGGGTATCTA-3') and S2d (5'-AGTGGAACTCCATGTGT-3'). Amplification products were purified using the UltraClean PCR clean-up kit (Mobio) and sequenced using an ABI-PRISM 377 (PE-Applied Biosystems) automated sequencer and the ABI PRISM BigDye Terminator cycle sequencing ready reaction kit (PE-Applied Biosystems).
A total of 11 strains (8, 59b, 68, 71, 88, 116, 154, 166, 203, 449 and 459) were sequenced. 16S rRNA gene sequences of about 1500 bp were compared with the full dataset of about 22 000 reference sequences of the ARB database (http://www.arb-home.de/) using the program package ARB (Ludwig et al., 2004). This database contains 250 almost full-length sequences corresponding to LAB. The respective ARB tools were used for automated sequence alignment. The alignment was checked visually and corrected manually using the sequence editor ARB_EDIT. Sequence similarity values were calculated by comparing nucleotides at the corresponding positions. A tree was derived from the corrected distance values by using the neighbour-joining method (Saitou & Nei, 1987). Maximum-parsimony- and maximum-likelihood-based trees, respectively using the full dataset or a selection of about 28 sequences, were obtained and compared in order to make an estimate of the confidence. The effects on the results of a base-frequency filter in the sequence comparison analysis were also evaluated.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The dendrogram constructed by numerical analysis of the phenotypic profiles of the heterofermentative lactobacilli is available as Supplementary Fig. A (dendrogram I) in IJSEM Online. Above 93 % similarity (S), 10 clusters could be delineated, named He1–He10, six of which were formed only by one reference strain (He5–He10). Cluster He1 comprised 20 wine strains, L. brevis CECT 216 and L. collinoides CECT 922T. Cluster He2 comprised 48 wine strains and L. hilgardii NCFB 264T. Cluster He3 and He4 comprised wild strains without any reference strain. The reproducibility level was 100 %S. The calculated cophenetic correlation value for the phenotypic profiles was 0·94, indicating an excellent level of reliability.
The dendrogram that was obtained after numerical analysis of the phenotypic profiles of homofermentative and facultatively heterofermentative lactobacilli is available as Supplementary Fig. A (dendrogram II) in IJSEM Online. Above 86 %S, 15 clusters could be delineated, named Ho1–Ho15, seven of which were formed only by one reference strain. Cluster Ho1 comprised one wine strain and L. coryniformis subsp. coryniformis CECT 982T. Cluster Ho3 comprised 25 wine strains and L. mali CECT 4149. Cluster Ho4 comprised Lactobacillus delbrueckii subsp. bulgaricus CECT 4005T and Lactobacillus helveticus CECT 4305T. Cluster Ho7 comprised nine wine strains and L. casei ATCC 334. Cluster Ho8 comprised six wine strains and two reference strains, Lactobacillus zeae ATCC 15820T and L. casei CECT 475T. Cluster Ho9 comprised 10 wine strains without reference strains. Cluster Ho12 comprised 32 wine strains and four reference strains: L. plantarum CECT 748T, Lactobacillus rhamnosus CECT 278T, L. paracasei subsp. paracasei CECT 4022T and L. nagelii DSM 13675T. Cluster Ho14 comprised only two wine strains. The reproducibility level was 100 %S. The calculated cophenetic correlation value for the phenotypic profiles was 0·83, indicating a good level of reliability.
All the wine strains that grouped without or with more than one reference strain were named as Lactobacillus spp. The two reference strains of L. casei were placed in different clusters: the type strain CECT 475T was in the same group as the type strain of L. zeae, ATCC 15820T. This incongruence in the results prevents the identification of wild strains grouped in clusters Ho7 and Ho8. These phenotypic traits allowed us to identify only three species of wine lactobacilli: L. hilgardii (48 strains), L. coryniformis subsp. coryniformis (1 strain) and L. mali (25 strains).
Numerical analysis of RFLP-PFGE patterns
Single digestions of DNA were performed with SfiI, NotI and SmaI, respectively yielding 1–30 fragments of 15–1000 kb, 1–35 fragments of 15–800 kb and more than 35 fragments of 4–100 kb. The level of reproducibility was 100 %S for digestion with SfiI and NotI and 92·07 %S for SmaI. As the SfiI and NotI fragments of L. delbrueckii subsp. bulgaricus CECT 4005T, L. delbrueckii subsp. delbrueckii CECT 286T, L. delbrueckii subsp. lactis CECT 282, L. fermentum CECT 4007T and L. cellobiosus CECT 562 were of approximately 4–97 kb, the electrophoretic conditions used to separate them were the same as for SmaI fragments.
Based on a computerized numerical analysis of combined RFLP-PFGE patterns, we grouped the strains into 22 clusters at a similarity level of 50 %S (Supplementary Fig. B in IJSEM Online). This technique did not enable the assignment of isolated strains to species because the majority of clusters contained none or more than one reference strain; only cluster M5 included 23 wine strains and L. hilgardii NCFB 264T. The global cophenetic correlation value for the PFGE was 0·89, indicating a good level of reliability. Analyses of the single PFGE-RFLP patterns with SfiI, NotI and SmaI are not shown, but the results were concordant with those obtained using the combined analysis.
The typing power of RFLP-PFGE was high; analysis of digestion bands obtained with SfiI, NotI and SmaI revealed 130, 118 and 130 distinct patterns, respectively, from 178 isolated Lactobacillus strains. One hundred and forty-three distinct patterns were found after combined numerical analysis; 122 strains showed a specific restriction pattern, whereas 56 strains were grouped in 21 different patterns. Strains sharing the same pattern came from the same sample, leading us to think that they represented the same strain isolated several times. Ferrero et al. (1996), Zapparoli et al. (1998) and Ventura & Zink (2002) reported discriminatory, clear and reproducible RFLP-PFGE patterns in various Lactobacillus species, which allowed its use for typing studies. Our results agree with their observations.
16S-ARDRA fingerprinting analysis
Based on a computerized numerical analysis of combined BfaI and MseI 16S-ARDRA patterns, we grouped the strains into 26 clusters at a similarity level of 98 % (Supplementary Fig. C in IJSEM Online). Cluster A1 comprised L. zeae ATCC 15820T and L. casei CECT 475T. Cluster A2 contained 13 wine strains, L. paracasei subsp. paracasei CECT 4022T and L. casei ATCC 334. Cluster A4 comprised Lactobacillus reuteri CECT 925T and Lactobacillus oris CECT 4021T. Cluster A5 comprised two wine strains and L. collinoides CECT 922T. Cluster A8 comprised 10 wine strains and L. mali CECT 4149. Cluster A12 comprised 71 wine strains and L. hilgardii NCFB 264T. Cluster A14 comprised 36 wine strains, L. plantarum CECT 748T and Lactobacillus pentosus CECT 4023T. Cluster A17 comprised one wine strain and L. coryniformis subsp. coryniformis CECT 982T. Cluster A18 comprised 19 wine strains and L. brevis CECT 216. Cluster A25 contained L. fermentum CECT 4007T and L. cellobiosus CECT 562T and cluster A26 grouped the three subspecies of L. delbrueckii. Clusters A6, A7, A9–A11 and A21 comprised one or more wine strains without a reference strain and were named Lactobacillus spp. The remaining clusters contained only reference strains. The reproducibility of the method was 98·52 %S for each enzyme. The calculated global cophenetic correlation value for the 16S-ARDRA patterns was 0·96, indicating an excellent level of reliability.
The 16S-ARDRA patterns with BfaI and MseI permitted us to distinguish between almost all reference strains and to identify six species of wine lactobacilli: L. paracasei subsp. paracasei (13 strains), L. collinoides (2 strains), L. mali (10 strains), L. hilgardii (71 strains), L. coryniformis subsp. coryniformis (1 strain) and L. brevis (19 strains).
The pairs of strains L. zeae ATCC 15820T–L. casei CECT 475T, L. paracasei subsp. paracasei CECT 4022T–L. casei ATCC 334, L. reuteri CECT 925T–L. oris CECT 4021T, L. plantarum CECT 748T–L. pentosus CECT 4023T and L. fermentum CECT 4007T–L. cellobiosus CECT 562T and the subspecies of L. delbrueckii could not be differentiated by BfaI and/or MseI 16S-ARDRA. However, we achieved a good discrimination of L. reuteri and L. oris from the different profiles obtained with AluI (Rodas et al., 2003). Giraffa et al. (1998) were able to discriminate L. delbrueckii subsp. bulgaricus and L. delbrueckii subsp. lactis by 16S-ARDRA with EcoRI. We confirmed their results with our reference strains. The discrimination of single species in the other strain pairs was impossible with AluI or other enzymes tested.
Our results show that L. plantarum and L. pentosus grouped together in the same cluster. These species share more 99 % 16S rRNA gene sequence similarity (Collins et al., 1991; Quere et al., 1997) and therefore it is impossible to discriminate them by 16S-ARDRA. Despite their high 16S rRNA gene sequence similarity, Zanoni et al. (1987) demonstrated that they are separate species on the basis of DNA–DNA hybridization studies. The 36 wine strains grouped together with these species could be assigned to either of these species.
We could also not discriminate between L. fermentum and L. cellobiosus. Chenoll et al. (2003) published the 16S rRNA gene sequence of the former species and showed that this sequence shared more than 99 % similarity with that of L. fermentum. Other data supporting their taxonomic proximity are the high phenotypic similarities, the complete DNA–DNA relatedness found by Kandler & Weiss (1986) and their identity when they were analysed by the tDNA-PCR technique (Baele et al., 2002). Recently, Dellaglio et al. (2004) proposed the reclassification of L. cellobiosus as a later heterotypic synonym of L. fermentum.
The L. casei group comprises three species: L. casei, L. paracasei and L. rhamnosus (Collins et al., 1989). However, the taxonomic position of the L. casei group remains unclear. Dellaglio et al. (2002) requested an Opinion of the Judicial Commission of the International Committee on Systematics of Prokaryotes to replace the type strain of L. casei, ATCC 393T, with ATCC 334, to reject the name L. paracasei, to include L. paracasei strains in L. casei and to reclassify strain ATCC 393T as a member of L. zeae. Thirteen of our wine strains clustered together with the type strain of L. paracasei subsp. paracasei and L. casei ATCC 334, whereas type strains of L. casei CECT 475T and L. zeae ATCC 15820T clustered at 100 %S in another cluster (Supplementary Fig. C in IJSEM Online). These results support the request of Dellaglio et al. (2002). Our wine strains will remain named L. paracasei while the Judicial Commission considers the request of Dellaglio et al. (2002).
16S-ARDRA generates species-specific patterns in the majority of the studied species, but it is not useful for typing purposes because the 16S rRNA gene sequence is highly conserved at the species level.
RAPD fingerprinting analysis
Reproducibility of RAPD fingerprinting was evaluated in preliminary experiments using a subset of strains. Primer 16R gave poor reproducibility with homofermentative and facultatively heterofermentative lactobacilli, but produced reproducible species-specific profiles with heterofermentative lactobacilli. Thus, we decided to use this primer to analyse the latter group. Primer 17R showed clear and reproducible specific patterns for all reference species. Primer COC (Cocconcelli et al., 1995) was chosen to analyse homofermentative lactobacilli, as it generated species-specific and highly reproducible patterns. The reproducibility of the method ranged between 93·10 and 98·53 %S for these three primers. An example of the RAPD patterns obtained from some reference strains and some isolates is available as Supplementary Fig. D in IJSEM Online. Amplification with primer 16R generated 2–18 fragments of 200–4000 bp, primer 17R generated 5–20 fragments of 100–5000 bp and primer COC generated 10–30 fragments of 100–4000 bp.
On the basis of a computerized numerical analysis of combined RAPD patterns, we grouped strains into 34 clusters at a similarity level of 50 % (Supplementary Fig. E in IJSEM Online), named R1–R34. Cluster R4 was formed by L. cellobiosus CECT 562T and L. fermentum CECT 4007T. Cluster R6 comprised 10 wine strains and L. mali CECT 4149. Cluster R11 comprised 28 wine strains and L. plantarum CECT 748T. Cluster R17 comprised 16 wine strains and L. brevis CECT 216. Cluster R24 consisted of 71 wine strains and L. hilgardii NCFB 264T. Cluster R28 contained 13 wine strains and two reference strains: L. paracasei subsp. paracasei CECT 4022T and L. casei ATCC 334. Cluster R31 comprised two wine strains and L. collinoides CECT 922T. Cluster R34 comprised two wine strains and L. pentosus CECT 4023T. Clusters R5, R7, R9, R12, R14, R18, R19, R23 and R30 grouped one or more wine strains without a reference strain and were named as Lactobacillus spp. The remainder of the clusters comprised only single reference strains. The calculated cophenetic correlation value for the RAPD patterns was 0·94, indicating an excellent level of reliability.
The RAPD patterns allowed us to identify six species of wine lactobacilli: L. mali (10 strains), L. plantarum (28 strains), L. brevis (16 strains), L. hilgardii (71 strains), L. paracasei subsp. paracasei (13 strains), L. collinoides (2 strains) and L. pentosus (2 strains).
The technique permitted discrimination between all reference strains except the pair L. cellobiosus and L. fermentum; this result is concordant with the results obtained by 16S-ARDRA. The pair L. plantarum–L. pentosus, which could not be separated with 16S-ARDRA, were differentiated with RAPDs, and 30 of the 36 strains grouped with these two species were identified; two strains belonged to L. pentosus and 28 strains belonged to L. plantarum. This is the first time that L. pentosus has been reported in wine. The six strains clustered in R12 were adjacent to L. plantarum (44·74 %S) but did not show the species-specific pattern of this species. The group L. casei could be differentiated with the RAPD technique, except L. paracasei subsp. paracasei CECT 4022T and L. casei ATCC 334, supporting our previous results and the request of Dellaglio et al. (2002).
Analysis of the results with primer 17R revealed 123 distinct patterns from 178 wine Lactobacillus strains. The analysis of 92 wine heterofermentative lactobacilli with primer 16R revealed 17 patterns, whereas primer COC gave 63 distinct patterns from 86 wine homofermentative lactobacilli. The combined analysis of amplification profiles obtained with the three primers showed 131 distinct patterns from the 178 isolated strains.
The RAPD technique has been described as a useful technique for both identification and typing (Cocconcelli et al., 1995; Daud Khaled et al., 1997; Nigatu et al., 2001). In our study, the three selected primers show both characteristics, although primer 17R is of wider application than the other two, because it can be used for all wine lactobacilli and yielded more complex patterns, exhibiting a high typing ability. The discriminatory power of RAPD is lower than that of PFGE.
Ribotyping analysis
Ribotyping was applied to two different sets of wine strains. One set consisted of L. plantarum CECT 748T, L. pentosus CECT 4023T and 10 selected wine strains from cluster A14 in Supplementary Fig. C in IJSEM Online and clusters R11, R12 and R34 in Supplementary Fig. E. The second set consisted of L. paracasei subsp. paracasei CECT 4022T, L. casei ATCC 334 and 10 selected wine strains from cluster A2 in Supplementary Fig. C and cluster R28 in Supplementary Fig. E. L. casei CECT 475T, L. rhamnosus CECT 278T and L. zeae ATCC 15820T were also analysed in order to clarify the species assignment of the wine strains.
The dendrogram obtained from the numerical analysis of ribotypes grouped the strains into six clusters at a similarity level of 49 %S (Supplementary Fig. F in IJSEM Online), named E1 – E6. Cluster E1 comprised 10 wine strains and two reference strains: L. paracasei subsp. paracasei CECT 4022T and L. casei ATCC 334. Cluster E2 comprised nine wine strains and L. plantarum CECT 748T. Cluster E4 comprised one wine strain and L. pentosus CECT 4023T. The reproducibility level was 97 %S. The calculated cophenetic correlation value for the ribotypes profiles was 0·93, indicating an excellent level of reliability.
Ribotyping technique enabled us to confirm the identification of 20 studied strains; 10 strains belonged to L. paracasei subsp. paracasei, nine to L. plantarum and one to L. pentosus. We observed a specific-species band pattern for L. plantarum; however, within the cluster we could also differentiate two subclusters, as also observed with RAPD (clusters R11 and R12 in Supplementary Fig. E). Differences were evident among the patterns of the reference species.
Ribotyping results enabled the assignment to species of those strains grouped in the same cluster with L. pentosus and L. plantarum by 16S-ARDRA. Ribotyping confirmed the results obtained with RAPD and clarified the taxonomic status of some strains not grouped by this last technique (cluster R12 in Supplementary Fig. E). As in previous dendrograms, some wine strains were grouped with the type strain of L. paracasei subsp. paracasei and also with L. casei ATCC 334. The grouping of these latter reference strains was observed in RAPD, 16S-ARDRA and ribotyping, supporting the proposal of Dellaglio et al. (2002).
Global analysis
On the basis of a computerized numerical analysis of all combined patterns, a dendrogram was constructed (Fig. 1). Above 69 %S, 30 clusters could be delineated. Cluster T8 comprised 10 wine strains and L. mali CECT 4149. Cluster T13 contained 71 wine strains and L. hilgardii NCFB 264T. Cluster T17 comprised 19 wine strains and L. brevis CECT 216. Cluster T18 contained two wine strains and L. collinoides CECT 922T. Cluster T21 consisted of 13 wine strains, L. paracasei subsp. paracasei CECT 4022T and L. casei ATCC 334. Cluster T23 contained one wine strain and L. coryniformis subsp. coryniformis CECT 982T. Cluster T24 contained two wine strains and L. pentosus CECT 4023T. Cluster T25 comprised 34 wine strains and L. plantarum CECT 748T. Clusters T5, T6, T9, T10, T11, T12 and T15 comprised one or more wine strains without any reference strain and were named as Lactobacillus spp. The remaining clusters consisted of reference strains. The calculated global cophenetic correlation value was 0·96, indicating an excellent level of reliability.
|
Different authors have described difficulties with LAB phenotypic identification schemes (Vandamme et al., 1996; Björkroth et al., 2002). In our work, phenotypic identification allowed us to identify only three species. Moreover, of the 25 wine strains identified phenotypically as L. mali, only 10 grouped with this reference strain in the global analysis: 10 strains were close to L. helveticus CECT 4305T and five strains were close to L. mali CECT 4149 (Fig. 1
; clusters T8, T6 and T9–T12, respectively). All 48 wine strains identified phenotypically as L. hilgardii were assigned to this species by the global analysis, but this analysis also included 22 additional strains previously unidentified by phenotypic traits. Members of this last group ferment arabinose, while L. hilgardii is unable to ferment this carbohydrate according to Kandler & Weiss (1986) and Hammes et al. (1991). In agreement with our results, Sohier et al. (1999) found L. hilgardii wine strains that could ferment this sugar. Strain 449 was identified as L. coryniformis subsp. coryniformis by phenotypic traits, 16S-ARDRA and global analysis; however, RAPD clustering did not assign it to this species, though it was placed close to it. To verify the relationship between L. coryniformis subsp. coryniformis and strain 449, we sequenced the 16S rRNA gene from this strain. Phylogenetic analysis showed that the sequences share more than 99 % similarity (data not shown), confirming the results obtained by other techniques that classify strain 449 as L. coryniformis subsp. coryniformis.
The global analysis reflects the information from each experiment; however, the more congruent techniques with global analysis are RAPD and 16S-ARDRA. Amplification with the 17R primer shows a specific-species band pattern, and numerical analysis (data not shown) exhibits a similar clustering and identifies the same number of species from wine as global analysis. The 16S-ARDRA technique permits the identification of the reference strains and the assignment to species of our wine lactobacilli using the sequential analysis described by Rodas et al. (2003). A numerical confirmation of this statement was obtained by calculating the diversity indices to species level of the global analysis: values of 0·782 and 1·895 were obtained for the Simpson (1–
) and Shannon–Weaver (H') indices. When these calculations were applied to those individual tests that provided a correct identification of the majority of strains (RAPD with primer 17R and 16S-ARDRA with MseI), similar values of diversity were obtained to those of the global analysis [for RAPD-17R, the Simpson (1–) and Shannon-Weaver (H') indices were 0·800 and 2·082, respectively; for ARDRA-MseI they were 0·778 and 1·844]. For these reasons we propose that these techniques are useful to simplify and clarify the identification of lactobacilli. 16S-ARDRA has advantages over RAPD: it is less dependent on reaction conditions and the interpretation of results is easier.
Twenty-six wine strains remained unidentified after the global analysis. They are grouped in seven clusters: T5, T6, T9, T10–T12 and T15 (Fig. 1). We selected one representative strain of each cluster, except for cluster T15, from which four strains were chosen because preliminary results indicated that this group was more distant from reference strains (<95 % sequence similarity). The selected strains were 8, 59b, 68, 71, 88, 116, 154, 166, 203 and 459: their almost complete 16S rRNA gene sequences (approx. 1500 bp long) were subjected to similarity searches with the ARB database. The results obtained from the application of the three alternative treeing methods were congruent. All strains were assigned to the genus Lactobacillus. The analysis sited them specifically within the heterogeneous L. casei group as defined by Collins et al. (1991) and redefined into four groups by Schleifer & Ludwig (1995). The corresponding overall 16S rRNA gene sequence similarity matrix is available as Supplementary Table A in IJSEM Online. Fig. 2 displays a phylogenetic tree of Lactobacillus species, based on the neighbour-joining method, which included the 16S rRNA gene sequence data of these 10 strains. In the analysis, we have included Oenococcus oeni as the outgroup.
|
. This group represents a cluster of related species consisting of Lactobacillus alimentarius, Lactobacillus farciminis and the recently described Lactobacillus paralimentarius (Cai et al., 1999), Lactobacillus kimchii (Yoon et al., 2000), Lactobacillus mindensis (Ehrmann et al., 2003) and Lactobacillus versmoldensis (Kröckel et al., 2003). With the exception of L. kimchii, all these species have been isolated from sourdoughs, and not from wine. The closest relatives of strain 203 were L. kimchii (99·3 %), L. alimentarius (98·2 %) and L. paralimentarius (97·8 %) (Supplementary Table A). These three species were not included in this study because their habitats do not include wine (kimchi and sourdough). In further studies, we will compare strain 203 in depth with L. kimchii, L. paralimentarius and L. alimentarius by a polyphasic approach that will include DNA–DNA hybridization. Then, we will ascertain whether strain 203 could constitute a novel species or a new description from wine of one of these three species L. kimchii, L. paralimentarius and L. alimentarius.
Phylogenetic analysis placed the other nine wine strains within the L. salivarius group as defined by Schleifer & Ludwig (1995). This group represents a cluster of related species consisting of Lactobacillus agilis, Lactobacillus animalis, Lactobacillus aviarius, L. mali, Lactobacillus murinus, L. ruminis, Lactobacillus salivarius and the recently described L. nagelii (Edwards et al., 2000), Lactobacillus acidipiscis (Tanasupawat et al., 2000) and Lactobacillus equi (Morotomi et al., 2002). The majority of these species have been isolated from human or animal sources, while L. mali and L. nagelii have been isolated from wine. The closest relative of strains 8, 59b, 68 and 71 was L. mali, with 96·8, 97·8, 97·8 and 96·7 % sequence similarity, respectively; L. nagelii had approximately 95 % sequence similarity to these strains except strain 71, which exhibited 96·1 % similarity. Strain 88 shared 100 % 16S rRNA gene sequence similarity with L. mali (Supplementary Table A). Although this result and the phenotypic traits indicate that strain 88 should belong to L. mali, the rest of the tests and the global analysis do not allow assignment to this species. 16S rRNA gene sequences of strains 116, 154, 166 and 459 were almost identical (99·9 % similarity). The closest relatives of strains 116, 154, 166 and 459 were L. mali (approx. 95 %) and L. nagelii (approx. 95 %) (Supplementary Table A). DNA–DNA hybridizations must be done to reveal whether these nine strains belong to L. mali or L. nagelii or represent novel species.
After the global analysis, the number of different strains retained is 146 from 178 wine lactobacilli, PFGE being the most discriminatory technique (143 different patterns). One hundred and twenty-seven strains could be differentiated by global analysis and 51 strains were grouped in 19 groups. As has been stated before, some of these 178 strains could have been isolated several times from the same sample.
This vast study reveals that the majority of lactobacilli in wine belong to L. hilgardii (40 %), L. plantarum (19 %), L. brevis (10·7 %), L. paracasei subsp. paracasei (7·3 %) and L. mali (5·6 %), the first three being the most typical in wine. The species L. collinoides, L. coryniformis subsp. coryniformis and L. pentosus appear much less frequently and were always associated with spoiled wine. The latter three species and L. paracasei subsp. paracasei were described in wine for the first time by Rodas et al. (2003) and their presence is confirmed in the present work. Furthermore, the application of this polyphasic approach has allowed the detection of seven putative novel species and contributes to expanding our knowledge on the confusing genus Lactobacillus. Moreover, this approach allows evaluation of the taxonomic value of different techniques currently used, and clarifies controversial results obtained by the application of a single technique. The inclusion of a good set of reference strains is necessary to obtain consistent results. In addition to the taxonomic knowledge obtained in this work, we have demonstrated a simple identification scheme for wine lactobacilli: 16S-ARDRA with MseI and RAPD with primer 17R in cases of ambiguity.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Berthier, F. & Ehrlich, S. D. (1998). Rapid species identification within two groups of closely related lactobacilli using PCR primers that target the 16S/23S rRNA spacer region. FEMS Microbiol Lett 161, 97–106.[CrossRef][Medline]
Björkroth, K. J., Schillinger, U., Geisen, R., Weiss, N., Holzapfel, W. H., Korkeala, H. & Vandamme, P. (2002). Taxonomic study of Weissella confusa and description of Weissella cibaria sp. nov., detected in food and clinical samples. Int J Syst Evol Microbiol 52, 141–148.[Abstract]
Cai, Y., Okada, H., Mori, H., Benno, Y. & Nakase, T. (1999). Lactobacillus paralimentarius sp. nov., isolated from sourdough. Int J Syst Bacteriol 49, 1451–1455.
Cappuccino, J. G. & Sherman, N. (1987). Microbiology: a Laboratory Manual. Menlo Park, CA: Benjamin/Cummings.
Chagnaud, P., Machinis, K., Coutte, L. A., Marecat, A. & Mercenier, A. (2001). Rapid PCR-based procedure to identify lactic acid bacteria: application to six common Lactobacillus species. J Microbiol Methods 44, 139–148.[CrossRef][Medline]
Chenoll, E., Macián, M. C. & Aznar, R. (2003). Identification of Carnobacterium, Lactobacillus, Leuconostoc and Pediococcus by rDNA-based techniques. Syst Appl Microbiol 26, 546–556.[CrossRef][Medline]
Cocconcelli, P. S., Porro, D., Galandini, S. & Senini, L. (1995). Development of RAPD protocol for typing of strains of lactic acid bacteria and enterococci. Lett Appl Microbiol 21, 376–379.[Medline]
Collins, M. D. & Wallbanks, S. (1992). Comparative sequence analyses of the 16S rRNA genes of Lactobacillus minutus, Lactobacillus rimae and Streptococcus parvulus: proposal for the creation of a new genus Atopobium. FEMS Microbiol Lett 92, 235–240.[CrossRef]
Collins, M. D., Farrow, J. A. E., Phillips, B. A., Ferusu, S. & Jones, D. (1987). Classification of Lactobacillus divergens, Lactobacillus piscicola and some catalase-negative, asporogenous, rod-shaped bacteria from poultry in a new genus, Carnobacterium. Int J Syst Bacteriol 37, 310–316.
Collins, M. D., Phillips, B. A. & Zanoni, P. (1989). Deoxyribonucleic acid homology studies of Lactobacillus casei, Lactobacillus paracasei sp. nov., subsp. paracasei and subsp. tolerans, and Lactobacillus rhamnosus sp. nov., com. nov. Int J Syst Bacteriol 39, 105–108.
Collins, M. D., Rodrigues, U., Ash, C., Aguirre, M., Farrow, J. A. E., Martínez-Murcia, A., Phillips, B. A., Williams, A. M. & Wallbanks, S. (1991). Phylogenetic analysis of the genus Lactobacillus and related lactic acid bacteria as determined by reverse transcriptase sequencing of 16S rRNA. FEMS Microbiol Lett 77, 5–12.[CrossRef]
Collins, M. D., Samelis, J., Metaxopoulos, J. & Wallbanks, S. (1993). Taxonomic studies on some Leuconostoc-like organisms from fermented sausages: description of a new genus Weissella for the Leuconostoc paramesenteroides group of species. J Appl Bacteriol 75, 595–603.[Medline]
Daud Khaled, A. K., Neilan, B. A., Henriksson, A. & Conway, P. L. (1997). Identification and phylogenetic analysis of Lactobacillus using multiplex RAPD-PCR. FEMS Microbiol Lett 153, 191–197.[CrossRef][Medline]
Dellaglio, F., Felis, G. E. & Torriani, S. (2002). The status of the species Lactobacillus casei (Orla-Jensen 1926) Hansen and Lessel 1972 and Lactobacillus paracasei Collins et al. 1989. Request for an Opinion. Int J Syst Evol Microbiol 52, 285–287.[Abstract]
Dellaglio, F., Torriani, S. & Felis, G. E. (2004). Reclassification of Lactobacillus cellobiosus Rogosa et al. 1953 as Lactobacillus fermentum Beijerinck 1901. Int J Syst Evol Microbiol 54, 809–812.
Dicks, L. M. T. & van Vuuren, H. J. J. (1988). Identification and physiological characteristics of heterofermentative strains of Lactobacillus from South African red wines. J Appl Bacteriol 64, 505–513.
Dubernet, S., Desmasures, N. & Guéguen, M. (2002). A PCR-based method for identification of lactobacilli at the genus level. FEMS Microbiol Lett 214, 271–275.[CrossRef][Medline]
Edwards, U., Rogall, T., Blocker, H., Emde, M. & Bottger, E. C. (1989). Isolation and direct complete nucleotide determination of entire genes. Characterization of a gene coding for 16S ribosomal RNA. Nucleic Acids Res 17, 7843–7853.
Edwards, C. G., Haag, K. M., Collins, M. D., Hutson, R. A. & Huang, Y. C. (1998). Lactobacillus kunkeei sp. nov., a spoilage organism associated with grape juice fermentations. J Appl Microbiol 84, 698–702.[CrossRef][Medline]
Edwards, C. G., Collins, M. D., Lawson, P. A. & Rodriguez, A. V. (2000). Lactobacillus nagelii sp. nov., an organism isolated from a partially fermented wine. Int J Syst Evol Microbiol 50, 699–702.[Abstract]
Ehrmann, M. A., Martin, R. A. & Vogel, R. F. (2003). Molecular analysis of sourdough reveals Lactobacillus mindensis sp. nov. Int J Syst Evol Microbiol 53, 7–13.
Euzéby, J. P. (1997). List of bacterial names with standing in nomenclature: a folder available on the Internet (URL: http://www.bacterio.cict.fr/). Int J Syst Bacteriol 47, 590–592.
Ferrero, M., Cesena, C., Morelli, L., Scolari, G. & Vescovo, M. (1996). Molecular characterization of Lactobacillus casei strains. FEMS Microbiol Lett 140, 215–219.[CrossRef]
Frayne, R. F. (1986). Direct analysis of the major organic components in grape must and wine using high performance liquid chromatography. Am J Enol Vitic 37, 281–287.
Gevers, D., Huys, G. & Swings, J. (2001). Applicability of rep-PCR fingerprinting for identification Lactobacillus species. FEMS Microbiol Lett 205, 31–36.[CrossRef][Medline]
Giraffa, G., De Vecchi, P. & Rossetti, L. (1998). Identification of Lactobacillus delbrueckii subsp. lactis dairy isolates by amplified rDNA restriction analysis. J Appl Microbiol 85, 918–924.[CrossRef][Medline]
Hammes, W. P., Weiss, N. & Holzapfel, W. H. (1991). The genera Lactobacillus and Carnobacterium. In The Prokaryotes, 2nd edn, vol. 2, pp. 1535–1594. Edited by A. Balows, H. G. Trüper, M. Dworkin, W. Harder & K. H. Schleifer. New York: Springer.
Kandler, O. & Weiss, N. (1986). Genus Lactobacillus Beijerinck 1901, 212AL. In Bergey's Manual of Systematic Bacteriology, vol. 2, pp. 1208–1234. Edited by P. H. A. Sneath, N. S. Mair, M. E. Sharpe & J. G. Holt. Baltimore: William & Wilkins.
Kröckel, L., Schillinger, U., Franz, C. M. A. P., Bantleon, A. & Ludwig, W. (2003). Lactobacillus versmoldensis sp. nov. isolated from raw fermented sausage. Int J Syst Evol Microbiol 53, 513–517.
Leisner, J. J., Vancanneyt, M., Goris, J., Christensen, H. & Rusul, G. (2000). Description of Paralactobacillus selangorensis gen. nov., sp. nov., a new lactic acid bacterium isolated from chilli bo, a Malaysian food ingredient. Int J Syst Evol Microbiol 50, 19–24.[Abstract]
Lonvaud-Funel, A., Joyeux, A. & Ledoux, O. (1991). Specific enumeration of lactic acid bacteria in fermenting grape must and wine by colony hybridization with non-isotopic DNA probes. J Appl Bacteriol 71, 501–508.
Lortal, S., Valence, F., Bizet, D. & Maubois, J. L. (1997). Electrophoretic pattern of peptidoglycan hydrolases, a new tool for bacterial species identification: application to 10 Lactobacillus species. Res Microbiol 148, 461–474.[Medline]
Ludwig, W., Strunk, O., Westram, R. & 29 other authors (2004). ARB: a software environment for sequence data. Nucleic Acids Res 32, 1363–1371.
Matushek, M. G., Bonten, M. J. M. & Hayden, M. K. (1996). Rapid preparation of bacterial DNA for pulsed-field gel electrophoresis. J Clin Microbiol 34, 2598–2600.[Abstract]
Morotomi, M., Yuki, N., Kado, Y., Kushiro, A., Shimazaki, T., Watanabe, K. & Yuyama, T. (2002). Lactobacillus equi sp. nov., a predominant intestinal Lactobacillus species of the horse isolated from faeces of healthy horses. Int J Syst Evol Microbiol 52, 211–214.[Abstract]
Motoyama, Y., Funahashi, W. & Ogata, T. (2000). Characterization of Lactobacillus spp. by Ribotyping. J Am Soc Brew Chem 58, 1–3.
Nigatu, A., Ahrné, S. & Molin, G. (2001). Randomly amplified polymorphic DNA (RAPD) profiles for the distinction of Lactobacillus species. Antonie van Leeuwenhoek 79, 1–6.[CrossRef][Medline]
Nour, M. (1998). 16S-23S and 23S-5S intergenic spacer regions of lactobacilli: nucleotide sequence, secondary structure and comparative analysis. Res Microbiol 149, 433–448.[Medline]
Patarata, L., Pimentel, M. S., Pot, B., Kersters, K. & Faia, A. M. (1994). Identification of lactic acid bacteria isolated from Portuguese wines and musts by SDS-PAGE. J Appl Bacteriol 76, 288–293.
Quere, F., Deschamps, A. & Urdaci, M. C. (1997). DNA probe and PCR-specific reaction for Lactobacillus plantarum. J Appl Microbiol 82, 783–790.[CrossRef][Medline]
Rizzo, A. F., Korkeala, H. & Mononen, I. (1987). Gas chromatography analysis of cellular fatty acids and neutral monosaccharides in the identification of lactobacilli. Appl Environ Microbiol 53, 2883–2888.
Rodas, A. M., Ferrer, S. & Pardo, I. (2003). 16S-ARDRA, a tool for identification of lactic acid bacteria isolated from grape must and wine. Syst Appl Microbiol 26, 412–422.[CrossRef][Medline]
Saitou, N. & Nei, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4, 406–425.[Abstract]
Schillinger, U., Holzapfel, W. & Kandler, O. (1989). Nucleic acid hybridization studies on Leuconostoc and heterofermentative lactobacilli and description of Leuconostoc amelibiosum sp. nov. Syst Appl Microbiol 12, 48–55.
Schleifer, K. H. & Ludwig, W. (1995). Phylogeny of the genus Lactobacillus and related genera. Syst Appl Microbiol 18, 461–467.
Sohier, D. & Lonvaud-Funel, A. (1998). Rapid and sensitive in situ hybridization method for detecting and identifying lactic acid bacteria in wine. Food Microbiol 15, 391–397.[CrossRef]
Sohier, D., Coulon, J. & Lonvaud-Funel, A. (1999). Molecular identification of Lactobacillus hilgardii and genetic relatedness with Lactobacillus brevis. Int J Syst Bacteriol 49, 1075–1081.
Tanasupawat, S., Shida, O., Okada, S. & Komagata, K. (2000). Lactobacillus acidipiscis sp. nov. and Weissella thailandensis sp. nov., isolated from fermented fish in Thailand. Int J Syst Evol Microbiol 50, 1479–1485.[Abstract]
Tompkins, T. A., Stewart, R., Savard, L., Rusell, I. & Dowhanick, T. M. (1996). RAPD-PCR characterization of brewery yeast and beer spoilage bacteria. J Am Soc Brew Chem 54, 91–96.
Vandamme, P., Pot, B., Gillis, M., De Vos, P., Kersters, K. & Swings, J. (1996). Polyphasic taxonomy, a consensus approach to bacterial systematics. Microbiol Rev 60, 407–438.
Ventura, M. & Zink, R. (2002). Specific identification and molecular typing analysis of Lactobacillus johnsonii by using PCR-based methods and pulsed-field gel electrophoresis. FEMS Microbiol Lett 217, 141–154.[CrossRef][Medline]
Ventura, M., Casas, I. A., Morelli, L. & Callegari, M. L. (2000). Rapid amplified ribosomal DNA restriction analysis (ARDRA) identification of Lactobacillus spp. isolated from fecal and vaginal samples. Syst Appl Microbiol 23, 504–509.[Medline]
Weiss, N., Schillinger, U. & Kandler, O. (1983). Lactobacillus trichodes, and Lactobacillus heterohiochii, subjective synonyms of Lactobacillus fructivorans. Syst Appl Microbiol 4, 507–511.
Yoon, J. H., Kang, S. S., Mheen, T. I. & 7 other authors (2000). Lactobacillus kimchii sp. nov., a new species from kimchii. Int J Syst Evol Microbiol 50, 1789–1795.
Zanoni, P., Farrow, J. A. E., Phillips, B. A. & Collins, M. D. (1987). Lactobacillus pentosus (Fred, Peterson and Anderson) sp. nov., nom. rev. Int J Syst Bacteriol 37, 339–341.
Zapparoli, G., Torriani, S. & Dellaglio, F. (1998). Differentiation of Lactobacillus sanfranciscensis strains by randomly amplified polymorphic DNA and pulsed-field electrophoresis. FEMS Microbiol Lett 166, 325–332.[CrossRef]
Zhong, W., Millsap, K., Bialkowska-Hobrzanska, H. & Reid, G. (1998). Differentiation of Lactobacillus species by molecular typing. Appl Environ Microbiol 64, 2418–2423.
Zúñiga, M., Pardo, I. & Ferrer, S. (1993). An improved medium for distinguishing between homofermentative and heterofermentative lactic acid bacteria. Int J Food Microbiol 18, 37–42.[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  S. Rouse, C. Canchaya, and D. van Sinderen Lactobacillus hordei sp. nov., a bacteriocinogenic strain isolated from malted barley Int J Syst Evol Microbiol, September 1, 2008; 58(9): 2013 - 2017. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Cai, B. T. Rodriguez, W. Zhang, J. R. Broadbent, and J. L. Steele Genotypic and phenotypic characterization of Lactobacillus casei strains isolated from different ecological niches suggests frequent recombination and niche specificity Microbiology, August 1, 2007; 153(8): 2655 - 2665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Passoth, J. Blomqvist, and J. Schnurer Dekkera bruxellensis and Lactobacillus vini Form a Stable Ethanol-Producing Consortium in a Commercial Alcohol Production Process Appl. Envir. Microbiol., July 1, 2007; 73(13): 4354 - 4356. [Abstract] [Full Text] [PDF]
|
