Although the use of probiotics has increased in recent years, the positive health effects of consuming fermented foods were noted over 100 years ago by Elie Metchnikoff (1). The use of probiotics as farm animal feed supplements dates back to the 1970s (2). They were originally incorporated into animal feed to accelerate growth and improve health by increasing the animal's resistance to disease. They have also gradually replaced antibiotics and contributed to reducing excessive use of minerals in animal feeds, parallel to the increasing recognition of negative environmental and health effects of antibiotics and minerals when used as dietary additives (3).
Probiotics were recently defined as deliberately ingested preparations of living micro-organisms that exert health and physiological benefits on the host when administered in a sufficient amount (4). The most commonly used probiotics are Lactobacillus, Bifidobacterium, and Saccharomyces, especially L. casei, L. rhamnosus, L. acidophilus, L. johnsonii, L. plantarum, B. longum, B. bifidum, and S. cerevisiae boulardii (5, 6).
The beneficial application of probiotics in animals has been well documented (7). For instance, a mixture of L. acidophilus and Streptococcus faecium (synonym: Enterococcus faecium) lowers the numbers of pathogenic strains in lambs (8). Also, the administration of these probiotics to animals improves both the clinical and microbiological outcome of Salmonella typhimurium infection (8). The use of L. casei reduces the colonization levels of shiga toxin-producing Escherichia coli O157:H7 in rabbits (9). One study predicted that probiotics are not only useful as food, but also as a biological controller of fish disease and an activator of nutrient regeneration (10).
Many animal studies have demonstrated that probiotics are capable of stimulating the immune system and producing antimicrobial substances (11, 12). Thus, they are given to hosts in an attempt to maintain or restore normal microflora, promote resistance to colonization by pathogenic bacteria and prevent the overgrowth of microorganisms (11).
LAB-MOS (Alltech, Kentucky, USA) is an example of a probiotic preparation. It is a porcine food additive containing MOS, (an agent that may improve growth rate), L. acidophilus, and E. faecium in a heat-stable capsule (13). Ghenisson 22, (GHEN Corporation, Gifu, Japan) which contains L. acidophilus, is used as a probiotic preparation for poultry. In this study, the 16S rRNA gene sequence of the L. acidophilus included in LAB-MOS and Ghenisson 22 was determined, and those Lactobacillus species were reclassified on the basis of their 16S rRNA gene sequences.
The strains used in this study were L. acidophilus strain MNFLM01 (LAB-MOS) and L. acidophilus strain GAL-2 (Ghenisson 22). The bacterial strains were cultured on de Man, Rogosa, Sharpe (MRS) agar (Becton Dickinson, MA, USA) at 37°C in an anaerobic system (AnaeroPack; Mitsubishi Gas Chemical, Tokyo, Japan). Bacterial DNA was then extracted by boiling, and PCR of the 16S rRNA gene was performed using, for example, primers P0mod and PC5, as previously reported (14). The amplified products were sequenced, and the sequences compared with those of representative Lactobacillus species (including type strains and strains whose genome sequences have previously been determined), obtained from GenBank. Multiple alignment was performed up to 1000 times using default settings with ClustalX software (version 2.0), and the phylogenetic tree constructed using the free TreeViewX software (version 0.5.0) (http://taxonomy.zoology.gla.ac.uk/rod/treeview.html). A bootstrap value (1000) represents the accuracy of a branch in the phylogenetic tree (analysis was repeated 1000 times). Homology analysis was performed using the software BLAST (http://blast.ddbj.nig.ac.jp/top-e.html) and FASTA (http://fasta.ddbj.nig.ac.jp/top-j.html).
For conventional phenotype (carbohydrate fermentation) testing of Lactobacillus strains, the API 50 CHL system (bioMérieux, Marcy 1'Etoile, France), was used according to the manufacturer's instructions. In those experiments, type strains L. rhamnosus ATCC7469 (YIT 0105T), L. acidophilus ATCC4356 (YIT 0070T), and L. johnsonii ATCC33200 (YIT 0219T) were also employed, these were kindly provided by Dr. K. Watanabe (Yakult Central Institute for Microbiological Research, Tokyo).
The 16S rRNA gene sequences of strains MNFLM01 and GAL-2, determined using primers, for example, P0mod and PC5, were 1487 and 1508 base pairs respectively (GenBank accession numbers are AB288235 and AB295648 respectively). Those sequences were compared with the reported 16S rRNA gene sequences of Lactobacillus species (Fig. 1). In the phylogenetic tree analysis, Leuconostoc mesenteroides (type strain ATCC8293) was used as an outgroup. Strain MNFLM01 fell into the L. casei group (not into the L. acidophilus group), and it constructed the closest subcluster with the type strain ATCC7469 of L. rhamnosus (a bootstrap value for the branch consisting of MNFLM01 and ATCC7469 was 1000) (Fig. 1a, b). BLAST and FASTA analyses showed that strain MNFLM01 was 100% homologous with L. rhamnosus-type strain ATCC7469, while it was much less homologous with L. acidophilus-type strain ATCC4356 (86.1% homologous).
Strain GAL-2 fell into the L. acidophilus group, but constructed the closest subcluster with L. johnsonii (Fig. 1a); the bootstrap value for the branch consisting of GAL-2 and L. johnsonii strain NCC533, whose genome sequence was determined, was 566, and that for the branch consisting of GAL-2, NCC533, and L. johnsonii-type strain ATCC33200 was 979. BLAST and FASTA analyses showed that strain GAL-2 was 99.7% homologous with L. johnsonii-type strain ATCC33200 and 100% homologous with L. johnsonii strain NCC533, while it was much less homologous with L. acidophilus-type strain ATCC4356 (90.5% homologous).
Carbohydrate fermentation of Lactobacillus strains was examined using the API 50 CHL system (Table 1). This assay confirmed the identification of MNFLM01 as L. rhamnosus; however, it ambiguously identified GAL-2 as L. acidophilus. Moreover, L. johnsonii type strain ATCC33200 was assigned as L. acidophilus in this assay, indicating the limitation of species identification by this method.
|Strain||Identification (species)||% ID†|
|L. rhamnosus ATCC7469||L. rhamnosus||99.9|
|L. acidophilus ATCC4356||L. acidophilus||76.4|
|L. johnsonii ATCC33200||L. acidophilus||98.6|
For definition of bacterial species, although DNA-DNA association (with a homology value greater than 70%) serves as the gold standard, 16S rRNA sequencing (with a homology value greater than 97%) also plays a dominant role (15). DNA-DNA association and phylogenetic tree analysis of the 16S rRNA gene sequences have both been employed in Lactobacillus species (16–18). The API 50 CHL system (phenotypic [carbohydrate fermentation] test) is useful for identifying the genus of non-spore-forming anaerobic and facultative anaerobic gram-positive rods (including Lactobacillus) (19); however, because of a high level of phenotypic variability and lack of a sufficient database for Lactobacillus species, it fails to identify the species of Lactobacillus (19, 20).
There are several well-known groups in the genus Lactobacillus. L. acidophilus was first isolated from infant feces by Moro in 1900 (16), but the type strain was lost, and a neotype strain was designed by Hasen and Mocquot in 1970 (16, 21). Previously, from 1980, phenotypically identified L. acidophilus group strains were divided into various species (22–24) and more recently into six species (L. gasseri, L. johnsonii, L. gallinarum, L. acidophilus, L. amylovorus, and L. crispatus) (21) by DNA-DNA association. Consequently, L. gasseri is a major constituent of the L. acidophilus group, and current L. acidophilus, identified by DNA-DNA association, is a minor species in the group and rarely isolated from humans, except fermented milk products (21, 25). These six species can be discriminated from each other by phylogenetic tree analysis of 16S rRNA gene sequences (18, this study).
L. casei has been isolated from humans and from fermented milk products. In 1980, L. casei consisted of five subspecies casei, alactosus, pseudoplantarum, tolerans, and rhamnosus (26). Reclassification of L. casei based on DNA-DNA relatedness has been reported by two groups. Collins et al. in 1989 (27) presented three species and two subspecies: L. casei, L. paracasei subsp. paracasei, L. paracasei subsp. tolerans, and L. rhamnosus, and Dicks et al. in 1996 (28) proposed the following three species: L. zeae, L. casei, and L. rhamnosus. In both cases, all species of the L. casei group were well discriminated and now even from the L. acidophilus group by phylogenetic tree analysis of 16S rRNA gene sequences (17, 18, this study).
In contrast, the L. plantarum group is an exceptional case in which the species are barely discriminated by phylogenetic tree analysis of 16S rRNA gene sequences; however, it should be noted that the L. plantarum group itself is well separated from L. acidophilus and L. casei groups by phylogenetic tree analysis (18, this study). For differentiation of the species within the L. plantarum group, multiplex PCR assay for the recA gene has been developed (29).
Thus, recent classification of species belonging to the genus Lactobacillus is mainly based on information on their DNA. In this study, L. acidophilus (strain MNFLM01) from the LAB-MOS preparation was identified as L. rhamnosus (in the L. casei group) on the basis of 16S rRNA gene sequences. Similarly, L. acidophilus (strain GAL-2) from the Ghenisson 22 preparation was identified as L. johnsonii (in the L. acidophilus group). Such reclassification of L. acidophilus based on their DNA data has also been reported previously; for instance, Fujisawa et al. reported that most L. acidophilus strains were in fact L. gasseri (25). Misidentification and consequent misnaming of bacterial species used as probiotics should be avoided in regard to regulation of probiotic products and human or animal use.
Both L. rhamnosus and L. johnsonii are well-known species that have been described as probiotics. Regarding commercial use, LAB-MOS (including strain MNFLM01) is widely consumed on animal farms in Japan as a food supplement for poultry, pigs and cattle. To gain a better understanding of the effects of strains MNFLM01 and GAL-2, their activity as probiotics should be further studied at the molecular level.