Strain-dependent effects of inoculation of Lactobacillus plantarum subsp. plantarum on fermentation quality of paddy rice (Oryza sativa L. subsp. japonica) silage


  • Masanori Tohno,

    Corresponding author
    • National Agriculture and Food Research Organization, National Institute of Livestock and Grassland Science, Nasushiobara, Tochigi, Japan
    Search for more papers by this author
  • Hisami Kobayashi,

    1. National Agriculture and Food Research Organization, National Institute of Livestock and Grassland Science, Tsukuba, Ibaraki, Japan
    Search for more papers by this author
  • Kiyoshi Tajima,

    1. National Agriculture and Food Research Organization, National Institute of Livestock and Grassland Science, Tsukuba, Ibaraki, Japan
    Search for more papers by this author
  • Ryuichi Uegaki

    1. National Agriculture and Food Research Organization, National Institute of Livestock and Grassland Science, Nasushiobara, Tochigi, Japan
    Search for more papers by this author

Correspondence: Masanori Tohno, National Agriculture and Food Research Organization, National Institute of Livestock and Grassland Science, Nasushiobara, Tochigi 329-2793, Japan. Tel./fax: +81 287 37 7804; e-mail:


Paddy rice has been of particular interest as a forage crop in Japan. In this study, the isolated strains TO1000, TO1001, TO1002, and TO1003 were characterized by phenotypic and genotypic approaches. These strains were identified as Lactobacillus plantarum subsp. plantarum by species-specific PCR. Phenotypic characteristics varied among different strains of the same subspecies, and the strains represented unique and diverse phenotypes related to fermentation factors, such as carbohydrate assimilation and range of pH and temperature allowing growth. PCR analysis revealed that the patterns of presence/absence of known plantaricin genes differed in a strain-specific manner. Using these strains as inoculants for preparation of whole crop paddy rice silage, fermentation quality was significantly improved, as shown by lower pH, higher lactic acid content, and inhibition of the growth of undesirable microorganisms such as molds, coliform bacteria, and clostridia, after 30 and 60 days of storage, with effectiveness differing from strain to strain. These observations suggest that suitable candidates for bacterial inoculants in silage preparation should be screened at the strain level. Strain TO1002 may be useful for producing silage inoculants for the production of well-preserved whole crop paddy rice silage.


Paddy rice fields occupy over 11% of the total global cultivated area, and the major rice-producing countries of Asia account for over half of the world's population (Maclean et al., 2002). In Japan, there has been growing interest in paddy rice not only as a main dish for human consumption but also as a forage crop for livestock. As the result of population increase and urbanization in other Asian countries, the growth in demand for animal protein such as meat is rising, and may result in increased utilization of forage crops, such as paddy rice.

Silage with good quality depends on appropriate fermentation after storage, which results in the production of sufficient acid to inhibit the growth of microorganisms causing spoilage (McDonald et al., 1991). In general, well-preserved silage is characterized by different parameters, such as a pH value of approximately 4.2 or lower, high lactic acid content, low butyric acid and volatile basic nitrogen (VBN) concentrations, high dry matter (DM) recovery, and low counts of undesirable microorganisms (McDonald et al., 1991; Yunus et al., 2000). The lactic acid bacteria (LAB) play important roles in adequate acidification and production of higher-quality silage. Insufficient production of lactic acid by LAB results in poor-quality silage. To promote efficient fermentation in paddy rice silage, LAB should be added during the fermentation process. Some species of LAB used as silage additives, such as Lactobacillus plantarum, L. buchneri, L. acidophilus, L. brevis, L. rhamnosus, Pediococcus acidilactici, P. pentosaceus, and Enterococcus faecium, have proven effectiveness (McDonald et al., 1991; Yunus et al., 2000). Some in vitro differences in available carbohydrates, optimal growth pH and temperature, are observed among different LAB strains, even within the same species and subspecies (Tohno et al., 2012a). However, strain-dependent effects on fermentation quality of silage are not well understood.

In our previous study (Kobayashi et al., 2010) utilizing a L. plantarum strain, which has been used in the preparation of forage paddy rice in Japan, butyric acid fermentation caused by clostridia was observed in conditions such as lower storage temperature, lower available carbohydrates, and higher moisture content. Because the efficacy of LAB inoculants is highly influenced by their phenotypic characteristics, it might be useful to take advantage of a particular LAB strain that has a phenotypically appropriate growth potential in response to various environmental factors related to silage fermentation. Therefore, we isolated and characterized LAB strains inhabiting vegetative forage crops, taking particular interest in the development of novel inoculants contributing to good fermentation quality of paddy rice silage. Finally, we investigated differences in the fermentation quality of paddy rice silage inoculated with different conspecific strains, as well as the possibility that the isolates could aid efficient fermentation of the silage.

Materials and methods


The source of isolation was described in our previous studies (Kobayashi et al., 2010; Tohno et al., 2012a). The isolation process is described below. Grass silage (mixed pasture of timothy grass and orchardgrass), which was stored in a round bale for 300 days, was transferred into sterile homogenization bags, suspended 1 : 10 (w/v) in sterilized distilled water, and homogenized for 1 min in a Promedia SH-II M homogenizer (ELMEX, Tokyo, Japan). Serial dilutions were used for isolation of LAB using Lactobacilli de Man Rogosa Sharpe (MRS) agar (Difco, Detroit, MI) at 30 °C for 48 h under anaerobic conditions in a TE-HER Hard Anaerobox model ANX-1 (Hirosawa Ltd, Tokyo, Japan). Isolation and purification were as follows: colonies on MRS agar medium were picked and streaked to single colonies twice on MRS agar. The pure cultures were grown on MRS agar at 30 °C for 24 h, picked and transferred to nutrient broth (Difco) with 10% glycerol, and stored as stock cultures at −80 °C. The four isolated strains used were designated TO1000, TO1001, TO1002, and TO1003. These strains were deposited in the National Institute of Technology and Evaluation Biological Resource Center (Kisarazu, Chiba, Japan).

16S rRNA gene sequencing analysis

Molecular phylogeny analysis was conducted, and a phylogenetic tree was constructed based on about 1500 bases of 16S ribosomal RNA (rRNA) gene sequence as previously described (Tohno et al., 2012b).

PCR analysis

A recA multiplex PCR assay was performed to distinguish closely related species and subspecies of the L. plantarum group according to our previous report (Tohno et al., 2012b). PCR amplification of known plantaricin genes was conducted as described elsewhere (Omar et al., 2006). The primers used are listed in Supporting Information, Table S1.

Small-scale silage production

Paddy rice (Oryza sativa L. subsp. japonica) at the fully ripe stage was obtained from a local field at Kumagaya, Saitama, Japan, on October 25, 2010, by cutting using grass shears. In a small-scale fermentation system (Cai et al., 1997), approximately 100-g portions of the materials, chopped into about 20-mm lengths, were packed into 180 × 260 cm Hiryu KN-type plastic film bags (Asahikasei, Tokyo, Japan) with or without various bacterial inoculants (105 colony-forming units (CFU) g−1 fresh matter), and the bags were sealed with a BH 950 vacuum sealer (Panasonic, Osaka, Japan). Small-scale silage samples in a room at ambient temperature were collected at days 30 and 60 of the ensiling process. Ensiling was performed in triplicate.

Microbiological analysis

Samples (10 g) were blended with 90 mL of sterilized distilled water and chopped for 1 min in a Promedia SH-II M homogenizer. Serial dilutions were used for isolation of LAB using MRS agar at 30 °C for 72 h under anaerobic conditions. In addition, coliform bacteria were plated on blue light broth agar (Nissui Pharmaceutical Co. Ltd, Tokyo, Japan) and incubated at 30 °C for 72 h under aerobic conditions. Mold and yeast were incubated using potato dextrose agar (Nissui Pharmaceutical) adjusted to pH 3.5 with 10% tartaric acid at 30 °C for 72 h under aerobic conditions. Yeasts were distinguished from molds or bacteria by colony appearance and cell morphology. Aerobic bacteria were incubated on nutrient agar (Nissui Pharmaceutical) at 30 °C for 72 h. Homogenates of samples incubated at 75 °C for 15 min were used to count spore-forming clostridia and bacilli. Clostridia were counted on clostridia count agar (Nissui Pharmaceutical) after incubation in an anaerobic box at 30 °C for 3–5 days. Bacilli were detected on nutrient agar (Nissui Pharmaceutical) after aerobic incubation at 30 °C for 72 h. Colonies were counted as viable numbers of microorganisms [in CFU per gram of fresh matter (FM)].

Fermentation quality

Dry matter was analyzed according to method 934.01 of AOAC International. Fermentation products were extracted by sterilized distilled water as described above. The pH of the filtrate was measured with an MP230 glass electrode pH meter (Mettler Toledo, Columbus, OH). The organic acid contents were determined by high-performance liquid chromatography on an LC-2000Plus HPLC system (Jasco, Tokyo, Japan) as previously described (Cao et al., 2011). VBN was determined by steam distillation in a Kjeltec 2400 automatic distillation titration system (FOSS, Hillerød, Denmark); 10 mL of filtrate was steam distilled, and the VBN was absorbed in 2% (w/v) boric acid and then titrated with 0.01 M HCl solution in the presence of methyl red and bromocresol green indicators.

Statistical analysis

Differences in means were analyzed by one-way analysis of variance aided by prism software (Prism Software Co., Irvine, CA), and P values equal to or < 0.05 were considered statistically significant.

Results and discussion

The taxonomic position of the four strains was first investigated. The four strains were grouped on the phylogenetic tree with L. pentosus, L. plantarum subsp. plantarum, L. plantarum subsp. argentoratensis, and L. paraplantarum (Fig. S1). 16S rRNA gene sequence similarity is not sufficient to certify the species and subspecies in the L. plantarum group (Torriani et al., 2001; Bringel et al., 2005). Because the recA gene is more variable and can thus help differentiate within this group, the four strains were distinguished by means of recA gene amplification. Analysis by a recA-specific multiplex PCR revealed that the PCR products of all tested strains were similar to those of L. plantarum subsp. plantarum JCM 1149T, indicating that these strains are L. plantarum subsp. plantarum (Fig. 1).

Figure 1.

Amplification products obtained from a recA multiplex PCR assay. Lanes: M, marker 2-kb plus DNA ladder (Wako Pure Chemical Industries, Ltd, Tokyo, Japan); 1, L. casei JCM 1134T; 2, Lactobacillus paraplantarum DSM 10667T; 3, Lactobacillus pentosus JCM 1558T; 4, L. plantarum subsp. plantarum JCM 1149T; 5, L. plantarum subsp. argentoratensis JCM 16169T; 6–9; TO1000, TO1001, TO1002, and TO1003.

Phenotypic analysis revealed that these gram-positive, catalase-negative, and rod-shaped strains had the ability to grow at 10–45 °C and that only strains TO1002 and TO1003 had weak growth at pH 3.0 (Table 1). The four strains had a wider range of viable temperature and pH conditions than L. plantarum chikuso-1, which can grow at 15–45 °C and at pH 3.5–6.0 (Cai et al., 2003), or L. plantarum NGRI0320, which can grow at 15 °C but not 45 °C (Tanaka et al., 2000). Therefore, these four strains may be useful for developing an advanced L. plantarum subsp. plantarum-containing inoculant.

Table 1. Phenotypic characteristics and PCR amplification of plantaricin genes of TO strainsa
  1. a

    +, positive; −, negative; w, weakly positive.

Gram stain++++
Growth at
10 °C++++
15 °C++++
30 °C++++
45 °C++++
50 °C
Growth at
pH 3.0ww
pH 3.5++++
pH 4.0++++
pH 4.5++++
pH 5.0++++
pH 5.5++++
pH 6.0++++
Plantaricin A-related genes
plnA +++
plnB +++
plnC +++
plnD +++
plnEF +++
plnI +++
plnJ +++
plnK +++
plnG +++
plnN +++
Plantaricin NC8 structural gene
Plantaricin S structural gene
Plantaricin W structural gene

In rice grains, glucose, maltose, maltotriose, sucrose, raffinose, stachyose, fructose, xylose, raffinose, and arabinose are detectable (Murata et al., 1966; Singh & Juliano, 1977). In hydrolysates of rice straw, glucose, xylose, fructose, and arabinose are major monosaccharides, whereas small amounts of fucose, mannose, galactose, and rhamnose also are present (Sugahara et al., 1992; Sulbaran-de-Ferrer et al., 2003). In the analysis of their carbohydrate utilization, the tested strains had unique fermentation patterns compared with the type strains of the L. plantarum group (Table 2). In addition, differences in carbohydrate fermentation patterns were found among the L. plantarum subsp. plantarum strains in spite of the high similarity of their genetic backgrounds. For example, strains TO1000 and TO1001 showed positive reactions for utilization of l-arabinose, whereas TO 1002 and TO 1003 were negative. TO1001 had no ability to use l-rhamnose. Only TO1000 was able to assimilate starch, which is a major constituent of rice grains (Baun et al., 1970; Perdon et al., 1975; Perez et al., 1975). The potential to utilize carbohydrates might be an important factor in effectiveness of LAB inoculants on silage fermentation quality.

Table 2. Carbohydrate fermentation patterns of TO strains and related type strainsa
Acid sourceTO strainType strain
  1. All strains produced acid from d-galactose, d-glucose, d-fructose, d-mannose, d-mannitol, N-acetyl-glucosamine, amygdalin, arbutin, esculin ferric citrate, salicin, d-cellobiose, d-maltose, d-lactose, d-melibiose, d-sucrose, d-trehalose, gentiobiose, and gluconate, but failed to produce from erythritol, d-arabinose, L-xylose, D-adonitol, methyl-β-D-xylopyranoside, dulcitol, inositol, methyl-α-D-glucopyranoside, inulin, glycogen, xylitol, D-lyxose, D-tagatose, D-fucose, L-fucose, L-arabitol, 2-keto-gluconate, 5-keto-gluconate.

  2. a

    +, positive; −, negative; w, weakly positive.


Next, we evaluated the four strains as additives for whole crop paddy rice silage. The DM of paddy rice materials used was 43.0%. The pH value of homogenates of the materials was 6.24. Organic acids such as lactic acid, propionic acid, and n-butyric acid were not present at detectable levels. The VBN content was 0.02 g kg−1 FM. Before ensiling, the microbiological composition was LAB (6.66 log CFU g−1 FM), coliform bacteria (6.62), yeasts (8.26), aerobic bacteria (8.28), clostridia (3.00), bacilli (3.18), and molds (4.70).

As shown in Table 3, all strains increased fermentation rates in whole crop paddy rice silage, resulting in a significant pH decrease after 30 days of storage. Even within the same subspecies, a significant difference in pH after fermentation was observed between TO1000 and TO1002. Likewise, differences in the content of organic acids and VBN were also found among the treatments (Table 3). For example, the lactic acid content in LAB-treated samples was significantly higher than in the untreated samples, and strain TO1000 had the highest concentration. Because only TO1000 had the ability to use starch, which consists of a large number of glucose polymers (Table 1), efficient assimilation of starch might be an early event during fermentation by TO1000. In addition, the use of LAB was effective in decreasing the VBN content (Table 3). It has been reported that homofermentative LAB inoculants can decrease wasteful fermentation end products including ammonium nitrate and volatile fatty acids, which cause higher DM losses (Pahlow & Honig, 1994). DM is a material remaining after removal of water and contains the main nutrients found in feeds for animal growth (McDonald et al., 1991). TO1002 was useful for keeping a significantly higher DM, and the DM recovery also differed in a strain-dependent manner. Similarly, the numbers of viable microorganisms differed (Table 3). The LAB-inoculated samples maintained significantly higher numbers of LAB and had lower numbers of aerobic bacteria as well as undetectable levels of molds and yeasts. These results indicate that lower pH-resistant L. plantarum subsp. plantarum can survive in silage with acidic conditions for 30 days and inhibit the growth of undesirable microorganisms such as molds and yeasts. The viability of coliform bacteria, bacilli, and clostridia in the TO1000- and TO1001-containing samples fell below detectable levels, whereas those in the TO1002 and TO1003 samples tended to be detectable but were significantly or moderately depressed. Considering the differences in organic acid contents and pH values among different strains of the same subspecies, the distinct growth-inhibitory activities of organic acids might influence the survival of microorganisms in fermentative processes.

Table 3. Chemical composition and counts of viable microorganisms of paddy rice silage after 30 days of storagea
  1. DM, dry matter; FM, fresh matter; VBN, volatile basic nitrogen; cfu, colony-forming unit; nd, not detected.

  2. a

    Values are means of three samples. Means within columns with different superscript letters differ significantly (P < 0.05).

DM (%)29.6 ± 0.8bc27.1 ± 2.5bc27.0 ± 2.0bc35.2 ± 2.2a32.8 ± 0.2ab
pH6.14 ± 0.04a4.04 ± 0.02c4.15 ± 0.02bc4.23 ± 0.09b4.15 ± 0.07bc
Organic acid (g kg−1 FM)
Lactic acid1.2 ± 0.1c24.0 ± 2.0a18.5 ± 1.5ab17.0 ± 3.3b16.3 ± 1.8b
Acetic acid0.8 ± 0.0ab0.9 ± 0.1a0.9 ± 0.1a0.8 ± 0.2ab0.7 ± 0.1b
Propionic acidndndndndnd
n-Butyric acidndndndndnd
VBN (g kg−1 FM)0.14 ± 0.02a0.07 ± 0.01b0.09 ± 0.02b0.10 ± 0.03a0.07 ± 0.01b
Microorganisms (log cfu g−1 FM)
Lactic acid bacteria7.83 ± 0.28b8.78 ± 0.20a8.64 ± 0.06a8.88 ± 0.08a8.86 ± 0.07a
Molds4.46 ± 0.16ndndndnd
Yeasts2.80 ± 0.17ndndndnd
Aerobic bacteria8.12 ± 0.05a4.30 ± 0.35b4.51 ± 0.49b5.09 ± 0.68b5.03 ± 0.58b
Coliform bacteria7.19 ± 0.17andnd3.67 ± 0.68b3.44 ± 0.68b
Bacilli3.83 ± 1.09andndnd2.80 ± 0.17a
Clostridia4.42 ± 0.54andnd2.80 ± 0.17b2.80 ± 0.17b

After 60 days of storage, all LAB-inoculated samples showed significantly lower pH values than the no-additive group, reflecting significantly higher lactic acid content (Table 4). The VBN content in all LAB-treated samples was slightly lower than the control sample (Table 4). Silage treated with TO1002 or TO1003 showed significantly higher DM recovery (Table 4). The numbers of LAB in LAB-treated samples were maintained after 60 days and were significantly higher than the control (Table 4). Using LAB inoculants, the survival of unfavorable microorganisms such as molds, aerobic bacteria, coliform bacteria, bacilli, and clostridia was significantly suppressed or had dropped to below detectable levels. Bacilli and clostridia, which can generate dormant and highly resistant spore-forming cells in response to severe external environments (Setlow, 2006; Driks, 2007), were detected in the TO1000-treated samples (Table 4). In the case of TO1001, yeasts were detected at the same level as the control (Table 4). Certain yeasts survive and keep their intracellular pH between 6.0 and 7.5 when the extracellular pH varies from 3.5 to 9 (Salhany et al., 1975; Borst-Pauwels & Peters, 1977; Eraso & Gancedo, 1987). Thus, the ability of LAB inoculants to improve the whole crop paddy rice silage differed depending on the strain.

Table 4. Chemical composition and counts of viable microorganisms of paddy rice silage after 60 days of storagea
  1. DM, dry matter; VBN, volatile basic nitrogen; cfu, colony-forming unit; nd, not detected.

  2. a

    Values are means of three samples. Means within columns with different superscript letters differ significantly (P < 0.05).

DM (%)22.9 ± 2.4b28.6 ± 2.5b28.8 ± 2.0b30.0 ± 1.0a30.6 ± 3.7a
pH6.05 ± 0.13a4.19 ± 0.05b4.09 ± 0.05bc3.95 ± 0.04c4.14 ± 0.01bc
Organic acid (g kg−1)
Lactic acid0.7 ± 0.2a13.7 ± 2.0bc14.2 ± 0.8c15.4 ± 1.3c10.6 ± 0.8b
Acetic acid0.5 ± 0.1a0.7 ± 0.1a0.6 ± 0.1a0.5 ± 0.0a0.5 ± 0.0a
Propionic acidndndndndnd
n-Butyric acidndndndndnd
VBN (g kg−1)0.16 ± 0.04a0.12 ± 0.03a0.10 ± 0.02a0.10 ± 0.01a0.12 ± 0.02a
Microorganisms (log cfu g−1)
Lactic acid bacteria7.80 ± 0.25b8.43 ± 0.18a8.37 ± 0.14a8.67 ± 0.07a8.58 ± 0.12a
Molds3.22 ± 0.07ndndndnd
Yeasts3.70 ± 0.69and3.85 ± 0.30andnd
Aerobic bacteria9.35 ± 0.13a3.74 ± 0.04c3.71 ± 0.15c5.81 ± 0.16b5.78 ± 0.09b
Coliform bacteria6.78 ± 0.03ndndndnd
Bacilli4.57 ± 0.46a3.67 ± 0.51bndndnd
Clostridia4.61 ± 0.57a2.80 ± 0.17bndndnd

Some L. plantarum strains are able to produce antimicrobial bacteriocins, such as plantaricins A, EF, JK, N, NC8, S, and W (Stephens et al., 1998; Holo et al., 2001; Maldonado et al., 2003; Diep et al., 2009). Strain-related differences in bactericidal activity affect the susceptibility of other microorganisms to plantaricins and organic acids (Ehrmann et al., 2000; Omar et al., 2006; Nielsen et al., 2010). None of the strains had genes for plantaricins NC8, S, or W (Table 1). With the methodology used, plantaricin A-, EF-, JK-, and N-related genes were detectable in all strains except for TO1001 (Table 1). Similar to the case of TO1001, L. plantarum strain 3.9.1, isolated from an African fermented food, does not have any of these plantaricin genes (Omar et al., 2006). Certain L. plantarum strains show the following different types of plantaricin-related gene combinations: (1) plnEF and plnW; (2) plnD, plnEF, plnI, and plnG; (3) plnD, plnJ, plnK, and plnG; (4) plnD, plnEF, plnI, plnK, and plnG; (5) plnA, plnC, plnD, plnEF, plnI, plnJ, plnK, and plnN (Omar et al., 2006; Moghadam et al., 2010). Thus, the characteristics of the gene combinations carried for the production of plantaricins in TO1000, TO1002, and TO1003 are unique among the known L. plantarum strains isolated from fermented products. The synthesis of plantaricin A is observed from early exponential to early stationary phase. During stationary phase, the amount of plantaricin A strikingly declines (Diep et al., 1994). The addition of sucrose to the medium enhances production of nisin, another bacteriocin produced by Lactococcus lactis, (Devuyst & Vandamme, 1992). Thus, bacterial growth rate and available nutrients are associated with antimicrobial activity. In fact, the rates of fermentation differed among the four strains at 30 and 60 days of storage (Tables 3 and 4), suggesting that, in addition to the divergence in the available carbohydrates, the capacity for production of organic acids, and the pH and temperature preferences for growth, antimicrobial activity may also be an important factor in the regulation of silage fermentation quality. Further studies are needed both to elucidate the production of plantaricins by the TO strains inoculated in silage and to understand their roles in the improvement of silage quality.

In conclusion, phenotypic and genotypic differences were present among LAB strains in spite of their belonging to the same species and subspecies, and the fermentation quality of silage inoculated with different conspecific strains differed significantly, supporting the idea that suitable LAB inoculants should be selected on a strain basis. Because TO1002 most effectively improved the fermentation quality in terms of pH decrease, regulation of undesirable microorganisms, and high DM recovery, this strain should be the most suitable inoculant for longer storage of paddy rice silage. The selected L. plantarum subsp. plantarum TO1002 strain could help in the development of functional silage additives contributing to good fermentation quality of whole crop paddy rice silage.


We sincerely thank Mr. T. Sugita for his kind gifts of paddy rice. This study was partly supported by a Grant-in-Aid for Young Scientists (Start-up) (No. 21880053) from the Japan Society for the Promotion of Science, and a research grant for production of valuable livestock by feeding self-sufficient forage crops from the Ministry of Agriculture, Forestry and Fisheries of Japan.