**3. Results and discussion**

## **3.1. Chemical composition of materials and silage**

The contents of DM, crude protein, ether extract, nonfibous carbohydrate, ash, and neutral detergent fiber in molasses were 72.7, 4.3, 0.7, 83.6, 11.4, and 0%, respectively (Table 1). The contents of organic matter and crude protein in WCR were 86.5 and 5.3%, respectively. The contents of crude protein, nonfibous carbohydrate, and neutral detergent in the tofu cake were 30.1, 15.8, and 37.7%, respectively. And the chemical composition of vegetable residues is shown in Table 2. The DM of the four types of vegetable residues was less than 6%. Their OM contents were more than 70% of DM, and the crude protein and neutral detergent fiber contents were approximately 20% and 30% of DM, respectively. The water-soluble carbohydrates (including sucrose, glucose, and fructose) contents ranged from 8.4 to 21.7% of DM; the highest and lowest values were observed in white cabbage and lettuce residues, respectively.

Preparing TMR silage is one practice whereby food by-products are stored and utilized as ruminant feeds in Japan. It can avoid energy costs associated with drying, and may improve odors and flavors of unpalatable feed resources through fermentation in a silo (Cao et al., 2009a; Imai 2001; Wang & Nishino 2008). We have performed a number of experiments to investigate the effects of food by-products including tofu cake, rice bran, and


1Whole crop rice; 2Formula feed ("Koushi Ikusei Special Mash" made by Zenno with 120g kg–1 crude protein in fresh matter); 3Tofu cake; 4Wang & Goetsch (1998); 5Dry matter; 6Crude protein; 7Ether extract; 8Non-fibrous carbohydrate (100 – crude protein – ether extract – neutral detergent fiber – ash); 9Acid detergent fiber; 10Neutral detergent fiber.

**Table 1.** Chemical composition of WCR, concentrate, beet pulp and tofu cake used in total mixed ration silages (Cao et al., 2010b)



1Values are means ± SD.

368 Lactic Acid Bacteria – R & D for Food, Health and Livestock Purposes

**3. Results and discussion** 

silages (Cao et al., 2010b)

Tukey test was used to identify differences (*P* < 0.05) between means.

**3.1. Chemical composition of materials and silage** 

way analysis of variance. Data on chemical compositions, fermentative characteristics, in vitro ruminal DM digestibility, and fermentation products after 6-h incubation of silages opened on d 60 were analyzed using a completely randomized design with a 4 × 4 (vegetable residues × additive treatment) factorial treatment structure. The general linear model procedure of SAS version 9.0 (SAS Institute, Inc., Cary, NC) was used for the analysis, and the model included the main effects of vegetable residues and additive treatment, and their interactions. Sealing time and ensiling duration were excluded from the model because the 60-d silages, wastes processed, and silos were made only one time. The

The contents of DM, crude protein, ether extract, nonfibous carbohydrate, ash, and neutral detergent fiber in molasses were 72.7, 4.3, 0.7, 83.6, 11.4, and 0%, respectively (Table 1). The contents of organic matter and crude protein in WCR were 86.5 and 5.3%, respectively. The contents of crude protein, nonfibous carbohydrate, and neutral detergent in the tofu cake were 30.1, 15.8, and 37.7%, respectively. And the chemical composition of vegetable residues is shown in Table 2. The DM of the four types of vegetable residues was less than 6%. Their OM contents were more than 70% of DM, and the crude protein and neutral detergent fiber contents were approximately 20% and 30% of DM, respectively. The water-soluble carbohydrates (including sucrose, glucose, and fructose) contents ranged from 8.4 to 21.7% of DM; the highest

and lowest values were observed in white cabbage and lettuce residues, respectively.

investigate the effects of food by-products including tofu cake, rice bran, and

Preparing TMR silage is one practice whereby food by-products are stored and utilized as ruminant feeds in Japan. It can avoid energy costs associated with drying, and may improve odors and flavors of unpalatable feed resources through fermentation in a silo (Cao et al., 2009a; Imai 2001; Wang & Nishino 2008). We have performed a number of experiments to

 WCR1 Concentrate2 Beet pulp TC3 Molasses4 DM5 (%) 36.0 88.2 90.7 91.3 72.7 CP6 (% DM) 5.3 16.7 8.4 30.1 4.3 EE7 (% DM) 2.2 3.8 0.7 12.2 0.7 NFC8 (% DM) 32.1 60.1 33.7 15.8 83.6 Ash (% DM) 13.5 5.1 5.1 4.3 11.4 ADF9 (% DM) 30.2 8.7 25.6 22.2 – NDF10 (% DM) 48.0 14.4 52.1 37.7 0 1Whole crop rice; 2Formula feed ("Koushi Ikusei Special Mash" made by Zenno with 120g kg–1 crude protein in fresh matter); 3Tofu cake; 4Wang & Goetsch (1998); 5Dry matter; 6Crude protein; 7Ether extract; 8Non-fibrous carbohydrate (100 – crude protein – ether extract – neutral detergent fiber – ash); 9Acid detergent fiber; 10Neutral detergent fiber. **Table 1.** Chemical composition of WCR, concentrate, beet pulp and tofu cake used in total mixed ration **Table 2.** Chemical composition1 of vegetables residues (Cao et al., 2011)


1Molassess; 2Lactic acid bacteria (Lactobacillus plantarum); 3Standard error of means; 4Fresh matter; 5Dry matter; 6Formula feed ("Koushi Ikusei Special Mash" made by Zenno; total digestible nutrients, 70.0%; crude protein, 12.0% in fresh matter); 7Commercial vitamin-mineral supplement product (Snow brand seed, Iwate, Japan); 8Non-fibrous carbohydrate (100 – crude protein – ether extract – neutral detergent fiber – ash).

**Table 3.** Ingredient and chemical composition of total mixed ration silages (Cao et al., 2010b)

wet green tea waste on fermentation quality of WCR-containing TMR silage. Our previous study (Cao et al., 2009a, b) showed that silages with 30% tofu cake had higher lactic acid content, compared to those with rice bran and green tea waste. Therefore, we prepared TMR silage using tofu cake, and in order to investigate if adding LAB or molasses can further increase lactic acid content of the silages with tofu cake, LAB and molasses were added into these silages in this study. LAB can increase the lactic acid content of a silage (Cai, 2001; Cai et al., 2003), and was well used to prepare silage. Molasses is a fermentable carbohydrate (Maiga & Schingoethe, 1997) and many researchers (Alli et al., 1984) have reported its successful use with grass silage. In addition, molasses is a food by-product of sugar beet and sugarcane production. Molasses with high water-soluble carbohydrates is used as a major energy source for meat or milk production (Araba et al., 2002; Granzin & Dryden, 2005; Sahoo & Walli, 2008; Wang & gotsch, 1998).

Ruminal Digestibility and Quality of Silage Conserved via Fermentation by Lactobacilli 371

The pH, lactic acid, acetic acid, and ammonia-N were affected not only by vegetable, but also by addition and vegetable × addition. Comparison among the four types of vegetable silages revealed that the pH was the lowest (*P* < 0.001) in silage with white cabbage residue, followed by red cabbage, Chinese cabbage, and lettuce silages. The lactic acid content was highest in white cabbage silage (*P* < 0.001), whereas the acetic acid content was highest in lettuce silage (*P* < 0.001), followed by red cabbage, white cabbage, and Chinese cabbage silages. Propionic and butyric acids were not detected among the four types of vegetable silages; the ammonia-N concentration of white cabbage, red cabbage, and lettuce silages was lower (*P* < 0.001) than that of Chinese cabbage silage. Comparison of the treated silages showed that all silages treated with LAB or BP had lower (*P* < 0.001) pH values and ammonia-N concentrations, but higher (*P* < 0.001) lactic acid contents compared with the

Alli et al. (1984) assessed the effects of molasses on the fermentation of chopped whole-plant Leucaena. Silages were treated with molasses at a rate of 2.25% or 4.5% fresh weight at the time of ensiling, which led to increased rates of lactic acid production, lower pH, decreased DM loss, and reduced levels of ammonia-N compared to Leucaena to which no molasses was added. In the present experiment, WCR-containing TMR silages were treated with or without molasses at the rate of 4% fresh weight. Although the addition of molasses did not significantly influence the chemical composition, it increased DM and Non-fibrous carbohydrate contents by 3.06 and 3.56%, respectively. Adding molasses did not increase lactic acid content significantly, Adding LAB and molasses+LAB, however, increased lactic acid content significantly. This is probably because that, even if no molasses, there was enough fermentable sugars in TMR silage, and the LAB may have converted more fermentable sugars to lactic acid (Cai 2001; Cai et al., 2003). This study showed the advantage of LAB over molasses. Alli et al. (1984) reported that adding molassesf reduced the ammonia-N of silage, although in present study, ammonia-N did not differ among the

Furthermore, our previous study (Cao et al., 2011) has determined the effect of LAB inoculant and beet pulp addition on silage fermentation quality and *in vitro* ruminal DM digestion of vegetable residues. The silage treated with LAB or beet pulp had a lower pH and a higher lactic acid content than the control silage (Table 4). After 6 h of incubation, the LAB-inoculated silage had the highest DM digestibility and the lowest methane production. Weinberg et al. (2003) and Filya et al. (2007) reported that LAB inoculants affected the in vitro digestibility of alfalfa hay and corn silage, respectively, after 48 h incubation. In the present study, LAB inoculants not only increased (*P* < 0.01) the silage DM digestibility after 6 h in vitro incubation but also decreased ruminal methane production, which decreases the energy loss of feed (Cao et al., 2010a). Furthermore, LAB inoculants improve the fermentation quality of vegetable silage, which might decrease the degradation of crude protein in the silage; therefore, LAB-treated silage had a high concentration of ruminal ammonia-N. Adding beet pulp to vegetable silage did not affect the DM digestibility after 6 h in vitro incubation, but did increase the production of acetic acid, propionic acid, and even methane, while decreasing the production of butyric acid and ammonia-N. We cannot

control silage.

four silages.

### **3.2. Fermentation quality**

As indicated by the low pH value (around 4.0) and ammonia-N/total N content (2.83– 2.97%), high lactic acid content (2.49–2.87%), and V-score (99.8) for the silages, the four TMR silages were well preserved (Table 4). Although the levels of moisture, pH, acetic acid, propionic acid, butyric acid, and ammonia-N/total N and V-score did not differ significantly, lactic acid contents for the silages with LAB and molasses+LAB were higher (*P* = 0.005) than the control and molasses silages.

It is well established that LAB play an important role in silage fermentation, and LAB values have become a significant factor in predicting the adequacy of silage fermentation and in determining whether to apply bacterial inoculants to silage materials. Generally, when LAB reaches at least 105 (CFU/g of FM), silage can be well preserved (Cai 2001; Cai et al., 1999; Cai et al., 2003). However, the LAB values below 105 and aerobic bacteria values above 106 present in most WCR suggest that high-quality silage fermentation may need to be controlled using certain inoculants. The inoculant strain used in this study was Lactobacillus plantarum Chikuso-1; this strain promotes lactic acid fermentation and can grow well in low-pH environments. Therefore, silage prepared using this strain can promote the propagation of LAB, decrease pH, inhibit the growth of clostridia and aerobic bacteria, and improve silage quality (Cai et al., 1999).


1Molassess; 2Lactic acid bacteria (Lactobacillus plantarum); 3Standard error of means; 4Fresh matter; Means within a row with different letters (a, b) differ (*P* < 0.05).

**Table 4.** Fermentative characteristics of total mixed ration silages (Cao et al., 2010b)

The pH, lactic acid, acetic acid, and ammonia-N were affected not only by vegetable, but also by addition and vegetable × addition. Comparison among the four types of vegetable silages revealed that the pH was the lowest (*P* < 0.001) in silage with white cabbage residue, followed by red cabbage, Chinese cabbage, and lettuce silages. The lactic acid content was highest in white cabbage silage (*P* < 0.001), whereas the acetic acid content was highest in lettuce silage (*P* < 0.001), followed by red cabbage, white cabbage, and Chinese cabbage silages. Propionic and butyric acids were not detected among the four types of vegetable silages; the ammonia-N concentration of white cabbage, red cabbage, and lettuce silages was lower (*P* < 0.001) than that of Chinese cabbage silage. Comparison of the treated silages showed that all silages treated with LAB or BP had lower (*P* < 0.001) pH values and ammonia-N concentrations, but higher (*P* < 0.001) lactic acid contents compared with the control silage.

370 Lactic Acid Bacteria – R & D for Food, Health and Livestock Purposes

**3.2. Fermentation quality** 

(*P* = 0.005) than the control and molasses silages.

al., 2003). However, the LAB values below 105

row with different letters (a, b) differ (*P* < 0.05).

wet green tea waste on fermentation quality of WCR-containing TMR silage. Our previous study (Cao et al., 2009a, b) showed that silages with 30% tofu cake had higher lactic acid content, compared to those with rice bran and green tea waste. Therefore, we prepared TMR silage using tofu cake, and in order to investigate if adding LAB or molasses can further increase lactic acid content of the silages with tofu cake, LAB and molasses were added into these silages in this study. LAB can increase the lactic acid content of a silage (Cai, 2001; Cai et al., 2003), and was well used to prepare silage. Molasses is a fermentable carbohydrate (Maiga & Schingoethe, 1997) and many researchers (Alli et al., 1984) have reported its successful use with grass silage. In addition, molasses is a food by-product of sugar beet and sugarcane production. Molasses with high water-soluble carbohydrates is used as a major energy source for meat or milk production (Araba

As indicated by the low pH value (around 4.0) and ammonia-N/total N content (2.83– 2.97%), high lactic acid content (2.49–2.87%), and V-score (99.8) for the silages, the four TMR silages were well preserved (Table 4). Although the levels of moisture, pH, acetic acid, propionic acid, butyric acid, and ammonia-N/total N and V-score did not differ significantly, lactic acid contents for the silages with LAB and molasses+LAB were higher

It is well established that LAB play an important role in silage fermentation, and LAB values have become a significant factor in predicting the adequacy of silage fermentation and in determining whether to apply bacterial inoculants to silage materials. Generally, when LAB reaches at least 105 (CFU/g of FM), silage can be well preserved (Cai 2001; Cai et al., 1999; Cai et

most WCR suggest that high-quality silage fermentation may need to be controlled using certain inoculants. The inoculant strain used in this study was Lactobacillus plantarum Chikuso-1; this strain promotes lactic acid fermentation and can grow well in low-pH environments. Therefore, silage prepared using this strain can promote the propagation of LAB, decrease pH, inhibit the

Treatment SEM3 *P*-value

pH 3.99 3.92 4.01 4.03 0.0391 0.585

V-score 99.8 99.8 99.8 99.8 0.0090 0.970

1Molassess; 2Lactic acid bacteria (Lactobacillus plantarum); 3Standard error of means; 4Fresh matter; Means within a

**Table 4.** Fermentative characteristics of total mixed ration silages (Cao et al., 2010b)

Lactic acid (% FM4) 2.49a 2.52a 2.84b 2.87b 0.0880 0.005 Acetic acid (% FM) 0.09 0.09 0.10 0.09 0.0032 0.934 Propionic acid (% FM) 0.003 0.003 0.001 0.001 0.0009 0.574 Butyric acid (% FM) 0.002 0.003 0.003 0.003 0.0001 0.776 ammonia-N/total-N (%) 2.97 2.91 2.92 2.83 0.1775 0.956

growth of clostridia and aerobic bacteria, and improve silage quality (Cai et al., 1999).

Control M1 LAB2 M+LAB

and aerobic bacteria values above 106 present in

et al., 2002; Granzin & Dryden, 2005; Sahoo & Walli, 2008; Wang & gotsch, 1998).

Alli et al. (1984) assessed the effects of molasses on the fermentation of chopped whole-plant Leucaena. Silages were treated with molasses at a rate of 2.25% or 4.5% fresh weight at the time of ensiling, which led to increased rates of lactic acid production, lower pH, decreased DM loss, and reduced levels of ammonia-N compared to Leucaena to which no molasses was added. In the present experiment, WCR-containing TMR silages were treated with or without molasses at the rate of 4% fresh weight. Although the addition of molasses did not significantly influence the chemical composition, it increased DM and Non-fibrous carbohydrate contents by 3.06 and 3.56%, respectively. Adding molasses did not increase lactic acid content significantly, Adding LAB and molasses+LAB, however, increased lactic acid content significantly. This is probably because that, even if no molasses, there was enough fermentable sugars in TMR silage, and the LAB may have converted more fermentable sugars to lactic acid (Cai 2001; Cai et al., 2003). This study showed the advantage of LAB over molasses. Alli et al. (1984) reported that adding molassesf reduced the ammonia-N of silage, although in present study, ammonia-N did not differ among the four silages.

Furthermore, our previous study (Cao et al., 2011) has determined the effect of LAB inoculant and beet pulp addition on silage fermentation quality and *in vitro* ruminal DM digestion of vegetable residues. The silage treated with LAB or beet pulp had a lower pH and a higher lactic acid content than the control silage (Table 4). After 6 h of incubation, the LAB-inoculated silage had the highest DM digestibility and the lowest methane production. Weinberg et al. (2003) and Filya et al. (2007) reported that LAB inoculants affected the in vitro digestibility of alfalfa hay and corn silage, respectively, after 48 h incubation. In the present study, LAB inoculants not only increased (*P* < 0.01) the silage DM digestibility after 6 h in vitro incubation but also decreased ruminal methane production, which decreases the energy loss of feed (Cao et al., 2010a). Furthermore, LAB inoculants improve the fermentation quality of vegetable silage, which might decrease the degradation of crude protein in the silage; therefore, LAB-treated silage had a high concentration of ruminal ammonia-N. Adding beet pulp to vegetable silage did not affect the DM digestibility after 6 h in vitro incubation, but did increase the production of acetic acid, propionic acid, and even methane, while decreasing the production of butyric acid and ammonia-N. We cannot


explain the mechanism of these effects. More research is needed to elucidate the probiotic effect of adding LAB or beet pulp to vegetable silage in ruminants.

Ruminal Digestibility and Quality of Silage Conserved via Fermentation by Lactobacilli 373

bacteria, yeasts, and molds. When silage was treated with LAB or BP, the fermentation tended to ensure rapid and vigorous results with the faster accumulation of lactic acid (Table 3) and lower pH values at earlier stages of ensiling, and it also inhibited the production of acetic acid and ammonia-N during silage fermentation and thus improved vegetable-residue conservation. The transitional behavior of the VFA in the silage during fermentation indicated sharp decreases in pH, and corresponding increases in lactic acid contents at earlier stages (7 d of ensiling) were typical of a good fermentation process and were in agreement with previous studies. Subsequently, a steady reduction in pH depicted stability, while lactic acid contents gradually stabilized after a decrease during storage. Some studies (Cai, 2001; Cai et al., 1999; Cai et al., 2003) have shown that the development of LAB peaks in the first 7 d in parallel with the rise in lactic acid concentration of silage, and this is followed by decreases in LAB numbers; however, no apparent decrease in LAB numbers was observed in this study. The d 60 LAB-treated silage had higher organic matter and crude protein, but lower water-soluble carbohydrates than did control silages. During silage fermentation, LAB could effectively utilize water-soluble carbohydrates to produce sufficient lactic acid to reduce pH and inhibit the growth of harmful bacteria; therefore, the resulting silage was of good quality. Furthermore, the moisture content of silage material is also a major factor influencing silage fermentation (Cai et al., 2003). An intrinsic characteristic of vegetable residues is their very high moisture content (95 to 98% of FM), and this is a major limitation to its use as livestock feed. Although dried vegetable residues can easily be incorporated into rations, the energy cost associated with drying wet vegetable residues has been increasing. Moreover, the risk of effluent production is high because of the low DM content. Therefore, pressed vegetable residues have been preferred for adjusting moisture with other feed to ensile. Cai et al. (1999) reported that high-moisture silage is more beneficial to lactic acid fermentation and has less risk of heat damage than low-moisture silage. In this study, according to our preliminary experiment and taking into consideration the cost of feed, the moisture of BP-treated silage was adjusted to 70%, and most beet pulp-treated silages had lower pH and ammonia-N and higher lactic acid content compared with control silage. It is possible that this is because the addition of BP not only adjusted the moisture content of the vegetable residues but also increased the water-soluble carbohydrates content; therefore, silages with added BP could greatly contribute to better lactic acid fermentation. Furthermore, we used a small-scale system of silage fermentation; all silages stored well and maintained high quality without aerobic deterioration in this

**3.3.** *In vitro* **DM digestibility, and methane and VFA production** 

After in vitro 6 h incubation, DM digestibility, total VFA, acetic acid, isovaleric acid, valeric acid, and the acetic to propionic acid ratio did not differ significantly among the treatments (Table 6). However, methane production for the LAB silage and the molasses silage tended (*P* = 0.065) to decrease and increase, respectively, propionic acid for the LAB silage tended (*P* = 0.061) to increase, and butyric acid for the control silage was higher (*P* = 0.008) than the

study.

other silages.

1Means within columns with different letters (a-c) differ (*P* < 0.05); 2 Propionic and Butyric acids were not detected; 3Lactobacillus plantarum (Chikuso-1, Snow Brand Seed, Sapporo, Japan); 3Beet pulp.

**Table 5.** Fermentation profile of vegetable residue silages prepared with LAB and BP after 60 days of storage (Cao et al., 2011)

The pH, lactic acid, acetic acid, and ammonia-N were affected not only by vegetable, but also by addition and vegetable × addition (Table 5). Comparison among the four types of vegetable silages revealed that the pH was the lowest (*P* < 0.001) in silage with white cabbage residue, followed by red cabbage, Chinese cabbage, and lettuce silages. The lactic acid content was highest in white cabbage silage (*P* < 0.001), whereas the acetic acid content was highest in lettuce silage (*P* < 0.001), followed by red cabbage, white cabbage, and Chinese cabbage silages. Propionic and butyric acids were not detected among the four types of vegetable silages; the ammonia-N concentration of white cabbage, red cabbage, and lettuce silages was lower (*P* < 0.001) than that of Chinese cabbage silage. Comparison of the treated silages showed that all silages treated with LAB or BP had lower (*P* < 0.001) pH values and ammonia-N concentrations, but higher (*P* < 0.001) lactic acid contents compared with the control silage.

The factors involved in fermentation quality include not only the physiological properties of epiphytic bacteria but also the chemical composition of the silage material (Cai et al., 1999). In this study, the four types of vegetable residues had relatively high water-soluble carbohydrates contents; the epiphytic LAB transformed water-soluble carbohydrates into organic acids during the ensiling process, and as a result, the pH was reduced, which inhibited the growth of some microorganisms, such as bacilli, coliform bacteria, aerobic bacteria, yeasts, and molds. When silage was treated with LAB or BP, the fermentation tended to ensure rapid and vigorous results with the faster accumulation of lactic acid (Table 3) and lower pH values at earlier stages of ensiling, and it also inhibited the production of acetic acid and ammonia-N during silage fermentation and thus improved vegetable-residue conservation. The transitional behavior of the VFA in the silage during fermentation indicated sharp decreases in pH, and corresponding increases in lactic acid contents at earlier stages (7 d of ensiling) were typical of a good fermentation process and were in agreement with previous studies. Subsequently, a steady reduction in pH depicted stability, while lactic acid contents gradually stabilized after a decrease during storage. Some studies (Cai, 2001; Cai et al., 1999; Cai et al., 2003) have shown that the development of LAB peaks in the first 7 d in parallel with the rise in lactic acid concentration of silage, and this is followed by decreases in LAB numbers; however, no apparent decrease in LAB numbers was observed in this study. The d 60 LAB-treated silage had higher organic matter and crude protein, but lower water-soluble carbohydrates than did control silages. During silage fermentation, LAB could effectively utilize water-soluble carbohydrates to produce sufficient lactic acid to reduce pH and inhibit the growth of harmful bacteria; therefore, the resulting silage was of good quality. Furthermore, the moisture content of silage material is also a major factor influencing silage fermentation (Cai et al., 2003). An intrinsic characteristic of vegetable residues is their very high moisture content (95 to 98% of FM), and this is a major limitation to its use as livestock feed. Although dried vegetable residues can easily be incorporated into rations, the energy cost associated with drying wet vegetable residues has been increasing. Moreover, the risk of effluent production is high because of the low DM content. Therefore, pressed vegetable residues have been preferred for adjusting moisture with other feed to ensile. Cai et al. (1999) reported that high-moisture silage is more beneficial to lactic acid fermentation and has less risk of heat damage than low-moisture silage. In this study, according to our preliminary experiment and taking into consideration the cost of feed, the moisture of BP-treated silage was adjusted to 70%, and most beet pulp-treated silages had lower pH and ammonia-N and higher lactic acid content compared with control silage. It is possible that this is because the addition of BP not only adjusted the moisture content of the vegetable residues but also increased the water-soluble carbohydrates content; therefore, silages with added BP could greatly contribute to better lactic acid fermentation. Furthermore, we used a small-scale system of silage fermentation; all silages stored well and maintained high quality without aerobic deterioration in this study.

372 Lactic Acid Bacteria – R & D for Food, Health and Livestock Purposes

Vegetable residue means

Additive treatment means

storage (Cao et al., 2011)

with the control silage.

effect of adding LAB or beet pulp to vegetable silage in ruminants.

Control 95.4a 4.46a 8.2c

3Lactobacillus plantarum (Chikuso-1, Snow Brand Seed, Sapporo, Japan); 3Beet pulp.

White cabbage 82.4 3.59c

BP 69.1b 3.77c

LAB+BP 70.3b 3.68c

Significance of main effects and interactions

1Means within columns with different letters (a-c) differ (*P* < 0.05); 2

Item Moisture pH Lactic

Chinses cabbage 82.9 4.26a 10.8b 1.3c

explain the mechanism of these effects. More research is needed to elucidate the probiotic

Red cabbage 80.8 3.74b 12.7b 2.0b 0.26b Lettuce 83.9 4.27a 10.4b 3.3a 0.28b

LAB 95.1a 3.95b 10.8b 1.9b 0.26b

Vegetable residues (V) 0.249 <0.001 <0.001 <0.001 <0.001 Additive treatment (A) <0.001 <0.001 <0.001 <0.001 <0.001 V × A 0.841 <0.001 <0.001 <0.001 <0.001

**Table 5.** Fermentation profile of vegetable residue silages prepared with LAB and BP after 60 days of

The pH, lactic acid, acetic acid, and ammonia-N were affected not only by vegetable, but also by addition and vegetable × addition (Table 5). Comparison among the four types of vegetable silages revealed that the pH was the lowest (*P* < 0.001) in silage with white cabbage residue, followed by red cabbage, Chinese cabbage, and lettuce silages. The lactic acid content was highest in white cabbage silage (*P* < 0.001), whereas the acetic acid content was highest in lettuce silage (*P* < 0.001), followed by red cabbage, white cabbage, and Chinese cabbage silages. Propionic and butyric acids were not detected among the four types of vegetable silages; the ammonia-N concentration of white cabbage, red cabbage, and lettuce silages was lower (*P* < 0.001) than that of Chinese cabbage silage. Comparison of the treated silages showed that all silages treated with LAB or BP had lower (*P* < 0.001) pH values and ammonia-N concentrations, but higher (*P* < 0.001) lactic acid contents compared

The factors involved in fermentation quality include not only the physiological properties of epiphytic bacteria but also the chemical composition of the silage material (Cai et al., 1999). In this study, the four types of vegetable residues had relatively high water-soluble carbohydrates contents; the epiphytic LAB transformed water-soluble carbohydrates into organic acids during the ensiling process, and as a result, the pH was reduced, which inhibited the growth of some microorganisms, such as bacilli, coliform bacteria, aerobic

acid

20.9a 1.5c

% g/kg of FM

Acetic acid

Ammonia-

N

0.26b

0.41a

2.4a 0.52a

17.6a 2.5a 0.23bc

18.3a 1.4b 0.20c

Propionic and Butyric acids were not detected;

#### **3.3.** *In vitro* **DM digestibility, and methane and VFA production**

After in vitro 6 h incubation, DM digestibility, total VFA, acetic acid, isovaleric acid, valeric acid, and the acetic to propionic acid ratio did not differ significantly among the treatments (Table 6). However, methane production for the LAB silage and the molasses silage tended (*P* = 0.065) to decrease and increase, respectively, propionic acid for the LAB silage tended (*P* = 0.061) to increase, and butyric acid for the control silage was higher (*P* = 0.008) than the other silages.


Ruminal Digestibility and Quality of Silage Conserved via Fermentation by Lactobacilli 375

In the present experiment, there was a non-significant increasing trend in DM digestibility with molasses, LAB and molasses+LAB. Ruminal methane production and the molar proportion of propionic acid for silage with LAB decreased by 8.6% and increased by 4.8%, respectively. These might be because that adding LAB increased lactic acid content in the silage, when the silage containing high lactic acid content was incubated in vitro, there are two known mechanisms for the conversion of lactic acid or pyruvic acid to propionic acid, and when lactate acid is secondarily fermented by lactateutilizing bacteria such as Megasphaera elsdenii, Selenomonas ruminantium, and Veillonella parvula, propionate is generally produced as a major product (Dawson et al., 1997) and this can reduce methanogenesis because electrons are used during propionate formation. But adding molasses, which has a high sugar content, may augment methane production in the rumen (Hindrichsen et al., 2005), perhaps because of which, molasses per se canceled (compensated) the effect of lactic acid content on methane production. A further research is necessary about the effect of molasses and the complex effects of molasses and LAB concerning the methane production in TMR silage. Furthermore, it is not yet clear why adding molasses or LAB decreased in vitro ruminal butyric acid in this

study.

Vegetable residue means

Additive treatment means

1Acetic acid/propionic acid ratio. 2Digestible dry matter.

4Beet pulp.

2011)

Item DM

Significance of main effects and interactions

Means within columns with different letters (a-c) differ (P < 0.05).

3Lactobacillus plantarum (Chikuso-1, Snow Brand Seed, Sapporo, Japan).

digestibility

Red cabbage 44.3 44.8a 27.5ab 6.7c

Total VFA

White cabbage 44.9 43.0ab 24.8b 9.1a 2.7b 112.7a Chinese cabbage 38.6 41.9b 28.0ab 7.4bc 3.8a 88.3b

Lettuce 41.6 44.7a 29.2a 7.6b 4.0a 75.1c

Control 41.3ab 40.1b 24.1b 6.0b 4.2a 122.4b LAB 47.5a 42.2b 25.7b 6.5b 4.1a 164.1a BP 39.8b 45.6a 29.3a 9.1a 3.2b 40.3c BP+LAB 40.6b 46.4a 30.4a 9.2a 3.4b 41.2c

V × A <0.001 0.009 0.01 <0.001 <0.001 <0.001

Vegetable residues (V) 0.056 0.014 0.014 <0.001 <0.001 <0.001 Additive treatment (A) 0.001 <0.001 <0.001 <0.001 <0.001 <0.001

**Table 7.** Measurements of dry matter digestibility, methane production, VFA concentration and ammonia-N after 6-h in vitro incubation with rumen fluid of vegetable residue silage (Cao et al.,

Acetic acid

Propionic

% mmol/L mg/L

acid A/P1 Ammonia-

4.3a 91.8b

N

1Molassess; 2Lactic acid bacteria (Lactobacillus plantarum); 3Standard error of means; 4Dry matter; 5Digestible dry matter; Means within a row with different letters (a, b) differ (*P* < 0.05).

**Table 6.** *In vitro* dry matter digestibility, methane production and volatile fatty acid concentration after 6 hours incubation of total mixed ration silages (Cao et al., 2010b)

DM digestibility, VFA, and ammonia-N concentrations of vegetable-residue silage after 6 h incubation in vitro are shown in Table 7. Although DM digestibility was not influenced by vegetable, it was influenced by addition and by vegetable × addition; VFA, and ammonia-N were influenced by vegetable, addition, and vegetable × addition. DM digestibility did not differ among silages. However, ruminal CH4 production of white and Chinese cabbage silages was lower (*P* < 0.001) than that of red cabbage and lettuce silages, and the total VFA production of red cabbage and lettuce residue silages was higher (*P* = 0.014) than that of Chinese cabbage silage. The acetic acid production of the lettuce silage was higher (*P* < 0.001) than that of the white cabbage silage. The propionic acid production of white cabbage was the highest (*P* < 0.001) among the four vegetable residues, followed by lettuce, which showed higher propionic acid production (*P* < 0.001) than did red or Chinese cabbage; the last two silages did not differ in this regard. Red cabbage had higher (*P* < 0.001) butyric acid production than Chinese cabbage and lettuce, and butyric acid production was higher in white cabbage (*P* < 0.001) than in Chinese cabbage. The A:P ratio of white cabbage silages was the lowest (*P* < 0.001) among the four types of vegetable silages. The highest and lowest (*P* < 0.001) ammonia-N production was found in white cabbage and lettuce silages, respectively. The LAB-treated silage had a higher (*P* < 0.001) DM digestibility than BP- and beet pulp+LAB-treated silages; it also had the highest ammonia-N production. Together with the control silage, LAB treated silage had lower (*P* < 0.001) total VFA, acetic acid, and propionic acid production, but higher (*P* < 0.001) butyric acid production and acetic acid:propionic acid ratio ratio compared with beet pulp- and LAB+beet pulp-treated silages.

In vitro DM digestibility was higher in silage with LAB than without LAB because LAB reduces DM loss in silage fermentation (Cai 2001; Cai et al., 2003). Furthermore, although there are some reports that adding molasses has no effect on DM digestibility (Granzin & Dryden 2005; Wang & Goetsch 1998), many more studies (Shellito et al., 2006; Sahoo & Walli 2008) have reported that diets with molasses have higher ruminal DM digestibility. In the present experiment, there was a non-significant increasing trend in DM digestibility with molasses, LAB and molasses+LAB. Ruminal methane production and the molar proportion of propionic acid for silage with LAB decreased by 8.6% and increased by 4.8%, respectively. These might be because that adding LAB increased lactic acid content in the silage, when the silage containing high lactic acid content was incubated in vitro, there are two known mechanisms for the conversion of lactic acid or pyruvic acid to propionic acid, and when lactate acid is secondarily fermented by lactateutilizing bacteria such as Megasphaera elsdenii, Selenomonas ruminantium, and Veillonella parvula, propionate is generally produced as a major product (Dawson et al., 1997) and this can reduce methanogenesis because electrons are used during propionate formation. But adding molasses, which has a high sugar content, may augment methane production in the rumen (Hindrichsen et al., 2005), perhaps because of which, molasses per se canceled (compensated) the effect of lactic acid content on methane production. A further research is necessary about the effect of molasses and the complex effects of molasses and LAB concerning the methane production in TMR silage. Furthermore, it is not yet clear why adding molasses or LAB decreased in vitro ruminal butyric acid in this study.


Means within columns with different letters (a-c) differ (P < 0.05).

1Acetic acid/propionic acid ratio.

2Digestible dry matter.

3Lactobacillus plantarum (Chikuso-1, Snow Brand Seed, Sapporo, Japan).

374 Lactic Acid Bacteria – R & D for Food, Health and Livestock Purposes

matter; Means within a row with different letters (a, b) differ (*P* < 0.05).

LAB+beet pulp-treated silages.

6 hours incubation of total mixed ration silages (Cao et al., 2010b)

Treatment SEM3 *P*-value

Control M1 LAB2 M+LAB

A/P 0.9 1.0 0.9 1.0 0.02 0.271

1Molassess; 2Lactic acid bacteria (Lactobacillus plantarum); 3Standard error of means; 4Dry matter; 5Digestible dry

**Table 6.** *In vitro* dry matter digestibility, methane production and volatile fatty acid concentration after

DM digestibility, VFA, and ammonia-N concentrations of vegetable-residue silage after 6 h incubation in vitro are shown in Table 7. Although DM digestibility was not influenced by vegetable, it was influenced by addition and by vegetable × addition; VFA, and ammonia-N were influenced by vegetable, addition, and vegetable × addition. DM digestibility did not differ among silages. However, ruminal CH4 production of white and Chinese cabbage silages was lower (*P* < 0.001) than that of red cabbage and lettuce silages, and the total VFA production of red cabbage and lettuce residue silages was higher (*P* = 0.014) than that of Chinese cabbage silage. The acetic acid production of the lettuce silage was higher (*P* < 0.001) than that of the white cabbage silage. The propionic acid production of white cabbage was the highest (*P* < 0.001) among the four vegetable residues, followed by lettuce, which showed higher propionic acid production (*P* < 0.001) than did red or Chinese cabbage; the last two silages did not differ in this regard. Red cabbage had higher (*P* < 0.001) butyric acid production than Chinese cabbage and lettuce, and butyric acid production was higher in white cabbage (*P* < 0.001) than in Chinese cabbage. The A:P ratio of white cabbage silages was the lowest (*P* < 0.001) among the four types of vegetable silages. The highest and lowest (*P* < 0.001) ammonia-N production was found in white cabbage and lettuce silages, respectively. The LAB-treated silage had a higher (*P* < 0.001) DM digestibility than BP- and beet pulp+LAB-treated silages; it also had the highest ammonia-N production. Together with the control silage, LAB treated silage had lower (*P* < 0.001) total VFA, acetic acid, and propionic acid production, but higher (*P* < 0.001) butyric acid production and acetic acid:propionic acid ratio ratio compared with beet pulp- and

In vitro DM digestibility was higher in silage with LAB than without LAB because LAB reduces DM loss in silage fermentation (Cai 2001; Cai et al., 2003). Furthermore, although there are some reports that adding molasses has no effect on DM digestibility (Granzin & Dryden 2005; Wang & Goetsch 1998), many more studies (Shellito et al., 2006; Sahoo & Walli 2008) have reported that diets with molasses have higher ruminal DM digestibility.

DM4 digestibility (%) 42.2 44.5 44.8 44.5 1.03 0.313 Methane production (L kg–1 DDM5) 10.5 11.2 9.6 10.2 0.30 0.065 Total VFA (mmol 100 ml–1) 5.3 5.8 5.9 5.7 0.16 0.340 Acetic acid (A) (mol %) 37.0 38.9 38.0 38.8 0.55 0.142 Propionic acid (P) (mol %) 39.9 40.4 41.8 40.5 0.35 0.061 Butyric acid (mol %) 19.9a 17.8b 17.0b 17.8b 0.34 0.008 Isovaleric acid (mol %) 0.4 0.3 0.3 0.3 0.06 0.844 Valeric acid (mol %) 2.8 2.5 2.5 2.6 0.10 0.416

<sup>4</sup>Beet pulp.

**Table 7.** Measurements of dry matter digestibility, methane production, VFA concentration and ammonia-N after 6-h in vitro incubation with rumen fluid of vegetable residue silage (Cao et al., 2011)
