Influence of inoculating forage with lactic acid bacterial strains that produce ferulate esterase on ensilage and ruminal degradation of fiber

Available online at www.sciencedirect.com

Animal Feed Science and Technology 145 (2008) 122–135

Influence of inoculating forage with lactic acid bacterial strains that produce ferulate esterase on ensilage and ruminal degradation of fiber夽 Victor L. Nsereko ∗ , Brenda K. Smiley, William M. Rutherford, Annette Spielbauer, Kimberly J. Forrester, George H. Hettinger, Elizabeth K. Harman, Bradley R. Harman Pioneer Hi-bred International Inc., 7300 NW 62nd Avenue, Johnston, IA 50131, USA Accepted 3 June 2007

Abstract By releasing ferulic acid from cell wall arabinoxylans, ferulate esterase (FE) can increase susceptibility of plant cell walls to enzymatic hydrolysis. As some lactic acid bacteria (LAB) produce FE, we investigated effects of ensiling perennial ryegrass (PRG) with LAB that produce FE on ruminal NDF degradation (NDFD) of the resulting silages. Among the 10,000 LAB screened, approximately 500 produced FE and 8 (i.e., Lactobacillus buchneri PTA-6138 and NRRL B-30866; Lactobacillus crispatus NRRL B-30868, 30869 and 30870; Lactobacillus reuteri NRRL B-30867; Lactobacillus brevis NRRL B-30865 and an unidentified Lactobacillus NRRLB-30871) were studied in detail. The PRG was harvested and ensiled with or without (control) inoculation with each individual LAB, in triplicate laboratory silos for 30 days. Silages were analyzed for fermentation characteristics, dried, ground (6 mm) and incubated in situ in Dacron bags (50 ␮m pore size) for 48 h in the rumens of three ruminally cannulated steers adapted to a diet of grass silage. L. buchneri strains increased silage pH

Abbreviations: ADF, acid detergent fiber; ADFD, ADF degradation; cfu, colony forming units; DM, dry matter; FE, ferulate esterase; FFW, forage fresh weight; LAB, lactic acid bacteria; MRS, De Man Rogosa Sharpe; aNDF, neutral detergent fiber; NDFD, NDF degradation; pNP, P-nitrophenyl; PRG, perennial ryegrass; ROT, rise over temperature; VFA, volatile fatty acids; WPCF, whole plant corn forage; WPCS, whole plant corn silage. 夽 This paper is part of a special issue entitled “Enzymes, Direct Fed Microbials and Plant Extracts in Ruminant Nutrition” guest edited by R. J. Wallace, D. Colombatto and P. H. Robinson. ∗ Corresponding author. Tel.: +1 515 727 4019. E-mail address: [email protected] (V.L. Nsereko). 0377-8401/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.anifeedsci.2007.06.039

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

123

and acetate (P<0.05) and reduced lactate concentrations (P<0.05). With the exception of NRRLB30871, all LAB increased (P<0.05) NDFD by 9–11%. To examine effects of combining L. buchneri PTA-6138 with Lactobacillus paracasei tolerans PTA-6135 (X11C38) on NDFD of whole plant corn silage (WPCS), forage (DM, 337 g/kg) was harvested and ensiled with or without inoculation in two silos of 2 tonnes each and allowed to ferment for 180 days. Fresh silage from each silo was fed to three ruminally cannulated steers in a cross-over experiment with two periods of 14 days that included 10 days for adaptation. WPCS was incubated in situ, as described above, in all six steers in a split–plot experiment with diet as the main plot and inoculation of silage as the subplot. Feeding inoculated silage did not influence NDFD and there was no interaction between the silage fed and inoculation of the silage incubated in situ. However, inoculation of forage with X11C38 increased NDFD by 15.8% (P=0.019). Since L. buchneri can improve aerobic stability (AS) of silage, we also determined effects of X11C38 on AS of WPCS. Forage from four hybrids was harvested, and each ensiled with or without inoculation in four laboratory silos that were stored for 50–57 days. AS was determined by recording silage temperature with oxygen exposure in a thermostable environment. Inoculation with X11C38 extended the duration of AS for all hybrids by 42–105 h (P<0.05). Inoculation with LAB that produce FE improve NDFD in ensiled PRG and X11C38 extended AS and improved NDFD of WPCS, suggesting that higher NDFD was due to changes in forage during storage. © 2007 Elsevier B.V. All rights reserved. Keywords: Ferulate esterase; NDFD; Lactobacillus buchneri

1. Introduction Ensiling is the anaerobic process of preserving moist crops by lactic acid fermentation. Under optimal ensiling conditions, epiphytic lactic acid bacteria (LAB) predominantly ferment endogenous plant water soluble carbohydrates into lactic acid that acidifies the crop and minimizes activity of aerobic organisms to preserve forage nutrients (McDonald et al., 1991). Effective traditional silage inoculants can accelerate or enhance lactic acid fermentation and improve preservation of forage because epiphytic LAB are often insufficient for efficient lactate fermentation (McDonald et al., 1991). Treatment of forage at ensiling with an effective bacterial inoculant under favourable ensiling conditions often improves animal performance, but these effects are typically attributed to improvements in fermentation (Muck, 1993). However, Keady and Steen (1995) observed that inoculating grass at ensiling with a commercial silage inoculant consisting of a single strain of Lactobacillus plantarum improved performance of beef cattle fed a ration containing the silage, in spite of there being no apparent change in silage fermentation characteristics. Indeed, these researchers were unable to relate any characteristic of the bacterial inoculant to the improved animal performance. More recently, LAB inoculants have been shown to survive ruminal conditions (Weinberg et al., 2003) and that LAB are consumed with silage to enter the rumen (Weinberg et al., 2004). Hence, feeding inoculated silage to ruminants can deliver direct-fed microorganisms to potentially enhance rumen function. Because of the possibility that LAB silage inoculants do more than preserve forage nutrients, efforts are underway to discover novel LAB strains with specific animal nutrition-enhancing properties that can be used as silage inoculants (Weinberg et al., 2004).

124

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

Animal performance can be improved by increasing rate and/or extent of ruminal digestion of the fiber components of forage. Lignin limits degradation of NDF (NDFD) by hydrolytic enzymes (Brown, 1985; Jung and Dietz, 1993), and improving NDFD can enhance dry matter (DM) intake and improve milk yield when DM intake is limited by gut fill (Dado and Allen, 1995). The NDF fraction of grasses consists predominantly of structural polysaccharides in the plant cell wall, which can be cross-linked or linked to lignin by phenolic acids such as ferulic acid (Grabber et al., 1998). Ferulic acid is esterified to the C5-hydroxyl of ␣-l-arabinose moieties of grass xylans (Kato and Nevis, 1985; Mueller-Harvey et al., 1986), and xylans are cross-linked by oxidative coupling of ferulate monomers into dehydromers, and by incorporation of ferulate into lignin (Jung and Dietz, 1993). Ferulic acid esters of arabinoxylans are thought to interfere with degradation of polysaccharides by impeding access of xylanases to xylan (Gorbacheva and Rodinova, 1977). Grabber et al. (1998) demonstrated that ferulate cross-linkages limited enzymatic degradation of synthetically lignified primary cell walls of maize cell suspensions, with a 70% reduction in ferulate-lignin cross-linking in cell wall suspensions increasing carbohydrate solubilization by fungal enzymes by 24–46%. Ferulate esterases (FE) release ferulic acid from arabinoxylans, and this increases susceptibility of cell wall to enzymatic degradation (Bartolome et al., 1995). Donaghy et al. (1998) reported that several subspecies of Lactobacillus produce FE. Thus, we sought to determine whether inoculation of forage at ensilage with LAB that produce FE would render the fibrous component of the silage more suscetable to degradation by ruminal polysaccharidases. The main aims of this study were to identify strains of LAB that produce FE(s), and to investigate effects of inoculating forage at ensiling with these strains on ensilage and ruminal degradability of the resulting silages. 2. Materials and methods 2.1. Lactic acid bacteria Approximately 10,000 isolates from the LAB culture collection of Pioneer Hi-Bred Int. (Johnston, IA, USA) were screened for their ability to produce FE by measuring zones of clearance when De Man Rogosa Sharpe (MRS; DifcoTM Lactobacilli MRS; Becton Dickinson and Company, Sparks, MD, USA) agar containing 12 ml/l of 100 g/l ethyl 4hydroxy-3-methoxycinnamate inoculated with these organisms as described by Donaghy et al. (1998). About 500 strains were positive for production of FE (experiments not shown), and a selection of them were studied further for quantitative determination of FE activity. The LAB strains that actively produced FE included Lactobacillus buchneri strains PTA-6138 and NRRL B-30866; Lactobacillus crispatus strains NRRL B-30868, NRRL B-30870 and NRRL B-30869; Lactobacillus reuteri strain NRRL B-30867; Lactobacillus brevis strain NRRL B-30865 and a yet unidentified Lactobacillus strain NRRL B-30871. 2.2. Determination of FE activity using 4-nitrophenyl ferulate FE activities of LAB strains PTA-6138, NRRL B-30866, NRRL B-30867, NRRL B-30865, NRRL B-30868, NRRL B-30870 and NRRL B-30871 were determined quantita-

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

125

tively using 4-nitrophenyl ferulic acid as the substrate (Mastihuba et al., 2002), which was purchased from the Institute of Chemistry of the Slovak Academy of Sciences (Dubravska, Cesta, Slovakia). Bacterial cells were grown for 24–48 h in MRS broth (10 ml) and harvested by centrifugation (3200 × g for 20 min at 20 ◦ C). Cells were re-suspended in 1 ml of a lysis buffer consisting of 100 mM HEPES (pH 7.0), sodium azide (10 ␮g/ml) and 5 ␮l of Dnase (Roche Diagnostics Corporation, Indianapolis, IN, USA) and then lysed using a French Press (French Press Cell, Pressure, SIM-AMINCO Spectronic Instruments Inc., Rochester, NY, USA). FE activities of microbial cell lysates were determined using the assay described by Mastihuba et al. (2002) with modifications as detailed below. The substrate was first dissolved in dimethylsulphoxide and then diluted to the final working substrate solution (2.5 mM in 0.5 M KPO4 ; pH 7.0). ESubstrate solution, 80 ␮l, was dispensed into a 96-well microtitre plate containing 20 ␮l of cell lysates, solutions were thoroughly mixed and incubated at 37 ◦ C for 30 min. Controls wells consisting of substrate solution in buffer (2.5 mM in 0.5 M KPO4 ; pH 7.0) and cell lysates in the same buffer were included and treated as the reaction mixtures. Following incubation, 20 ␮l from the reaction mixture or control wells was withdrawn using an eight-channel micropipette and added to a fresh microtitre plate well containing 180 ␮l of KPO4 (pH 8.0). The final volume in each microtiter plate well was 200 ␮l. Solutions were mixed thoroughly and optical densities determined at 405 nm using a microtiter plate reader (Vmax Kinetic Microplate Reader, Molecular Devices, Menlo Park, CA, USA). Reaction mixture absorbance readings were corrected for absorbance readings of controls prepared as described above. P-Nitrophenol (0, 0.025, 0.05, 0.1, 0.15, 0.2 and 0.25 mM in 0.5 M KPO4 (pH 8); 200 ␮l; Sigma Chemical Company, St. Louis, MO, USA) was used as a standard for the FE assay. Protein concentrations of the cells were determined using Bradford reagent (Sigma Chemical Company). FE activities of the cell lysates were expressed as nanomoles of P-nitrophenyl (pNP) released/min/mg of protein. 2.3. Ensilage of perennial ryegrass Perennial ryegrass (PRG), first cutting, was harvested at the Pioneer Livestock Nutrition Center (PLNC; Sheldahl, IA, USA). L. buchneri PTA-6138 was grown and freeze dried under contract by a supplier to Pioneer Hi-bred International, but all other strains were 24–48 h fresh cultures grown on MRS broth as described above. The PRG was either left uninoculated (control) or inoculated with the LAB at an estimated rate of 1 × 105 colony forming units (cfu)/g fresh forage weight (FFW). Lactobacilli were applied to forage as aqueous solutions (10 ml/4.54 kg FFW) and thoroughly mixed with the forage in a 114 l capacity plastic bag. Forage, 1.36 kg, was ensiled in triplicate in polyethylene packet silos that were vacuum packed and heat sealed as described by Dennis et al. (1999). Packet silos were incubated at 21 ± 2 ◦ C for 30 days. 2.4. Ensilage of whole plant corn in laboratory silos Pioneer® brand corn hybrids 33P67, 34G13, 37B35 and 34G13 (Pioneer Hi-Bred International Inc., Des Moines, IA, USA), harvested at respective DM values of 384, 321, 395 and 318 g/kg FFW were ensiled with or without inoculation with L. buchneri PTA-6138 or

126

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

a combination of L. buchneri PTA-6138 and Lactobacillus paracasei tolerans strain PTA6135 (X11C38) in four experiments designed to evaluate effects of inoculation on aerobic stability of whole plant corn silage (WPCS). L. buchneri strain PTA-6138 and L. paracasei tolerans strain PTA-6135 were grown, stabilized and lyophilized under contract by a suppler to Pioneer Hi-bred International. L. buchneri PTA-6138 was applied to forage as an aqueous solution at 2.2 ml/kg to deliver 1 × 105 cfu/g FFW. Inoculation with X11C38 was achieved by applying L. paracasei tolerans PTA-6135 (2 × 104 cfu/g FFW) and L. buchneri PTA-6138 (1 × 105 cfu/g FFW) to forage at 2.2 ml/kg in order to deliver a final cfu application rate of 1.2 × 105 cfu/g FFW. For each treatment, four experimental polyvinyl chloride pipe silos (10 cm × 35 cm) were filled and packed at 0.7 of maximum packing density (approximately 160 kg DM/M3 ) with a hydraulic press. This packing density was selected to allow for air retention in the silo to enhance the aerobic challenge to silage stability (Pahlow et al., 1999). Experimental silos were fitted with rubber quick caps at each end, and the top cap was equipped with a Bunsen valve to allow gas escape. Each silo was equipped with a hole (8 mm i.d.) at the top and bottom of the silo walls which were sealed with plastic tape. Silos were stored in an environmentally controlled room at 21 ± 2 ◦ C for 30 days. At 28 and 42 day of ensiling, the plastic tape covering the two silo wall holes was removed for 24 h to allow air to penetrate the forage and induce aerobic instability as described by Pahlow et al. (1999). Silos were opened after 50–57 days. 2.5. Ensilage of WPCS in 2 tonnes silos This experiment investigated effects of inoculating whole plant corn forage (WPCF) with X11C38 (1.2 × 105 cfu/g FFW) on NDFD of WPCS. WPCF was harvested with a John Deere brand 3950 forage chopper at a moisture content of 667 g/kg FFW and at approximately two-third milk line, the theoretical chop length of the forage was 0.95–1.2 cm. A single 2 tonnes silo was assigned to each treatment, being uninoculated WPCF (Control) or WPCF inoculated with X11C38. The inoculant was sprayed onto WPCF as a soluble suspension as the forage was dropped into the silo from a conveyor belt. The biomass were stomped manually as it was loaded into the silo, sealed with a sheet of plastic and covered with a plywood lid weighted with a 227 kb concrete weight. Silos were opened after 180 days and the crust on the surface of the silos removed and discarded. Silage samples collected from the top, middle and bottom of each silo was composited and prepared for in situ ruminal analysis as described below. 2.6. Silage analyses 2.6.1. Determination of silage DM, pH and fermentation products The laboratory silos were opened, emptied and the forage thoroughly mixed to produce a uniform mass. Aqueous extracts of silage from each silo were prepared by diluting 10 g of wet silage in 90 ml of sterile distilled water and agitating the mixture in a stomacher (Stomacher 400, Seward Ltd., London, England) for 1 min at the medium setting. Silage pH was determined on each extract immediately following preparation. The DM concentrations were determined by drying a wet sample to a constant weight in a forced air oven at 62 ◦ C.

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

127

Silage lactate and VFA concentrations were determined on the PRG silages by HPLC (Siegfried et al., 1984). 2.6.2. Determination of aerobic stability of WPCS For determination of aerobic stability of WPCS, the procedure of Honig (1990) was used with some modifications. Fresh silage (100 g of DM) was weighed into a plastic tub (100 mm i.d.; 150 mm height) equipped with a hole (10 mm i.d.) on both the lid bottom. A temperature probe was inserted into the center of the silage mass through a hole in lid. Each tub was placed into a Styrofoam cooler which provided 30 mm of foam insulation to the top and bottom and 60 mm to the side of each tub. The cooler was placed in to a temperature-controlled room (28 ◦ C) and silage temperature was recorded every 3 h for 168 h by a data logger (Model No. CR10, Campbell Scientific Inc., Logan, UT, USA). The time (h) for silage temperature to rise 1.7 ◦ C above ambient was recorded (ROT). 2.6.3. In situ ruminal incubations and fiber analyses A portion of silages from each packet or silo was dried to constant weight at 62 ◦ C and ground to pass a 6 mm screen for determination of NDFD in situ. Ground silage (1.5 g), prepared as described above, was weighed into tared in situ Dacron bags (5.5 cm × 5.5 cm; 40 ± 15 ␮m; Ankom Technology Corp., Fairport, NY, USA), that were sealed, dried and reweighed. In situ analysis of PRG silages was conducted by incubating the Dacron bags containing the ground silage into the rumens of three ruminally cannulated steers that had been fed and adapted to a grass silage diet for 2 weeks. For each silo, three replicate bags were incubated in each of the three ruminally cannulated steers for 48 h on the same day (i.e., a total of 27 bags were incubated per treatment). Following ruminal incubation, residues from the in situ Dacron bags were composited among steers within silo and analyzed for aNDF and ADF as described below. For WPCS, fresh silage from each 2 tonnes silo (control or X11C38) was fed to three ruminally cannulated steers in a cross-over experiment with two periods of 14 days that included 10 days for adaptation. The dried and ground WPCS was incubated in situ in Dacron bags for 48 h in the rumens of all six steers in a split–plot experiment with diet as the main plot and inoculation of in situ silage as the subplot. Fifty Dacron bags were prepared for each silage, with 25 Dacron in situ bags from each treatment incubated per steer. Following incubation, samples were composited within steer and period for analysis. The aNDF concentrations of the silage were determined on samples before and after the 48 h ruminal incubation using the Ankom Fiber Analyzer (Ankom Technology Corp.), with a heat stable ␣-amylase and sodium sulfite, as described by the manufacturer. Digestion coefficients for aNDF or ADF were calculated as the difference in weight of aNDF or ADF weight before versus after ruminal incubation divided by the weight of aNDF or ADF before ruminal incubation. 2.7. Statistical analysis Data were analyzed by one-way analysis of variance. FE activities of individual bacterial strains were compared using Tukey–Kramer HSD test (JMP 5.1). For comparisons

128

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

pertaining to the PRG experiments, treatment means were compared using Dunnett’s test (Dunnett, 1955). Results from the split–plot experiment measuring in situ digestion of corn silage with or without inoculation within the rumen of each of six animals (fed control or inoculated silage in two periods) were analyzed as a split–plot. Diet was tested using steer error mean square while the inoculant effect and the diet × inoculant interaction was tested using the diet by steer interaction error mean.

3. Results 3.1. FE activities of Lactobacillus strains Mean FE activities of the Lactobacillus strains (Fig. 1) ranged from 2.2 to 23.0 nmol pNP released/min/mg protein. Of the strains examined for FE activity, L. buchneri strain NRRL B-30866 had the highest (P<0.05) activity. FE activity of L. buchneri strain PTA-6138 did not differ from that of L. crispatus NRRL B-30868, but was higher (P<0.05) than that of L. crispatus NRRL B-30870, L. crispatus NRRL B-30869, L. brevis NRRL B-30865, L. reuteri NRRL B-30867 and unknown Lactobacillus NRRL B-30871. L. crispatus strains had comparable levels of FE activity, with values between 7 and 10 nmol pNP released/min/mg protein. L. brevis NRRL B-30865 had lower (P<0.05) FE activity than PTA-6138, NRRL B-30866 and NRRL B-30868, but similar levels of activity to other strains.

Fig. 1. Mean FE activities of LAB where error bars are standard deviations from the means of three independent assays and pNP is P-nitrophenol.

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

129

Table 1 Effects of inoculation with FE producing LAB on pH, DM (g/kg FFW), ammonia N (g/kg total N) and other fermentation characteristics (g/kg DM) of ryegrass silagesa pH Control L. brevis NRRL B-30865 L. buchneri PTA-6138 L. buchneri NRRL B-30866 L. reuteri NRRL B-30867 L. crispatus NRRL B-30868 L. crispatus NRRL B-30870 L. crispatus NRRL B-30869 Unknown Lactobacillus NRRL B-30871

4.55 4.35* 4.72* 4.87* 4.79* 4.60 4.53 4.60 4.11*

S.E.

0.035

a *

DM 305 281* 278* 306 326* 336* 314 305 320

Total N 15.6 15.4 16.0 15.9 16.2 15.6 15.2 14.9 15.4

3.87

0.51

Ammonia N 181 175 204 182 213 269 195 202 198 18.5

Lactic acid

Acetic acid

42.7 55.4 13.6* 20.8* 31.1 49.6 38.6 39.5 45.1

13.1 19.6 43.2* 21.7* 8.33 16.7 14.0 14.3 9.21

4.51

1.76

Values are the mean of three silos per treatment. Significantly different from control at P<0.05.

3.2. Perennial rye grass fermentation characteristics At harvest, PRG forage had a pH of 6.67 and DM of 329 g/kg. Compared to the control silage (Table 1), inoculation with either L. brevis strain NRRL B-30865 or unknown Lactobacillus strain NRRL B-30871 resulted in silage with a lower (P<0.05) pH, which is reflected by numerically higher lactic acid concentrations. In contrast, inoculation with either L. buchneri strains PTA-6138 and NRRL B-30866 and L. reuteri strain NRRL B-30867 increased (P<0.05) silage pH, PTA-6138 and NRRL B-30866 increased (P<0.05) acetate concentration and reduced (P<0.05) lactate concentration. Total and ammonia N concentrations (Table 1) were not affected by inoculation. However compared to the control, silages inoculated with L. brevis NRRL B-30865 and L. buchneri PTA-6138 had lower (P<0.05) DM while those treated with L. crispatus NRRL B-30868 and L. reuteri NRRL B-30867 had higher (P<0.05) DM. Additionally, silages inoculated with PTA-6138 (Table 2) had a higher (P<0.05) concentration of aNDF than control silages. Inoculation with other FE producing LAB did not alter concentrations of aNDF or ADF versus uninoculated silages. 3.3. Fermentation and aerobic stability of WPCS Control WPCS was well fermented with pH ranging from 3.80 to 3.89 among the four silage hybrids (Table 3). In general, inoculation with L. buchneri PTA-6138 or X11C38 did not substantially influence terminal silage pH values. Inoculating with L. buchneri PTA-6138 alone improved (P<0.05) aerobic stability in hybrid 33P67, and tended (P<0.10) to improve aerobic stability with two additional hybrids. However, with all four hybrids, inoculating with X11C38 improved (P<0.05) aerobic stability and these improvements ranged from 42 and 105 h versus the corresponding control silages.

130

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

Table 2 Fibrer components (g/kg DM) of ryegrass silagesa

Control L. brevis NRRL B-30865 L. buchneri PTA-6138 L. buchneri NRRL B-30866 L. reuteri NRRL B-30867 L. crispatus NRRL B-30868 L. crispatus NRRL B-30870 L. crispatus NRRL B-30869 Unknown Lactobacillus NRRL B-30871 S.E. a *

aNDF

ADF

571 596 618* 575 579 583 577 588 572

324 329 345 330 326 320 319 326 318

7.64

8.69

Values expressed are the mean of three silos per treatment. Significantly different for control at P<0.05.

3.4. In situ fiber digestion of PRG silages Inoculation of PRG with FE producing L. buchneri strains PTA-6138 and NRRL B30866, L. reuteri strain NRRL B-30867 and L. brevis strain NRRL B-30865 at ensilage (Table 4) increased in situ NDFD of the resulting silages at 48 h by 9–11% (P<0.05; Table 4). Similarly, inoculation with FE producing L. crispatus strains NRRL B-30868, NRRL B30869 and NRRL B-30870 increased (P<0.05) in situ NDFD at 48 h by 7–8%. In situ ADFD at 48 h was enhanced (P<0.05) by inoculation with L. buchneri PTA-6138, L. buchneri NRRL B-30866 and L. reuteri NRRL B-30867. 3.5. In situ fiber digestion of corn silage Feeding inoculated WPCS (Table 5) did not influence in situ NDFD at 48 h and no interaction between feeding (0.42 versus 0.36) and source of silage incubated in situ (0.44 versus 0.36) occurred. However, inoculation of WPCS at ensiling with X11C38 increased (P<0.05) in situ ruminal NDFD at 48 h by 15.8%.

4. Discussion We investigated whether use of LAB that produce FE as inoculants at ensiling would enhance ruminal degradation of fiber. Inoculation of PRG at ensiling with several FE producing LAB substantially increased in situ ruminal NDFD of PRG silages at 48 h, and inoculation of WPCF with X11C38 improved NDFD of WPCS. Since, in the latter study, feeding WPCS inoculated with X11C38 did not influence in situ NDFD of either inoculated or control WPCS at 48 h, the increased NDFD associated with inoculation occurred during silo fermentation. This is consistent with the proposed FE mode of action for LAB inoculants. Oba and Allen (1999) reported that increasing in vitro digestibility of NDF by 1 unit results in a potential increase in DM intake of 0.17 kg/day and an increase in milk yield of 0.25 kg/day. Extrapolating their observations to our results suggests that increases

Hybrid 33P67

Hybrid 34G13

Hybrid 37B35

Hybrid 34G13

pH

DM

ROTb

pH

DM

ROT

pH

DM

ROT

pH

DM

ROT

Control L. buchneri PTA-6138 X11C38 (1.2 × 105 cfu/g)

3.81 3.93 4.22*

336 319* 333

90.0 134* 132*

3.89 3.85 3.85

310 326 307

21.0 24.0 126*

3.88 3.88 3.88

389 374 374

31.5 78.0 100*

3.86 3.84 3.83

322 322 317

72.0 129 160*

S.E.

0.463

19.7

0.008

13.4

0.312

a b *

2.93

Values are means of four silos per treatment. Time in hours for silage to rise 1.7 ◦ C above ambient. Significantly different from control at P<0.05.

2.18

0.129

0.44

6.04

2.43

19.7

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

Table 3 Effects of inoculation with X11C38 on pH, DM (g/kg FFW) and aerobic stability (h) of WPCSa

131

132

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

Table 4 Effects of inoculation with FE producing LAB on mean NDF and ADF digestion coefficients (g/kg) of PRGa NDFD

ADFD

Control L. brevis NRRL B-30865 L. buchneri PTA-6138 L. buchneri NRRL B-30866 L. reuteri NRRL B-30867 L. crispatus NRRL B-30868 L. crispatus NRRL B-30870 L. crispatus NRRL B-30869 Unknown Lactobacillus NRRL B-30871

0.525 0.583* 0.575* 0.577* 0.596* 0.573* 0.578* 0.570* 0.566

0.549 0.586 0.607* 0.602* 0.602* 0.590 0.588 0.577 0.565

S.E.

0.01083

0.01209

a *

Values expressed are the mean of three silos per treatment. Significantly different from control at P<0.05.

in NDFD resulting from inoculation with FE producing LAB should be economically beneficial. Ruminal degradation of most components of forage fiber must have been enhanced by inoculation with FE producing LAB. Of the strains examined for production of FE, L. buchneri strains NRRL B-30866 and PTA-6138 (Fig. 1) produced the highest levels of enzyme. Because there was no relationship between level of FE activity of individual strains (r2 = 0.02) and NDFD, the level of activity may have limited impact on efficacy of ferulate esterase in the silo. Instead, the ability of the organism to grow and become established in the silo, and reproduce during fermentation, may be key to the effectiveness of any strain at enhancing NDFD. A primary objective of ensiling is to maximize preservation of crop DM by lactic acid fermentation (McDonald et al., 1991). Typically, LAB inoculants have been selected to increase the rate at which forage is acidified thereby limiting activity of aerobic organisms (Muck, 1993). While effects of other Lactobacillus strains on rates of PRG fermentation were not investigated in our study, the relatively high terminal silage pH values suggest that, with the exception of NRRL B-30865 and NRRL B-30871 which reduced terminal pH, most strains did not enhance lactate accumulation. This suggests that the task of incorporating Table 5 Influence of applying a silage inoculant containing FE producing PTA-6138 and L. paracasei tolerans PTA-6135 on in situ ruminal NDFDa of WPCS in steers fed uninoculated and or silages treated with the same inoculant Silage in the diet Control

Inoculated-X11C38

Average

S.E.M.

Control Inoculated-X11C38

0.362 0.422

0.363 0.440

0.363 0.431*

0.018

Average S.E.M.

0.392 0.022

0.402

* a

Significantly different from control at P<0.05. Values are least square means.

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

133

LAB strains that produce FE into commercial silage inoculants must include homofermentative LAB that enhance fermentation, followed by additional examination to ensure that the NDFD enhancing attribute is maintained. To our knowledge, effects of inoculating forage with L. crispatus strains on fermentation and nutritive value of the resulting silage are novel and have not been previously documented. However, L. buchneri strains have been the subject of substantial amounts of research as silage inoculants. Being a hetero-fermentative LAB, L. buchneri have a unique ability to ferment lactic acid to acetic acid, ethanol and 1,2 propanediol (Oude Elfrink et al., 2001). Consistent with our observations, their use as single strain silage inoculants results in accumulation of acetic acid and higher silage pH (Driehuis et al., 2001; Oude Elfrink et al., 2001). Acetic acid is an anti-fungal compound, and its accumulation over time, in silages inoculated with L. buchneri (Taylor et al., 2002), is believed to be responsible for restricting aerobic fermentation (Muck, 1996). Indeed, in the present study, inoculating WPCF with FE producing L. buchneri PTA-6138 alone, or in combination with L. paracasei tolerans PTA-6135 (X11C38), at ensiling tended to improve aerobic satiability of WPCS. Data indicates that FE producing L. buchneri strains, such as PTA-6138, can act as dual purpose inoculant strains to improve both aerobic stability and ruminal fiber fermentation. Indeed, inoculating with X11C38 was more effective in increasing aerobic stability than inoculating with L. buchneri PTA-6138 alone. Perhaps L. paracasei tolerans PTA-6135 increased lactic acid concentrations in silage and provided additional substrate for L. buchneri PTA-6138 to convert into acetic acid. Diferulic acid cross-linking of xylans, and ferulic acid cross-linking of xylans to lignin limit enzymic degradation of grass cell wall polysaccharides (Gorbacheva and Rodinova, 1977; Mitsuishi et al., 1998; Grabber et al., 1998; Grabber, 2005). However, it is not clear whether adding FE to lignified plant cell wall material can improve enzymatic hydrolysis of cell wall polysaccharides. Working with purified FE from Aspergillus niger and a recombinant Pseudomonas fluorescens FE, Bartolome et al. (1997) observed that these enzymes released 1–9% of the ferulic acid and 0.7% or less of the diferulic acid residues from barley and wheat cell walls, and incubating cell walls with xylanases dramatically increased release of ferulic acid, but not diferulic acid, by esterases. These researchers found that FE specifically released only 5-5 diferulic acid residues from either xylanase treated or insoluble fractions of cell walls, demonstrating that FE alone can release diferulic acid from cell walls. Grabber (2005) concluded that FE had a limited ability to improve plant cell wall degradability because they could not cleave an important proportion of crosslinkages within lignified cell walls. However, Garcia-Conseca et al. (1999) demonstrated that ferulate esterase A from A. niger could cleave at least one ester bond from diferulic acid cross-linkages in plant cell wall polymers without release of the free dimer. Thus conclusions about the effectiveness of FE alone for improving hydrolysis of lignified cell walls based on release of ferulic acid dimers from cell wall can be misleading because ferulic acid cross-linkages can apparently be cleaved without release of free diferulic acid. Supporting the efficacy of FE for improving polysaccharide hydrolysis, Borneman et al. (1993) demonstrated that combining isolated fungal FE to other fungal polysaccharidase mixtures was more effective in hydrolyzing Bermuda grass cell walls than the polysaccharidase mixture alone. In our study, FE producing LAB were effectively cultivated in forage during ensilage over a period of at least 30 days and, on exposure to ruminal

134

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

polysaccharidases, the fiber component of the resulting silages was more fermentable in the rumen. We postulate that in addition to enhancing DM preservation, silage inoculants can improve animal nutrition and performance by at least three broad mechanisms. First, through inoculation, nutrition-enhancing silage inoculants grow in silage and produce a metabolite, such as a bacteriocin which, if stable in the silo, can influence rumen microorganisms when the silage is fed. LAB such as Lactococcus lactis strains are known to produce nisin, which has been reported to have monensin-like effects on ruminal fermentation (Callaway et al., 1997). Secondly, nutrition-enhancing silage inoculants established in silage have a probiotic, and/or direct fed microbial, effect when fed to ruminants. Hence, the silage can both culture and deliver doses of specific organisms to the rumen (Weinberg et al., 2004). Finally, the organism being inoculated in forage may produce an active metabolite in silage that modifies it during storage thereby improving its nutritive value either by reducing toxins and/or enhancing digestibility of silage components. Our research falls into this latter category. 5. Conclusions Inoculation of forage at ensiling with LAB that produce FE enhanced ruminal degradation from both PRG and WPCS. Future forage inoculants consisting of LAB that produce FE, to increase NDFD, when combined with conventional LAB that accelerate fermentation and enhance aerobic stability, could improve conservation and feeding value of ensiled crops. References Bartolome, B., Faulds, C.B., Tuohy, M., Hazlewood, G.P., Gilbert, H.J., Williamson, G., 1995. Influence of different xylanases on the activity of ferulic acid esterase of wheat bran. Biotechnol. Appl. Biochem. 22, 65–73. Bartolome, B., Faulds, C.B., Kroon, P.A., Waldron, K., Gilbert, H.J., Hazlewood, G.P., Williamson, G., 1997. An Aspergillus niger esterase (ferulic acid esterase III) and a recombinant Pseudomonas fluorescens subsp. cellulosa esterase (XylD) release a 5-5 ferulic dehydrodimer (diferulic acid) from barley and wheat cell walls. Appl. Env. Biol. 63, 208–212. Borneman, W.S., Ljunghal, L.G., Hartley, R.D., Akin, D.E., 1993. Ferulate and p-coumaryl esterases from the anaerobic fungus Neocallimastix strain MC-2: properties and functions in plant cell wall degradation. In: Coughlan, M.P., Hazelwood, G.P. (Eds.), Hemicellulose and Hemicellulases. Portland Press, Chapel Hill, NC, USA, p. 85. Brown, A., 1985. Review of lignin biomass. J. Appl. Biochem. 7, 371–387. Callaway, T.R., Carneiro De Milo, A.M., Russell, J.B., 1997. The effects of nisin and monensin on ruminal fermentations in vitro. Curr. Microbiol. 35, 90–96. Dado, R.G., Allen, M.S., 1995. Intake limitations, feeding behaviour, and rumen function of cows challenged with rumen fill from dietary fiber or inert bulk. J. Dairy Sci. 78, 118–133. Dennis, S., Hendrick, C., Miller, D., 1999. The effects of variety, cut, moisture level and microbial inoculant on terminal pH of ensiled alfalfa. In: Proc. XII Int. Silage Conf., Swedish Univ. of Agric. Sci., Uppsala, Sweden, p. 87. Donaghy, J., Kelly, P.F., McKay, A.M., 1998. Detection of ferulic acid esterase production by Bacillus spp. and Lactobacilli. Appl. Microbiol. Biotechnol. 50, 257–260. Driehuis, F., Oude Elfrink, S.J.W.H., Van Wikselaar, P.G., 2001. Fermentation characteristics and aerobic stability of grass silage inoculated with Lactobacillus buchneri, with or without homofermentative lactic acid bacteria. Grass Forage Sci. 56, 330–343.

V.L. Nsereko et al. / Animal Feed Science and Technology 145 (2008) 122–135

135

Dunnett, C.W., 1955. A multiple comparison procedure for comparing several treatments with a control. J. Am. Stat. Assoc. 50, 1096–1121. Garcia-Conseca, M.T., Kroon, P.A., Ralph, J., Mellon, F.A., Colquhoun, I.J., Sauliner, L., Thibault, J.F., Williamson, G., 1999. A cinnamoyl esterase from Aspergillus niger can break plant cell wall cross-links without release of free diferulic acid. Eur. J. Biochem. 266, 644–652. Gorbacheva, I.V., Rodinova, N.A., 1977. Studies on xylan degrading enzymes. II. Action pattern of endo-1,4␤-xylanase from Aspergillus niger str. 14 on xylan and xylooligosaccharides. Biochim. Biophys. Acta 484, 94–102. Grabber, J.H., 2005. How do lignin composition, structure, and cross-linking affect degradability? A review of cell wall model studies. Crop Sci. 45, 820–831. Grabber, J.H., Ralph, J., Hatfield, R.D., 1998. Ferulate cross-links limit the enzymatic degradation of synthetically lignified primary walls of maize. J. Agric. Food. Chem. 46, 2609–2614. Honig, H., 1990. Evaluation of aerobic stability. In: Lingvall, P., Lindgren, S. (Eds.), Proc. Eurobac. Conf. Swed. Univ. of Agric. Sci., Uppsala, Sweden, 12–16 August 1986, pp. 76–81 (Grass and Forage Report No. 3-1990). Jung, H.J.G., Dietz, D.A., 1993. Cell wall lignification and degradability. In: Jung, H.G., Buxton, D.R., Hatfield, R.D. (Eds.), Forage Cell Wall Structure and Digestibility. American Society of Agronomy, Madison, WI, USA, pp. 315–346. Kato, Y., Nevis, P.J., 1985. Isolation and identification of O-5(5-O-feruloyl-␣-arabinofuranosyl)-(1-3)-O-␤xylopyranosyl-(1-4)-D-xylopyranose as a component of Zea shoot cell walls. Carbohydr. Res. 137, 139–150. Keady, T.J.W., Steen, R.W.J., 1995. The effect of treating low dry matter, low digestibility grass with a bacterial inoculant on intake and performance of beef cattle and studies of its mode of action. Grass Forage Sci. 50, 217–226. Mastihuba, V., Kremnicky, L., Mastihubova, M., Willet, J.L., Cote, G.L., 2002. A spectrophotometric assay for feruloyl esterase. Anal. Biochem. 309, 96–101. McDonald, P., Henderson, A.R., Heron, S.J.E., 1991. The Biochemistry of Silage, second ed. Chalcombe Publications, Marlow, UK. Mitsuishi, Y., Yamanobe, Y., Yagisawa, M., 1998. The mode of action of three xylanases from mesophilic fungus strain Y-94 on xylooligosaccharides. Agric. Biol. Chem. 52, 921–927. Muck, R.E., 1993. The role of silage additives in making high quality silage. In: Silage Production from Seed to Animal. NRAES-67: Northeast Regional Agricultural Engineering Service, Syracuse, NY, USA, pp. 106–116. Muck, R.E., 1996. A lactic acid bacteria strain to improve aerobic stability of silages. Pages 42–43 in U.S. Dairy Forage Res. Center 1996 Res. Summaries. Madison, WI. Mueller-Harvey, L., Hartley, R.D., Harris, P.J., Curzon, E.H., 1986. Linkage of p-coumaryl and feruloyl groups to cell-wall polysaccharides of barley straw. Carbohydr. Res. 148, 71–85. Oba, M., Allen, M.S., 1999. Evaluation of the importance of digestibility of NDF from forage: effects on dry matter intake and milk yield of dairy cows. J. Dairy Sci. 82, 589–596. Oude Elfrink, S.J.W.H., Krooneman, J., Gottschal, J.C., Spoelstra, S.F., Faber, F., Dreihuis, F., 2001. Anaerobic conversion of lactic acid to acetic acid and 1,2 propanediol by Lactobacillus buchneri. App. Env. Microbiol. 67, 1–8. Pahlow, G., Ruser, B., Honig, H., 1999. Inducing aerobic stability in laboratory scale silos. In: Proc. XII Int. Silage Conf., Swedish Univ. of Agric. Sci., Uppsala, Sweden, p. 253. Siegfried, V.R., Ruckemmann, H., Stumpf, G., 1984. Method for the determination of organic acids in silage by high performance liquid chromatography. Landwirtsch. Forsch. 37, 298–304. Taylor, C.C., Ranjit, N.J., Mills, J.A., Neylon, J.M., Kung Jr., L., 2002. The effect of treating whole plant barley with Lactobacillus buchneri 40788 on silage fermentation, aerobic stability and nutritive value for dairy cows. J. Dairy Sci. 85, 1793–1800. Weinberg, Z.G., Muck, R.E., Weimer, P.J., 2003. The survival of inoculant lactic acid bacteria in rumen fluid. J. Appl. Microbiol. 94, 1066–1071. Weinberg, Z.G., Chen, Y., Gamburg, M., 2004. The passage of lactic acid bacteria from silage in to rumen fluid, in vitro studies. J. Dairy Sci. 87, 3386–3397.