Maataloustieteellinen Aikakauskirja Vol. 57: 284—292, 1985 RESEARCH NOTE Enzymes as silage additive. Effect on fermentation quality, digestibility in sheep, degradability in sacco and performance in growing cattle PEKKA HUHTANEN, KARI HISSA, SEIJA JAAKKOLA and ESKO POUTIAINEN 1 Department of Animal Husbandry, University of Helsinki, SF-00710 HELSINKI, Finland Abstract. Fungal glucose oxidase and cellulase were used as silage additives in labo- ratory (25 1), pilot (3 mJ) and farm scale (2501) silos. In 3m’ scale silos, pFI and the concentra- tion of acetic acid were lower and the concentrations of lactic acid and sugars were higher in the enzyme treated than in untreated silage. The concentration of butyric acid was equal to or lower than in formic acid treated silage in all experiments. Cell wall constituents were degraded in the silo by cellulase and thus more energy was available for lactic acid bacteria. With increasing levels of cellulase application, the disappearance oforganic matter (OM) from nylon bags incubated for 1 to 12h in the rumen of a dairy cow increased significantly (P< 0.001, PcO.Ol) in two different experiments. The level of cellulase had no effect on the ratio N/OM disappeared from nylon bags. Digestibilities of OM in sheep were 75.8, 76.5 and 77.2 % for untreated, formic acid and enzyme treated silage (P>0.05). On untreated silage diet N reten- tion tended to be reduced in sheep compared with formic acid or enzyme treated silages. In growing bulls there were no differences in feed intake, growth rate, feed conversion rate, car- cass characteristics or OM digestibility between formic acid (41/t) and enzyme treated silage. Index words: silage, additives, enzymes, fermentation, degradability in sacco, growing cattle Introduction Silage additives used to influence fermen- tation can act either as stimulants by encour- aging lactic acid fermentation or as inhibitors by inhibiting completely or partially microbial growth. Fermentation stimulants can be clas- Present address: Agricultural Research Centre, SF-31600 Jokioinen, Finland sified as bacterial cultures or as carbohydrate sources. The effects of bacterial inoculants on silage fermentation have been variable and in many experiments no benefits at all have been seen. Woolford and Sawczyc (1984) did not find any notable influence on the rate of acidi- fication or promotion of homolactic fermen- tation by selected cultures of lactic acid bac- 285 JOURNAL OF AGRICULTURAL SCIENCE IN FINLAND https://www.c-info.fi/en/info/?token=A-p3VrbVc0mBTwIH.XNPoiBlTlMX9u8w6MCfNLw.-9u8YPOs8GRTCeYnG-Y4fcJ8TN9ZCzM3yd3ZRaEobkucSwbmYIZgsLzaior2UVesDPneVvI9At5QgNR_RxE-EyDrJvtTWf8aF1YkdTvPNzQ5IfnwYgEHvEr8s56uh_-orEJ6UGkQFyKERvRlf_J_nxyJ2vTtjeY9LOKrvkpIFbisbrofjJPJ2R-WzjcuCphke_13FrahwYjZYXzqr71ooX_PcooTu-AdP5HbUw teriä and these also tended to increase the amount of soluble protein and deamination. Carbohydrate rich materials such as sugars, molasses and cereals have been added to si- lage crop in order to supply energy for lactic acid bacteria. In his review, McDonald (1981) notes that molasses has been shown to decrease pH and the amount of ammonia N in silage. Castle and Watson (1985) did not record any differences in the performance of dairy cows fed formic acid or molasses treated silages. A disadvantage of using carbohydrate rich materials is the relatively high concen- trations required. Forage plants have a large reserve of car- bohydrates, but these occur mainly in the form of polysaccharides which cannot be used as an energy source by lactic acid bacteria in silage. In theory, theaddition of enzymes able to hydrolyse forage polysaccharides would increase the fermentation capacity of silage by releasing fermentable substrate. The cellu- lolytic enzymes would also promote cellular breakdown and render cell contents more ac- cessible to silage microflora. The objectives of the present studies were to investigate the effect of enzymes on silage fermentation, degradation rate in sacco, di- gestibility in sheep and on the performance of growing bulls. The silages were made in labo- ratory, pilot and farm scale silos. Materials and methods In two experiments 5 kg of chopped grass was wilted for 4 h in order to increase dry matter (DM) content to about 25 %, and then ensiled into 25 I laboratory silos. The herb- age contained 19.0 °7o of crude protein and 22.6 % of crude fibre in Exp. 1 and 15.6 and 25.2 % in Exp. 2. The treatments were as fol- lows: untreated control, formic acid (5 1/t), glucose oxidase (GO) and GO + 100, 200, 400, 800, 1600 or 3200 ml/t of cellulase. Formic acid was applied as AIV II solution (80 °7o formic acid, 2 % orthophosphoric acid) in all the experiments. After 59 d (Exp. 1)and 86 d (Exp. 2) the silos were opened and analyzed for DM, organic matter (OM), ash, nitrogen (N), true protein, ether extract, crude fibre, neutral detergent fibre (i DF), acid detergent fibre (ADF), lignin (ADL), pH, sugars, lactic acid, volatile fatty acids (VFA), ammonia and soluble N. Nylon bag (porosity 40 /rm) incubations were made in the rumen of a dairy cow fed at the level of 2.8 times maintenance, with for- age 60 % of DM. The silages incubated in triplicate were withdrawn after 2,5, 9, 24 and 48 h in Exp. 1 and after 1,3, 6, 12, 24 and 48 h in Exp. 2. Residues were analysed for DM, OM and N and some treatments also for NDF. The values for degradation constants were calculated using the formula of orskov and McDonald (1979) p = a + b(I e~ cl). For the digestibility experiment (Exp. 3) si- lages were made in 3 m 3 glass fibre silos from first cut timothy sward. The grass was wilted for 4 h except the direct cut formic acid treated silage. The treatments were untreated control, formic acid direct cut (4.5 1/t), formic acid pre-wilted (4 1/t), GO + hemicellulase (150 ml/t) + cellulase (200 ml/t) (Enzyme A) and hemicellulase (150 ml/t) + cellulase (200 ml/t) (Enzyme B). The digestibility trial was carried out with 5 sheep using a 5 X 5 Latin square design. The material for the feeding trial (Exp. 4) was second cut timothy red clover sward (red clover 35—40 % of DM). After wilting for 4 h the grass was ensiled into a 250 t bunker silo either with formic acid (4 1/t) or 2—3 d later with enzyme mixture (GO + cellulase 150 ml/t). A feeding trial of 210 d was carried out with 24 bulls using 2x2 factorial design. Both silages were supplemented with barley or barley + fishmeal (FM). Barley was fed at the level of 40 g/kg W 0 on an air dry basis. In theFM group 0.25 kg of barley was replaced by FM. In vivo digestibility was determined at the liveweight of 250—300 kg using acid- insoluble ash as digestibility marker (Van Keulen and Young 1977). 286 6 Results and discussion The breakdown of various fibre fractions increased during the ensilage period with in- creasing level of cellulase (Fig. 1). The effect was proportional to the square root of cel- lulase activity. In Exp. 2 with longer ensilage period and more mature material, the effect of cellulase on ADF, cellulose and crude fibre was like that in Exp. 1, but the effect on NDF was greater indicating the hydrolysis of hemi- cellulose and the hemicellulose activity of the enzyme mixture. In the normal ensilage process the breakdown of hemicellulose is caused also by plant enzymes and later, when pH is low, by acid hydrolysis (McDonald 1981). All the silages were well preserved. Formic acid almost completely inhibited fermentation in the laboratory silos (4 kg). Presumably at the high levels of addition the bacteriostatic action of formic acid and consequent suppres- sion of lactic acid production allowed the pH to increase in response to the release of buf- fering constituents from plant tissue. With increasing levels of cellulase the concentration of residual sugars increased and there was also some increase in the concentration of lactic acid. No effect on true protein contents or the ratio ammonia N: total N was found with in- creasing levels of cellulase, but the ratio solu- ble N: total N increased, presumably because of the breakdown of cell walls and subsequent release of cell contents. The disappearance of OM from nylon bags increased with the level of cellulase addition, especially after shorter incubations (Figs. 2 and 3). After an incubationof 48 h there were no differences in the extent of OM degrada- tion between the treatments. The very rapidly degradable fraction a increased with the level of cellulase and in time degradable fraction Fig. I. The effect of enzyme level on NDF A , ADF ■. cellulose • and crude fibre □ contents of si- lage (% of DM) in Exp. 2. Fig. 2. The effect of enzyme level on silage OM disap- pearance (% of total OM) from the nylon bags in Exp. 1. 287 b decreased. The effect of treatments on c (the rate of degradation of fraction b) was incon- sistent in Exp. 1, but in Exp. 2 all enzyme treated silages had a higher rate of OM de- gradation than untreated or formic acid treated silages. Cellulase treatment had the same pattern of response on N as on OM dis- appearance. The ratio N:OM that disappea- red during the first 2 or 3 hours was 55—63 g N/kg OM in Exp. 1 and 43—51 g in Exp. 2 being slightly lower in formic acid treated silage than in other silages. This ratio was not affected by the level of cellulase. The effect of increased N loss after short incubations on N or microbial N flow might be negative. However, Chamberlain et al. (1982) did not record any response of N flow into the duo- denum when the level of formic acid was in- creased from 0 to 5.9 1/t, and the undegrad- able fraction of silage N was increased only from 22 to 28 %. It seems that partial limita- tion of proteolysis can be achieved in the silo by the addition of high levels of formic acid, but protein spared from hydrolysis in the silo seems to be degraded by the rumen microbes. Part of the response in the OM loss can be explained by increased N loss with higher levels of cellulase. However, the loss of N-free OM after incubations for 3, 6 and 12 h in- creased from 33.0 to 41.7 % (P< 0.001), from 39.9 to 46.8 % (PcO.001), and from 48.4 to 54.0 % (P<0.01) in Exp. 2. With the highest level of cellulase applica- tion the potential degradability of NDF was decreased (Table 1), but this could be ex- plained by the breakdown of cell walls in the silo. Lag time was reduced by cellulase treat- ment, but in Exp. 1 with young grass a very high level of cellulase was needed to achieve this response. By contrast, with more mature material in Exp. 2 the same response was found with the level of 400 ml/t. In Exp. 3 enzyme treatment decreased pH Table 1. The degradation constants of NDF disappearance from the nylon bags in Exp. 1 and Exp. 2 Exp. 1 a+ b c r Lag time (h) Res. s.d. Control 76.0 .0608 .9999 2.6 0.92 Formic acid 74.4 .0677 .9977 3.0 3.23 GO +C 400 ml 73.5 .0620 ,9998 2.5 1.54 GO +C 3200 ml 65.5 .0644 .9987 0.9 2.96 Exp. 2 Control 68.3 .0385 .9984 2.8 4.21 Formic acid 62.5 .0376 .9996 4.2 1.57 GO +C 400 ml 61.5 .0403 .9993 1.7 1.26 GO+ C 3200 ml 55.5 .0321 .9990 1.6 2.30 GO = glucose oxidase, C = cellulase Fig. 3. The effect of enzyme level on silage OM disap- pearance (% of total OM) from the nylon bags in Exp. 2. 288 Table 2. Fermentation quality of experimental silages in Exp. 3. Formic acid Formic acid Control Enzyme A Enzyme B SEM direct-cut pre-wilted pH 4.06c 3.98' 4.40“ 3.95' 3.96' 0.03 In dry matter, % Sugars 1.53d 3.62' 0.17' o.Bo* 0.89* 0.18 Lactic acid 6.34' 6.38' 9.31“ 13.26' 13.05' 0.54 Acetic acid 1.82* 1.38' 3.03' 2.33' d 2.33' d 0.17 Propionic acid 0.08 0.07 0.03 Ratio lactic:acetic 4.1 ab 4.8 ab 3.1 b 5.7a 5.6 a 0.47 In total N, % Ammonia N 6.3* 5.2' 11.3' 7.2“ 7.2“ 0.35 Soluble N 60.5““ 58.9“ 70.4' 67.9'“ 2.0 Means with different letters were significantly different a, b (P<0.05), c, d, e (P<0.01) Table 3. Feed intake and the performance data in growing bulls in Exp. 4. Formic Enzyme Silage Protein Contr. FM Contr. FM FA Enzyme Contr. FM SEM DM intake 1, kg/d Silage 3.68 3.81 3.57 3.69 3.74 3.63 3.63 3.75 0.11 Barley 2.29 2.09 2.23 2.11 2.19 2.17 2.26 2.10 0.07 Fishmeal 0.22 0.22 0.11 0.11 0.22 0.00 Total 5.97 6.12 5.80 6.02 6.05 5.91 5.88 6.07 0.16 DM g/kg W° 15 86.7 88.2 86.8 85.8 87.5 86.3 86.8 87.0 1.2 Initial weight, kg 181.0 175.5 179.3 178.0 178.2 178.7 180.2 176.8 12.1 Final weight, kg 401.0 406.2 386.7 411.3 403.6 399.0 393.8 408.7 11.5 Daily gain, g 1064 1137 1004 1141 1100 1072 1034» 1139» 35 Feed conversion Kg DM/kg gain 5.63 5.40 5.85 5.31 5.52 5.58 5.74 5.36 0.18 ME MJ/kg gain 63.6 61.2 64.9 59.2 62.4 62.1 64.2 60.2 2.1 Carcass weight, kg 205.2 206.7 197.0 210.8 205.9 203.9 201.1 208.8 7.2 Dressing-% 50.8 50.7 50.9 51.2 50.7 51.0 50.9 50.9 0.5 1 On oven DM basis, FM = fishmeal, FA = formic acid Means with different letters significantly different a, b (P<0.05) and the concentrations of acetic acid and am- monia N and increased the concentrations of lactic acid and residual sugars compared with untreated control (Table 2). Enzyme treatment encouraged homofermentative lactic acid bac- teria and theratio lactic acid: acetic acid was the most favourable in enzyme silages. Ap- parently in the absence of fermentation inhib- itors the sugars from cell wall hydrolysis are used mainly for lactic acid production. The digestibility of OM was 76.2, 76.5, 75.8, 77.1 and 77.2 % for direct cut formic acid, pre-wilted formic acid, untreated, en- zyme A and enzyme B treated silages and N retentions were 2.5, 2.9, 1.4, 2.0 and 2.9 g/d, respectively. Both silages in Exp. 4 were well preserved in the bunker silo. The enzyme treated silage had a higher concentration of lactic acid (11.2 vs. 8.6 %), acetic acid (3.8 vs. 2.1 ®7o), am- monia N (8.9 vs. 5.9 % of total N) and solu- ble N (49.4 vs. 41.8 % of total N) and lower concentration of sugars (1.1 vs. 0.5 %). However, DM intakes of the two silages were similar on oven DM basis (Table 3). Differ- ences between the silages in growth rate, feed conversion and carcass characteristics were small and insignificant. FM, on the other hand, improved significantly (P<0.05) the liveweight gain (Table 3). One animal in the 289 enzyme silage + barley group had leg in- juries, which partly explains the insignificant (P>0.05) interaction between the type of silage and protein supplementation. Enzyme treatment reduced significantly (P <0.001) the digestibility of crude fibre and improved (P< 0.001) that of the ether extract compared with formic acid treated silage. The lower crude fibre digestibility is most likely explained by enzymatic cell wall digestion in the silo, but probably the slightly later har- vesting time also had some effect. It seems that the cellulase system first attacks the most easily digestible fraction of fibre in the silo and the fraction which is left is less digestible in the animal than the original fibre. There was a significant (P<0.01) interaction between the silage and protein supplementation: FM clear- ly improved the digestibility of enzyme treated silage, but had a slightly negative effect with formic acid silage. Possibly enzyme silage re- quires a different supplement than acid treated silage to optimize cell wall digestion in the rumen. In conclusion, enzyme treatment encour- aged lactic acid fermentation and also reduced ammonia N compared with untreated control silage. No adverse effects relative to formic acid silage were found on N retention in sheep or feed intake and liveweight gain in growing bulls. The advantages of biological additives over acids are that they are safer to handle, less corrosive to machinery and easier to trans- port. References Chamberlain, D.G., Thomas, P.C. & Wait, M.K. 1982. The rate ofaddition of formic acid to grass at ensilage and the subsequentdigestion of the silage in the rumen and intestines of sheep. Grass and Forage Sci. 37: 159—164. Castle, M.E. & Watson, J.N. 1985. Silage and milk production: studies with molasses and formic acid as additives for grass silage. Grass and Forage Sci. 40: 85—92. McDonald, P. 1981. Biochemistry of silage. 226 p. Chi- chester. John Wiley & Sons. ORSKOV , E.R. & McDonald, I. 1979. The estimation of protein degradability in the rumen from incubation measurements weighted according to rate of passage. J. agric. Sci., Camb. 92; 499—503. Van Keulen, J. & Young, B.A, 1977. Evaluation of acid- insoluble ash as a natural marker in ruminant digest- ibility studies. J. Anim. Sci. 44: 282—287. Woolford, M.K. & Sawczyc, M.K. 1984. An investiga- tion into the effect of cultures of lactic acid bacteria on fermentation in silage. 2. Use of selected strains in laboratory scale silages. Grass and Forage Sci. 39: 149—158. Ms received December 18, 1985 SELOSTUS Entsyymit nurmirehun säilöntäaineena. Vaikutus käymiseen, sulavuuteen lampailla, rehun hajoamisnopeuteen in sacco sekä tuotantoon lihanaudoilla Pekka Huhtanen, Kari Hissa, Seija Jaakkola ja Esko Poutiainen 1 Helsingin yliopisto, kotieläintieteen laitos, 00710 Helsinki Homeiden tuottamaa glukoosioksidaasia ja sellulaasia käytettiin nurmirehujen säilöntäaineena laboratorio- Nykyinen osoite: Maatalouden tutkimuskeskus, 31600 Jokioinen (25 1), pilot (3 m') ja maatilasiiloissa (250 1) suoritetuissa kokeissa. Entsyymikäsiteltyjen rehujen pH ja etikkahap- popitoisuus olivat alhaisempia ja maitohappo- ja sokeri- pitoisuus korkeampia kuin painorehulla 3 m’:n siiloissa. Voihappopitoisuus oli kaikissa kokeissa entsyymirehuil- 290 la sama tai alhaisempi kuin muurahaishapporehulla (AIV-2). Sellulaasi pilkkoi rehun solunseinämäaineita sii- lossa ja vapautti siten sokereita maitohappobakteereiden käytettäväksi. Kahdessa eri nailonpussikokeessa lypsyleh- mällä orgaanisen aineen pötsihajoavuus kohosi merkit- sevästi (P0.05). Typen pi- dättyminen oli painorehulla hieman alhaisempi kuin muu- rahaishapolla tai entsyymeillä säilötyillä rehuilla. Liha- naudoilla rehun syönnissä, kasvussa, rehun hyväksikäy- tössä, teuraslaadussa ja orgaanisen aineen sulavuudessa ei ollut eroa AIV-2:lla (41/t) tai entsyymeillä säilöttyjen rehujen välillä. 291