Two incubation runs were conducted with Rusitec fermenters to investigate the effects of three additive treatments (mixed fibrolytic enzymes from Trichoderma longibrachiatum (FE), disodium fumarate (FUM) and both additives (MIX)) on rumen microbial growth and fermentation of a grass hay:concentrate (600 : 400 g/kg DM) substrate. Each fermenter received daily 20 g substrate DM. Application rate (per g substrate DM) was 34·3 endoglucanase, 0·57 exoglucanase, 24·7 xylanase and 5·51 amylase units for FE and 30 mg fumarate for FUM. MIX fermenters received both additives. Both FE and MIX increased (P < 0·05) daily production of acetate, butyrate and methane, substrate DM and fibre disappearance at 6 and 48 h incubation, daily flow of microbial-N, and the microbial colonisation of substrate at 6 h incubation. Compared to FE, MIX treatment increased (P < 0·05) propionate production by 28 % and decreased (P < 0·05) the acetate:propionate ratio, but no other differences between both treatments were found (P>0·05). Supplementing with FUM increased (P < 0·05) volatile fatty acid production by 11 % and decreased (P < 0·05) the acetate:propionate ratio, but did not affect (P>0·05) any other variable, thus suggesting that observed effects were due to fermentation of FUM itself. The lack of effects of FUM and the absence of differences between FE and MIX on most of the measured variables would indicate that beneficial effects found in MIX fermenters were mainly due to the action of FE. Combining FE and FUM as feed additives under the conditions of the present experiment did not further improve rumen fermentation, compared to FE alone.

The effects of disodium fumarate on microbial growth, CH4 production and fermentation of three diets differing in their forage content (800, 500 and 200 g/kg DM) by rumen micro-organisms in vitro were studied using batch cultures. Rumen contents were collected from four Merino sheep. Disodium fumarate was added to the incubation bottles to achieve final concentrations of 0, 4 and 8 mm-fumarate, and 15N was used as a microbial marker. Gas production was measured at regular intervals from 0 to 120 h of incubation. Fumarate did not affect (P>0·05) any of the measured gas production parameters. In 17 h incubations, the final pH and the production of acetate and propionate were increased linearly (P<0·001) by the addition of fumarate. Fumarate tended to increase (P=0·076) the organic matter disappearance of the diets and to decrease (P=0·079) the amount of NH3-N in the cultures. Adding fumarate to batch cultures tended (P=0·099) to decrease CH4 production, the mean values of the decrease being 5·4 %, 2·9 % and 3·8 % for the high-, medium- and low-forage diet, respectively. Fumarate tended to increase (P=0·082) rumen microbial growth for the high-forage diet, but no differences (P>0·05) were observed for the other two diets. These results indicate that the effects of fumarate on rumen fermentation depend on the nature of the incubated substrate, the high-forage diet showing the greatest response.

Fifteen potential precursors of propionate were tested for their ability to decrease CH4 production by ruminal fluid in vitro. Sodium acrylate and sodium fumarate produced the most consistent effects in batch cultures, with 50 % of the added precursors being fermented to propionate and CH4 production decreasing by between 8 and 17 %, respectively. Additives were more effective when added as free acids, but this also decreased the pH and may have inhibited fibre digestion. Changing the dietary substrate from predominantly grass hay to predominantly concentrate had no influence on the effectiveness of acrylate and fumarate. In an in vitro fermentor (the rumen simulating technique, Rusitec) with a grass hay—concentrate (50:50, w/w) diet as substrate, both compounds were again fermented to propionate (33 and 44 % conversion to propionate, respectively). However, fumarate appeared more effective as a H2 sink compound. It was calculated to capture 44 % of the H2 previously used for CH4 formation compared with a 22 % capture of H2 with acrylate. Fumarate also caused a stimulation in fibre digestion. Thus, sodium fumarate was the preferred propionate precursor for use as a feed ingredient to decrease CH4 emissions from ruminants.

Batch cultures of mixed rumen micro-organisms were used to study the effects of different concentrations of disodium fumarate on the fermentation of five concentrate feeds (maize, barley, wheat, sorghum and cassava meal). Rumen contents were collected from four Merino sheep fed lucerne hay ad libitum and supplemented with 300 g concentrate/d. Disodium fumarate was added to the incubation bottles to achieve final concentrations of 0, 4, 7 and 10 mm-fumarate. In 17 h incubations, the final pH and total volatile fatty acid production increased (P<0·001) linearly for all substrates as fumarate concentration increased from 0 to 10 mm. Propionate and acetate production increased (P<0·05), while the value of the acetate:propionate ratio decreased (P<0·05) linearly with increasing doses of fumarate. In contrast, l-lactate and NH3-N concentrations in the cultures were not affected (P>0·05) by the addition of fumarate. For all substrates, fumarate treatment decreased (P<0·05) CH4 production, the mean values of the decrease being 2·3, 3·8 and 4·8 % for concentrations of 4, 7 and 10 mm-fumarate respectively. Addition of fumarate did not affect (P>0·05) the total gas production. If the results of the present experiment are confirmed in vivo, fumarate could be used as a feed additive for ruminant animals fed high proportions of cereal grains, because it increased pH, acetate and propionate production and it decreased CH4 production.

The influence of sodium fumarate on rumen fermentation was investigated in vitro using batch and semi-continuous cultures of mixed rumen micro-organisms taken from three sheep receiving a basal diet of hay, barley, molasses, fish meal and a mineral–vitamin supplement (500, 299·5, 100, 91 and 9·5 g/kg DM respectively). Batch cultures consisted of 10 ml strained rumen fluid in 40 ml anaerobic buffer containing 200 mg of the same feed given to the sheep. Sodium fumarate was added to achieve a final concentration of 0, 5 or 10 mmol/l, as a result of the addition of 0, 250 or 500 μmol, equivalent to 0, 200 and 400 g/kg feed. CH4 production at 24 h (360 μmol in the control cultures) fell (P < 0·05) by 18 and 22 μmol respectively (SED 7·5). Total gas production was increased by the addition of fumarate without significant accumulation of H2. Substantial increases in acetate production (92 and 194 μmol; SED 26·7, P < 0·01) were accompanied by increases in propionate formation (212 and 396 μmol; SED 13·0, P < 0·001). Longer-term effects of fumarate supplementation on ruminal fermentation and CH4 production were investigated using the rumen simulation technique (Rusitec). Eight vessels were given 20 g basal diet/d, and half of them received a supplement of fumarate (disodium salt) over a period of 19 d. The response to the daily addition of 6·25 mmol sodium fumarate was a decrease in CH4 production of 1·2 mmol (SED 0·39, P < 0·05), equivalent to the consumption of 4·8 mmol H2, and an increase in propionate production of 4·9 mmol (from 10·4 to 15·3 (SED 1·05) mmol/d, P < 0·01). The inhibition of CH4 production did not decline during the period of time that fumarate was added to the vessels. Thus, the decrease in CH4 corresponded well to the fraction of the fumarate that was converted to propionate. Fumarate had no significant (P > 0·05) effect on total bacterial numbers or on the number of methanogenic archaea, but numbers of cellulolytic bacteria were increased (8·8 v. 23·9 (SED 2·49) × 105 per ml, P < 0·01). Fumarate also increased DM digestibility of the basal diet after 48 h incubation (0·476 v. 0·508 (SED 0·0123), P < 0·05). Thus, it was concluded that sodium fumarate may be a useful dietary additive for ruminants, because it diverts some H2 from CH4 production and because it is able to stimulate proliferation of cellulolytic bacteria and digestion of fibre.