In vitro gas production technique to evaluate associative effects among lucerne hay , rice straw and maize silage

In vitro gas technique was used to evaluate the associative effects between lucerne hay and rice straw (Trial 1) and between their mixture and maize silage (Trial 2) in terms of gas production and microbial protein. Signifi cant positive associative effect existed in gas production and microbial protein when lucerne hay and rice straw were mixed. The optimal ratio was 20:80 at which lucerne hay and rice straw had the highest associative effect. Markedly negative associative effects were found when lucerne hay/rice straw (20/80) mixture was mixed with maize silage. The associative effects based on gas production and microbial protein behaved inconsistent.


INTRODUCTION
The amount of nutrients that ruminant can extract from one feed can be modifi ed by the type and quantity of other feeds consumed simultaneously.This phenomenon is commonly referred to as associative effect.Prediction of associative effect is an important aspect of effi cient utilization of locally available feed resources.The effects of concentrates on the intake, digestion and utilization of relatively high quality forages have been thoroughly investigated (Doyle et al., 2005).However, there is limited information on the interactions that may occur when two or three forages are offered together.
Since high correlation existed between digestibility measured in vivo and predicted from an in vitro rumen gas production technique in combination with chemical composition, a considerable amount of researchers has used in vitro gas techniques to study associative effects of various types of feedstuffs, and examine infl uences on rumen fermentation (Liu et al., 2002;Getachew et al., 2003).The objectives of this study were to investigate the associative effects between lucerne hay (LH) and rice straw (RS) (Trial 1) and between their mixture and maize silage (MS) (Trial 2) using in vitro gas production (GP) technique.

Feedstuffs and experimental design
The feedstuffs (LH, RS and MS) were obtained from the experimental farm of Zhejiang University, Huajiachi Campus (China), and the CP of them were 17.1, 5.7 and 11.9% DM, respectively.
In trial 1, RS was incubated together with LH at proportions of 0, 20, 40, 60, 80 and 100% as a basis of dry matter and the optimal ratio was achieved when they had the highest associative effect.Trial 2 was conducted to investigate the interactions between MS and the mixture of LH and RS with highest associative effect.The levels of MS were 0, 25, 50, 75 and 100%.

In vitro gas production
The semi-automated in vitro GP technique was employed as described by Mauricio et al. (1999).About 750 mg substrates were accurately weighed into serum bottles (approximate 180 ml).Rumen fl uid was collected from three rumenfi stulated sheep before morning feeding.After being strained through a double-layer of cheesecloth under CO 2 in a water-bath at 39 o C, 10 ml rumen inoculum was added into each fl ask and mixed with 90 ml pre-injected buffer medium (Theodorou et al., 1994).The incubations were conducted in two consecutive runs, each involving quadruples of samples.Two runs were incubated for 24 and 96 h for sample collection and GP study, respectively.The GP parameters were estimated using the equation: (Ørskov and McDonald, 1979), where a, b and k were constants and GP was the gas production from the substrate at time t.

Measurement of in vitro fermentation parameters
The pH was measured using a portable pH meter (Model PB-20, Sartorius) after 24 h incubation and concentration of volatile fatty acids (VFA) was determined by gas chromatography (GC-2010, Shimadzu), as described elsewhere (Hu et al., 2005).Concentrations of microbial protein (MP) were estimated from the ratio of purines to N of isolated bacteria (Makkar and Becker, 1979).Yeast RNA was used as a standard.

Statistical analysis
Data were analysed using the general linear model (GLM) procedure of SAS (1996).The differences among means for treatments were tested using Duncan's new multiple range test.

RESULTS
The mean gas volumes at 24 h increased with increasing ratio of LH in mixtures (Table 1).Inclusion of 20% LH increased 24 h GP and GP rate by 81 and 116% (P<0.05),respectively, while no signifi cant differences existed in potential GP.Total VFA concentration increased with increasing levels of LH, while no signifi cant differences existed in MP among different LH inclusion levels.The GP showed positive associative effects (P<0.05),irrespective of the inclusion level of LH, being the highest (20.8%) at 20% of LH (Table 2).Similar to GP at 24 h, signifi cant positive associative effect existed in MP, being 31.7 and 34.5% when LH was at 20 and 40%, respectively.From above, it is indicated that optimal positive associative effect occurs in the mixture of LH and RS at 20:80.
Inclusion of 25% MS did not signifi cantly affect 24 h GP, potential GP, GP rate and total VFA, while higher levels (>50%) of MS induced notable increase in these parameters (Table 3).The MP increased with increasing levels of MS (P<0.05),being highest for MS alone (1.31 mg/ml).4. Difference (%) 1 between the values of gas production or microbial protein for maize silage and lucerne hay: rice straw (20:80) mixtures and that predicted from them separately

Parameters
Levels of maize silage, % 25 50 75 Gas production (GP) at 24 h -12.0 * -13.0 * -9.5 * Microbial protein (MP) -17.2 * -14.6 * -25.1 * 1 difference (%) = [(observed GP or MP -predicted GP or MP)/ predicted GP or MP] × 100 * signifi cant difference (P<0.05)Contrary to the effects between LH and RS, negative associative effects predominated in the interaction between MS and LH:RS mixture (20:80) (Table 4).Signifi cant negative associative effects, averaging -11.5 and -19.0%, existed in GP at 24 h and MP (P<0.05),respectively.DISCUSSION Fibre quality is extremely important for the complete action of cellulolytic bacteria, responsible for high rates of microbial colonization to the substrate, and resulting in effi cient energetic use of the evaluated feeds.Silva and Ørskov (1988) observed that the presence of a source of readily digestible cellulose and hemicellulose increased the numbers of ruminal fi brolytic microorganisms and, consequently, may improve digestibility of other less degradable fi bre sources in mixtures.The positive associative effect in GP and microbial protein when LH was mixed with RS may be attributed to the readily digestible fi bre of LH.
In this study, the associative effect in GP (20.8%) was most apparent when LH was mixed with RS at 20% (Table 2).Similar ratios (27 and 17% LH) have been reported when LH was mixed with treated wheat straw for dairy heifers (Atwell et al., 1991) and barley straw for ewes (Haddad, 2000).Since the LH is relatively scarce and expensive in less developed countries, it is cost-effective to supplement 20% to rice straw.
Effi cient microbial growth requires balance and synchronization between nitrogen and energy supply (Tebot et al., 2004).When LH was mixed with RS at 20%, the high protein solubility nature of LH may complement the defi ciency of RS and resulted in synchronized supply of nitrogen and energy to the ruminal microbes (Chernev et al., 2002).Additional inclusion of MS induced negative associative effect, which might result from that a certain number of fermentation end products in silages do not contribute any energy to microbial growth (Verbic et al., 1999) and hence the homeostasis being destroyed.In fact, the synchrony index (Is) of silages (mean Is=0.30) was generally lower than that of fresh forages (0.84) and hay (0.80) in his study.
Gas production is an indirect measure of substrate degradation and it is not always positively related to microbial mass production (Liu et al., 2002).In our study, the associative effects in terms of GP and MP were highest for 20 and 40% LH in trial 1 (Table 2) and lowest for 50 and 75% MS (Table 4), respectively.Such inconsistency has been demonstrated when associative effect was investigated between mulberry leaves and oilseed meal (Su, 2002).Since truly digested substrate is partitioned among VFA, gas and microbial biomass, it is necessary to incorporate different parameters to evaluate associative effect of mixtures and further information on fermentation products partitioning is needed.

IMPLICATIONS
Positive associative effect exists in in vitro gas production (GP) and microbial protein (MP) when lucerne hay (LH) was mixed with maize silage (MS), while the inclusion of MS in LH:RS (20:80) mixture induced markedly negative associative effects.The associative effects in terms of GP and MP behaved inconsistent, for which further study is needed.

Table 1 .
Gas production and fermentation parameters of lucerne hay and rice straw mixtures

Table 2 .
Difference (%) 1 between the values of gas production or microbial protein observed for lucerne hay and rice straw mixtures and that predicted from them fermented separately * signifi cant difference (P<0.05)

Table 3 .
Gas production and fermentation parameters of maize silage and lucerne hay: rice straw mixtures