To feed the raising human population, the Food and Agriculture Organization of the United Nations predicts that total agricultural production from crops and animals will need to be doubled than in 2005. The growth in income in low- and middle-income countries would accelerate a dietary transition towards higher consumption of animal products, requiring commensurate shifts in output and adding pressure on livestock production systems (Alexandratos and Bruinsma 2012). In South Asia and sub-Saharan Africa, animal production would need to more than double by 2050 to meet increased demand. However, constraints in availability of quality feed have pushed challenges to the animal production systems. In Asia and many parts of the world, animals are mostly raised on agricultural crop residues. However, the nutritional value of crop residue is very low to maintain production (Devendra and Sevilla 2002). In India, cereal crop residues contribute about 70% of overall feed resources used for animal feeding, followed by green fodder (23%) while concentrated feeds account for only 6% per cent (Swaminathan 2007). As grain production is also important for human population, improving the quality of crop residue to increase nutrient bioavailability is one of the major alternates of increasing animal productivity. Therefore, emphasis has been given on development of good quality crops without compromising grain production and improve fodder quality.

Sorghum (Sorghum bicolor L.) plays a significant role in global crop production. Besides food grain and fodder crop production, it is also used for biofuels and alcoholic beverages manufacture (Bhoyar and Thakare 2009). Sorghum stovers after harvesting grain have been used in India and other Asian countries for large ruminant feeding, especially for buffaloes (Devendra 1997). However, the quality of sorghum stovers are very poor, even for maintenance of animals (Valli et al., 2019). Therefore, several novel sorghum cultivars have been developed through plant breeding programmes, which have the high grain and forage yield with low lignin content (AICSIP 2011). Therefore, the present study was aimed to examine the quality and potential feeding value of three novel genotypes of sorghum stovers (normal sorghum, sweet sorghum and brown midrib sorghum) by in vitro fermentation technique with buffalo rumen fluid. 

The present study was carried out during the year 2015-16, in the Division of Animal Nutrition and Feed Technology, ICAR-Central Institute for Research on Buffaloes, Hisar, Haryana, India (29.1203_N, 75.8069_E).
 
Collection of stover samples
 
Out of many novel cultivars of sorghum (Sorghum bicolor L.) developed by Indian Institute of Millets Research, Hyderabad by plant breeding programme, three different genotypes viz.  normal sorghum (CSV- 27), brown midrib (bmr) sorghum (SPV- 2018) and sweet sorghum (CSH- 22SS) were collected and a comparative study was undertaken to assess their feeding value. For each genotype, four stover samples were collected from two different fields and dried and ground to 1 mm particle size. Representative samples for analysis and fermentation study were prepared by mixing samples in equal proportion from different fields within each cultivar.
 
Collection on rumen liquor
 
Rumen liquor (RL) was collected from three rumen fistulated adult male Murrah buffaloes (Live weight, 550±40 kg) before feeding and watering. The animals were maintained on ad libitum diet of wheat straw with limited amount of standard concentrate mixture and green fodder as maintenance diet. The sample was collected from different locations of rumen and at different depths, from each of fistulated buffalo to get a representative and homogenous sample. Equal volume of RL from each buffalo was pooled and strained through four layered muslin cloth into a pre-warmed thermos flask (1 L capacity) which was previously flushed with carbon dioxide to maintain anaerobic condition and immediately brought to laboratory for in vitro studies.

In vitro fermentation study
 
Feed sample (200 mg±5 mg) for in vitro fermentation study was weighed accurately in a scoop and transferred carefully at the bottom of the 100 ml calibrated glass syringes. The syringes were gassed with oxygen free carbon dioxide, after placing pistons lubricated with petroleum jelly. The quantities of buffer solution, macro-mineral solution and micro-mineral solutions were prepared and buffered mineral solution, required for in vitro study, was prepared by mixing appropriate amount of the same with distilled water (Menke and Steingass 1988). The buffered mineral solution was mixed with strained rumen fluid (2:1) under anaerobic condition and dispensed (30 mL) through automatic dispenser into the syringes containing stover samples. After recording the initial volume, the syringes were shaken gently and placed vertically in an incubator at 39°C for 24 h. Three replicates for each treatment were kept and three syringes with only buffered rumen fluid were run as blank. The syringes were shaken gently at different intervals throughout the incubation period.
 
Estimation of total gas and methane production
 
During the period of incubation, the cumulative gas production in each syringe was recorded by displacement of piston of graduated glass syringes at 4, 8, 12, 18 and 24 h. The gas produced in blank syringes were subtracted from the gas produced in the syringes with feed samples to get net gas production (Menke and Steingass 1988) over the time. After 24 h of incubation, the head space gas samples (200 µL) from each syringe was taken in an airtight Hamilton syringe and injected into gas chromatograph (NUCON- 5700) fitted with SS column packed with Porapak-Q and flame ionization detector (FID) for determination of methane concentration against the standard mixture (50:50) of methane and carbon dioxide. The total methane production was calculated by multiplying the methane concentration in head space gas with total gas produced.
 
In vitro dry matter degradability (TDDM)
 
After termination of incubation, syringe contents were transferred individually to 500 mL spoutless beakers and refluxed for 1 h. Repeated washings with neutral detergent solution (VanSoest et al., 1991) were performed for complete transfer. The residue was collected by vacuum filtration through sintered crucible (G1) and dried in hot air oven at 105°C for 24 h for determination of in vitro dry matter degradability.
 
Volatile fatty acid estimation
 
After the end of incubation, supernatant (1 mL) of each syringe contents were mixed with 0.20 mL metaphosphoric acid (25%) in a micro centrifuge tube and the mixture was incubated at room temperature for 2 h. The tubes were then centrifuged at 5000 × g for 10 min and supernatant was collected and stored at -20°C for subsequent analysis. A sample of 1 µL was injected into the gas chromatograph (NUCON- 5700) fitted with chromosorb 101 glass column and flame ionization detector (FID) to estimate individual volatile fatty acids productions (Cottyn and Boucque 1968). The initial and final temperature of column oven was set to 170°C and 230°C, respectively with a ramping rate at 3°C/min. The temperature of injector and detector of FID were fixed at 240°C and 250°C, respectively.
 
Ruminal enzyme activities
 
For extraction of ruminal enzyme, the syringe whole content (30 mL), after the end of incubation, was transferred to a 100 mL conical flask and mixed with 5 mL each of carbon tetrachloride and lysozyme solution (0.1 M, pH 6.8). After incubation of mixtures at 40°C for 3 h, sonication was done at 4°C for 6 min in ice bath. The samples were then centrifuged at 17000 × g for 20 min at 4°C and supernatant was used for the estimation of ruminal enzyme activities. Carboxymethyl cellulase (CMCase) and xylanase activities were measured spectrophotometrically at 575 nm (Miller 1959). The activity acetyl esterase was estimated according to Huggins and Lapides (1947). The procedure of Shewale and Sadana (1978) was followed for measuring β-glucosidase activity spectrophotometrically. The enzyme activities (mIU) were expressed (nano mole of reducing sugars or p-nitrpophenol released per mg of protein) after estimating the protein content of the extracted samples (Lowry et al., 1951).
 
RT-PCR based quantification of rumen microbial abundance
 
After completion of incubation (24 h), the whole syringe contents were used for extraction of metagenomic DNA (Yu and Morrison 2004). After evaluation of quality and quantity of DNA by agarose gel (1%) electrophoresis and nano-drop spectrophotometer (Invitrogen Corporation, Carlsbad, CA, USA), the samples were stored at -20°C for further analyses. The abundances of total bacteria and archaea were quantified using SYBR Green-based quantitative real time-PCR (qPCR) using an Applied Biosystem (Step One Plus) real time PCR machine (Patra and Yu 2014).
 
Chemical analysis
 
The stover samples of normal sorghum, bmr sorghum and sweet sorghum were analysed in triplicate (AOAC 1995) to determine dry matter (DM), organic matter (OM), crude protein (CP) and ash contents. The methods of VanSoest et al., (1991) were followed to determine the neutral detergent fibre (NDF), acid detergent fibre (ADF) and acid detergent lignin (ADL). Hemicellulose and cellulose contents were calculated by subtracting ADF from NDF and ADL from ADF, respectively.
 
Statistical analysis
 
Data were subjected to analysis of variance (ANOVA) using general linear model procedure (SPSS 17.0) as a completely randomized design. Differences among means were tested using Duncan’s multiple range tests (Snedecor and Cochran 1994).

Chemical composition of all the three genotypes of stovers varied significantly (P<0.05). The stover of brown midrib (bmr) sorghum contained highest organic matter, followed by normal sorghum and lowest in sweet sorghum (Table 1). However, the crude protein content was lowest in bmr sorghum and highest in sweet sorghum with normal sorghum as intermediate one. The neural detergent fibre (NDF) was highest in bmr sorghum and lowest in sweet sorghum. The stover sample of bmr sorghum contained lowest (P<0.05) acid detergent fibre (ADF) among the three cultivars examined, resulted highest (P<0.05) hemicellulose (37.72%). Lowest acid detergent lignin (ADL) was reported in stovers of brown midrib cultivar and highest in sweet sorghum cultivar. The distribution of leaves, stems and their ratio as well as structure of plant tissue determine the variation in chemical composition (Dhadheech et al., 2000; Marsalis et al., 2010). The fibre content more in stem than leaves and it depends on the characteristics of cultivars (Neves et al., 2015; Scully et al., 2016). Introduction of bmr gene to sorghum cultivar, which exhibits brown pigmentation in midrib of leaf as well as stem pith, affect fibre concentration and lignification (Oliver et al., 2005). The present examination (Table 1) corroborates earlier studies (Green et al., 2014; Scully et al., 2016) of lowest ADF and lignin content in bmr sorghum than other cultivars, suggesting higher nutritional potential as animal feed. The true degradability of bmr sorghum stover was higher (P<0.001) than both sweet and normal sorghum (Table 2). Similarly, organic matter digestibility (OMD, %) and metabolizable energy (ME) values for bmr sorghum was higher (P<0.001) than both the normal and sweet sorghum stovers, no differences (P>0.05) were found between sweet and normal sorghum stovers in these parameters. The higher hemicellulose contents with lower fibre and lignin contents for the bmr sorghum stover could have provided better substrate for microbial colonization and improved nutrient degradability. Further, it was recorded the higher (P<0.05) number of total bacterial population (Table 4) resulting high rate of fermentation and degradability than the other stover genotypes. Similar to our study, many researchers (Bean and McCollum 2006; Beck et al., 2007) also reported higher fermentability of bmr sorghum than others due to low lignification.
 

 

 
The rate of gas production (Fig 1) as well as total gas production (Table 2) from stovers of bmr sorghum was higher (P<0.01) than both stovers of sweet and normal sorghum. However, no difference (P>0.05) was observed between sweet and normal sorghum stovers. The higher hemicellulose and low fibre and lignin contents of bmr sorghum stovers provided more substrate for rapid fermentation, which ultimately represented increased gas production (Fig 1). As many researchers (Zerbini and Thomas 2003; Getachew et al., 2004) suggested negative relationship between fibre and lignin contents with degradability and gas production. However, no difference (P>0.05) in gas production between sweet and normal sorghum stovers, although differences in fibre and lignin contents in these two cultivars, could be due to characteristics of fibre and lignin with their association pattern (Jung and Allen 1995; Reddy and Yang 2007).
 

       
The MBP (mg/g DM) was similar (p=0.513) in all the three cultivars of sorghum stovers. As the gas production and substrate degradability differed among the cultivars, the partitioning factor (PF) also varied (P<0.05). PF, the efficiency of microbial biomass production, measured as the substrate truly degraded per unit of gas produced was lower (p<0.05) in stover of bmr sorghum than normal and sweet sorghum cultivars, suggesting partitioning of nutrients towards short chain fatty acids (SCFA) and gas production rather than MBP. Shifting of fermentation towards more SCFA and gas production, thereby low PF value by supplementation of Saccharomyces cerevisiae or organic acid salts were demonstrated (Elghandour et al., 2017a; Elghandour et al., 2017b). The higher (p<0.001) degradability, gas production of bmr sorghum cultivars were also reported by other workers (Aydin et al., 1999; Bean et al., 2013).
       
Methane concentration in head space gas (Fig 2) after 24 h fermentation of stovers were lowest (P<0.05) in sweet sorghum and highest in normal sorghum with bmr sorghum remained intermediate. However, total methane production (mL/g DM incubated or mL/ g DDM) was greater (p<0.05) in bmr sorghum than other stovers, with lowest in sweet sorghum and intermediate in normal sorghum stovers (Table 3). The increased methane production during fermentation of bmr sorghum stovers was described by higher gas production and true degradability than other cultivars. Although methanogenic archaeal population remained comparable (P>0.05) for all the stover cultivars (Table 4), more hemicelluloses and less fibre in bmr sorghum stovers offered substrate to mixed rumen microbes resulting higher fermentation and methane production. Very limited information are available on methane production for sorghum stovers varieties, however, many researchers (Mahmood and Honermeier 2012; Mahmood et al., 2013; Thomas et al., 2013) reported higher methane production by fodders and silages of sorghum, corn and barley with more degradability and gas production.
 

 

 

 
All the three major volatile fatty acids (acetate, propionate and butyrate) production were highest (p<0.05) in bmr sorghum stovers (Table 3), demonstrating better feeding value than the others to the animals. Although, acetate concentration was higher (P<0.05) in stovers of normal sorghum than sweet cultivars, the propionate and butyrate productions were comparable (P>0.05) between the stovers of these two cultivars. The greater degradability and gas production with lower fibre and lignin contents in bmr sorghum stovers evidenced higher (p<0.05) volatile fatty acids production. While studying fermentation of bmr sorghum for biohydrogen production, Prakasham et al., (2012) reported higher fermentation and volatile fatty acids production, whereas reduced VFA production was reported (Wedig et al., 1987) in normal sorghum cultivars. The specific activities of CMCase and xylanase in fermentation medium after 24h were highest (p<0.05) in bmr sorghum stover followed by normal and sweet cultivars. However, β-glucosidase activity remained comparable (p>0.05) for all the three stover cultivars (Table 4). The acetyl esterase activity remained higher (p<0.05) in fermentation medium of both bmr and normal sorghum than sweet cultivars, however, no difference (P>0.05) was observed between these two. The greater CMCase and xylanase activities on fermentation of bmr sorghum stovers could be due to improved fibre quality and lower lignin content (Table 1) as evidenced by colonization of abundant bacteria (Table 4). Present study concurs with the earlier reports (Sugoro et al., 2015; Vallejo et al., 2016) that availability good quality substrate increased ruminal fibrolytic enzymatic activities (Chen et al., 1995).

The study encompasses the variations among the stovers of three different sorghum genotypes in terms of chemical compositions and fermentation characteristics. The assessment of feeding value of stovers by their chemical compositions, gas production, degradability, microbial colonization and ruminal enzyme productions revealed that the quality of stovers from bmr sorghum cultivar is better than the other genotypes.

The authors declare no conflict of interest.

The research facilities provided by the Directors of ICAR- Central Institute for Research on Buffaloes, Hisar and ICAR-Indian Institute of Millets Research, Hyderabad is duly acknowledged.