Growing Temperature Influence Lignocellulose in Sorghum and Could Lead to a Significant Variation in Feed Value of Fodder Sorghum Genotypes

.

different types of sugars that are bound by α-1, 4 linkages (Brethauer et al., 2020). The third layer lignin is a polymer derived from the phenylpropanoid pathway in vascular plants and is formed as plants mature (Moore & Jung, 2001).
Lignin is less digestible and is, therefore, an undesirable component in forages due to its non-contributory role to nutrients availability but creates a physical barrier to enzymes in the digestive system which lowers the digestible energy (DE) value of forage (Moore & Jung, 2001). Lignification is a genetically controlled process and hence significant differences may occur in lignin concentration among genotypes (Moore & Jung, 2001). Large genotypic differences in lignin concentration have been reported in different crops. Arai-Sanoh et al. (2011) and Jahn et al. (2011) reported a variation in lignin content concentration in rice varieties which ranged from 1 to almost 12 percent. Lignification in sorghum has been shown to increase with increasing plant age (Grev et al., 2020;Nabi et al., 2006). Besides, abiotic stress factors such as extreme temperatures significantly affect the lignification in crops (Gindl et al., 2002;Xu et al., 2020;Zonetti et al., 2011). Because of the above, there is a need to evaluate sorghum genotypes for lignin content and how it relates to environmental factors. The current study was aimed at evaluating the influence of environment differentiated by growing temperatures on the lignocellulose content of fodder sorghum genotypes. It was hypothesized that variation in environmental temperature and stage of harvesting has no significant effect on the concentration of lignin, cellulose, and hemicellulose of sorghum genotypes.

Materials and Methods
The experiment was conducted in the March-September growing season in 2019 and repeated in the May-October season in 2020, in Nakuru and Baringo counties of Kenya. Egerton university research station (0°22′S; 35°55′E, 2286 m a.s.l. Nakuru county, Kenya) has a mean annual temperature of 14.9 °C and is located in agro-ecological zone (AEZ) III classified as an upper highland (UH) zone. The area has a bimodal pattern of rainfall with both long and short rains. The long rains occur around March and August while the short rains occur between October and December. The average rainfall is 1000-1200 mm per annum while the soils are deep sandy loam, mollic andasols with a pH range of 5.5-6.5. The second site was located in Rongai (0°23′N; 35°51′E, 1890 m a.s.l., Nakuru county, Kenya) with a mean annual temperature is 17.6 °C and annual rainfall of 900 mm. The soils at the site are classified as vitric andasols with silt loam texture and a pH range of 5.5-5.8 and moderate organic matter content. The third site was under irrigation at Kenya Agricultural and Livestock Research Organization (KALRO) in Marigat (0°46′N; 35°98′E, 1066 m a.s.l., Baringo County Kenya). The geographical data for the three locations are presented in Table 2. The soils are cambisols which are very deep, calcerous, saline, and sodic with a fine sandy loam-clay texture and acidic pH. The mean annual temperature in Marigat is 24.5 °C while minimum and maximum daily temperatures are 16.8 °C and 32.3 °C. Twenty sorghum genotypes that had the potential for fodders were evaluated. The characteristics included; high stem sugar content, brown midrib trait, stay green trait, and adaptation to the local environment (Table 1). Table 1. Sorghum genotypes evaluated   Sorghum genotype  Source  Attributes   IS11612  ICRISAT  -IS11442  ICRISAT  -IS25547  ICRISAT  -IS25557  ICRISAT  -IS11721  ICRISAT  -IS11838  ICRISAT  -IS2331  ICRISAT  -IS25563  ICRISAT  -IS9201 ICRISAT The experiment was arranged in a randomized complete block design (RCBD) and replicated three times at each of the sites. An experimental unit measured 4 m × 5 m and it accommodated five rows of sorghum plants. Sorghum seeds were sown at the rate of 10 kg ha -1 , at a spacing of 60 cm by drill and at a depth of 2 cm. Thinning was done later when the seedlings were at 20 cm height to maintain a spacing of 60 cm × 15 cm. Phosphorous fertilizer was applied during sowing at the rate of 60 kg ha -1 of P 2 O 5 . Top dressing was done by applying nitrogen at the rate of 60 kg N ha -1 when the crops were 4 weeks old. Weed control was manually done by in-plant tillage cultivation and pesticides to control shoot fly, aphids and fall army worm was applied when it was necessary. Bird scaring was done through engagement of bird scarers.
Standard laboratory procedures were followed in analyzing chemical properties of the soil. Sampled soils were air dried then sieved with 2 mm mesh and analyzed for pH (Soil: H 2 O), total N (Kjedahl method), total carbon (Walkley & Black, 1934), CEC (Chapman, 1965), and available P according to Okalebo et al. (2002). Exchangeable K was extracted with 1.0 M-ammonium acetate at pH 7 and measured by atomic adsorption spectrophotometer (©Analytic Jena). Soil analysis data is available supplementary materials.

Plant Tissue Sampling for Laboratory Analysis
Harvesting of plants for analysis was undertaken at two growth stages as described below, (i) Booting stage: when the flag leaf sheath was swollen, indicating emerging panicle.
(ii) Dough stage: when the grain had a dough-like consistency, the ideal stage for sorghum fodder harvesting as it is associated with optimum dry matter and nutritive value (Lyons et al., 2019).
Harvesting of samples at each of the growth stages above was done in the three middle rows of each experimental unit. From each of the three rows, four plants with the same height were harvested by cutting at the base, shredded using a fodder chopper, and the shredded material placed in a paper bag. The samples were weighed then labeled in the field and taken to the laboratory for oven drying at 65oC, milled, and stored for subsequent analysis

Data Analysis
Data obtained were subjected to analysis of variance (ANOVA) in R statistical package using a linear mixed model for RCBD. Treatment mean differences were separated and tested by Fisher's protected least significant difference (Lsd) at P = 0.05 significant level. Correlation analyses were done on individual treatment means using SAS (Statistical Software Version 9.1) to determine inter-character associations among lignin, plant height, and days to 50% flowering.
The following model was adopted; where, µ = Overall mean, Bi = effect due to the ith blocking, Gj = effect due to the kth sorghum genotype, B(L)ij = effect due to the interaction between ith block and jth location, LJ = effect due to the kth location, Sl = Effect of lth stage, Ym = effect of mth Year, GLkj = effect of the interaction between kth genotype and jth location, GSkl = effect of the interaction between kth genotype and lth stage, GYkm = effect of the interaction between kth genotype and mth year, SLlj = effect of the interaction between lth stage and jth location, SYlm = effect of the interaction between lth stage and jth year, LYjm = effect of the interaction between jth location and mth year, εijklm = Random error term.

Mean Temperature, Days to 50% Heading, and Plant Height
The days to 50% heading were significantly influenced by both genotype and location (Table 4). Plants grown in Marigat took a shorter time to attain 50% heading followed by those grown in Rongai and Egerton, respectively. Genotype E6518, EST 20, IS11721, and IS2331 took the longest time to attain 50% heading across the three locations while B35 took the shortest time (Table 4). Regression analysis showed an inverse relationship between the total sum of temperature and days to 50% flowering ( Figure 1).  Note. Values in a column, followed by different superscript letters are significantly different at p ≤ 0.05. CV = coefficient of variation.
Mean temperature in the Egerton, Rongai and Marigat were significantly different (Table 3) with Marigat showing the highest mean temperature followed by Rongai and Egerton, respectively. Sorghum plant height was significantly different among the sorghum genotypes and across locations in the year 2020.The tallest plants were observed in Marigat followed by Rongai and Egerton, respectively. Genotypes E6518, IS25547, IS25557, 1S11721 and IS2331 were tallest with their heights ranging from 192-258 cm while genotype B35 was the shortest at less than 100 cm across the three locations (Table 5). Note. Values in a column, followed by different superscript letters are significantly different at p ≤ 0.05. CV = coefficient of variation Note. Values in a column, followed by different superscript letters are significantly different at p ≤ 0.05. CV = coefficient of variation.

Lignin, Cellulose and Hemicellulose Content (Percent)
The effect of crop developmental stage, genotype, location, year and their interactions were highly significant for cellulose, hemicellulose and lignin except for Year*Stage (Table 6).  (Table 7B). Lignin, cellulose and hemicellulose generally increased during plant maturation from booting and dough stage (Tables 7A-7B, 8A-8B, 9A-9B). The highest lignin, cellulose and hemicellulose content in sorghum were observed in sorghum grown in Marigat followed by Rongai and lowest for sorghum at Egerton (Tables 7A-7B, 8A-8B, 9A-9B). Note. Values in a column, followed by different superscript letters are significantly different at p < 0.05. CV = coefficient of variation. Note. Values in a column, followed by different superscript letters are significantly different at p < 0.05. CV = coefficient of variation.  Vol. 15, No. 8; plants (Gev et al., 2020;Nabi et al., 2006). However, the results were contrary to those of Buttler and Muir (2003), and Miron et al. ( 2006) whose work indicates reduced lignocellulose content with advances in crop age.
Lignin, cellulose, and hemicellulose content also varied from one location to another due to differences in air temperature. The results were consistent with research reports on other crops (Papini-Terzi et al., 2009;Wacloworvsky et al., 2010;Xu et al., 2020;Zonetti et al., 2011). However, literature on the effect of increased air temperature on the lignocellulose content of sorghum fodder is limited as most research has mostly concentrated on its effect by low and freezing temperatures (Papini-Terzi et al., 2009;Waclawovsky et al., 2010;Xu et al., 2020).
Lignocellulose content increased from a cooler location (Egerton) to a warmer location (Rongai) and was highest in Marigat (a semi-arid location). Higher air temperature leads to an increase in atmospheric evaporative demand which could lead to water loss in plant tissue. A high evaporation rate has been shown to increase tissue thickening and enhanced lignification in sorghum plants (Waclawovsky et al., 2010). The observed increase in lignin, cellulose, and hemicellulose content with an increase in temperature among the test sorghum genotypes could be attributed to an increase in evapotranspiration in Rongai and Marigat as compared to the Egerton location. Changes in lignin, cellulose, and hemicellulose content could be linked to plant defense mechanisms against desiccation and/or increased heat load (Moura et al., 2010).

Conclusion
In conclusion, the stage of development, genotype, and air temperature influence lignin, cellulose, and hemicellulose concentration in fodder sorghum. Subsequently, feed value is expected to vary accordingly. Genotypes B35, EST 20, and EST 36 ranked low in lignin, cellulose, and hemicellulose content as they compared well with E6518 which is a commercial sorghum fodder genotype in Kenya. It was also evident that based on lignocellulose content, the best stage to harvest fodder sorghum for low lignin concentration is the booting stage.