2.2 Analyzing biomass stem composition

The composition of biomass derived from forage, grain, and sweet sorghums has been well characterized [28]. The research on exploiting forage sorghum as biofuel was initiated in 1980s, which led to the development of photoperiod-sensitiveenergy sorghum hybrids [29]. These are high biomass yielders [30]. Being relatively a recent introduction, the stem composition knowledge of energy sorghum is still limited. Up till now, a majority of research on sorghum biomass feedstock has focused more on improving yield than the quality components. So, there is a need to accurately conduct the biochemical analysis, since stem composition is the basic element influencing biofuel yield.

Plant cell walls are the main constituents of biomass that provide strength and limited plasticity to cell. The cell wall serves as a tough physical barrier, protecting interior of the cell against biotic and abiotic stresses. It is a multilayered structure composed of polysaccharides and proteins, which are important contributors of

biofuel quality and energy conversion processes. The polysaccharides are cellulose (a polymer of glucose), pectic compounds (polymers of galacturonic acid), and hemicellulose (a polymer of a variety of sugars including xylose, arabinose, and mannose). Cellulose is the largest source of glucose for biofuels. Glucuronoarabinoxylan (GAX) hemicellulose complex is linked to lignin. Since lignin component of plant cell wall provides structure, it cannot be converted to carbohydrates and hence is recalcitrant to conversion protocols. Likewise, ash content also reduces biomass to biofuel conversion reaction. Certain constituents of cell wall are water soluble like sugars, proteins, amino acids, mixed-linkage glucans, and phenolic glycosides, whereas chlorophyll, lipids, and waxes are water-insoluble ingredients that need ethanol extraction.

Different studies have reported various approaches for compositional analysis of energy sorghum leaves and stem. In some sorghum genotypes, proportion of cellulose can vary between 27 and 52%, while the range of hemicellulose content is 17–23% and lignin content is 6.2–8.1% [31, 32]. Along with the biomass yield, low lignin, high cellulose, and hemicellulose contents are also the desirable selection attributes for energy sorghum genotypes [33]. Such sorghums exhibit wide variations in biomass composition [34]. Now a days, near-infrared spectroscopic (NIR) analysis is routinely used for high-throughput computation of biomass composition [28].

Cellulosic bioethanol production requires three main steps: pretreatment, hydrolysis, and fermentation [35] (Figure 1). Pretreatment is performed to fractionate lignocellulose into different components via physical (boiling, steaming, and ultrasonication), chemical (acid, alkali, salts, etc.), physiochemical (ammonium fiber explosion or AFEX), and biological methods (bacteria and fungi). It increases porosity and surface area of the substrate. During hydrolysis, nonstructural carbohydrates are degraded in to sugars. Enzyme-based hydrolysis is preferred over acid hydrolysis being a mild and cost-effective process.

#### Figure 1.

Flow chart of sorghum cellulosic ethanol production process.

The process of fermentation proceeds under liquid or solid state in the presence of bacteria or yeast [36]. In a recent study, 24 sorghum genotypes (Table 1) were subjected to stem compositional analysis [37]. These genotypes had previously been selected on the basis of morphological traits [38].

The dried stem samples of these genotypes were grinded and used for measuring crude protein (%), ash contents (%), neutral detergent fiber (NDF %), acid detergent fiber (ADF %), hemicellulose (%), cellulose (%), and acid detergent lignin (ADL %), using the respective formulae:

$$\text{Crude protein } \%= \frac{0.1 \text{ N } \text{H}\_2\text{SO}\_4 \times 100 \times 6.25 \times (0.0014 \times \text{total diluted volume})}{\text{Weight of sample} \times \text{diluted digestibility material } (\text{ml})}$$

Ash% <sup>¼</sup> Weight of ash � <sup>100</sup>

Sr. # Genotype # Sr. # Genotype # 1. PI-609239-01-SD 13. PI-329875-03-SD 2. PI-620625-01-SD 14. PI-330039-02-SD 3. PI-648173-01-SD 15. PI-330022-01-SD 4. PI-648187-01-SD 16. PI-456415-03-SD 5. PI-454464-03-SD 17. PI-329488-02-SD 6. PI-570039-02-SD 18. PI-155871-02-SD 7. PI-525981-01-SD 19. PI-457393-02-SD 8. PI-329569-01-SD 20. PI-329480-02-SD 9. PI-583832-02-SD 21. PI-303658-02-SD 10. PI-329733-01-SD 22. PI-303656-01-SD 11. PI-456441-03-SD 23. NSL-54978 12. PI-329471-02-SD 24. PI-257595-01-SD

NDF% <sup>¼</sup> <sup>ð</sup>Weight of crucible residueÞ � Weight of crucible � <sup>100</sup>

ADF% <sup>¼</sup> <sup>ð</sup>Weight of crucible <sup>þ</sup> ADF residueÞ � Weight of crucible � <sup>100</sup>

Hemicellulose % <sup>¼</sup> <sup>ð</sup>NDF � ADFÞ � Weight of crucible � <sup>100</sup>

ðWeight of crucible þ ADF residueÞ � Weight:of crucible þ residue after 24 NH2SO4 � 100 Weight of sample

ðWeight of crucible þ residue of celluloseÞ � Weight of crucible þ residue after combustion � 100 Weight of sample

Statistical analysis indicated highly significant variations among all sorghum genotypes for crude protein, ash contents, NDF, ADF, ADL, hemicellulose, and

PCA analysis of different biochemical traits indicated three principle components (PC1, PC2, and PC3) having Eigen values greater than 1 (Table 3). The cumulative variability of three PCs was 82.94% for the studied genotypes. The total variability in traits shared by three PCs was 37.48, 27.37, and 18.096%, respectively. Different biomass-related traits added more than 34% of variation factor in PC1 such as: ash contents (43.7%), ADL (47.6%), cellulose (45.5%), hemicelluloses (37.4%), and NDF (48.5%). PC1 showed weak and positive correlation with crude protein (0.000%) and ADF (0.012%). The PC2 contributed for 27.37% of total variability. PC2 showed positive and strong correlation with the traits such as ADL (38.4%), ADF (50.1%), and cellulose (46.8%). Weak and negative correlation was

Cellulose% ¼

Table 1.

Sorghum genotypes used for stem compositional analysis.

Sorghum an Important Annual Feedstock for Bioenergy DOI: http://dx.doi.org/10.5772/intechopen.86086

Lignin=ADL %ð Þ¼

165

cellulose contents (Table 2).

Weight of sample (2)

Weight of sample (3)

Weight of sample (4)

Weight of sample (5)

(6)

(7)

#### Sorghum an Important Annual Feedstock for Bioenergy DOI: http://dx.doi.org/10.5772/intechopen.86086


#### Table 1.

biofuel quality and energy conversion processes. The polysaccharides are cellulose (a polymer of glucose), pectic compounds (polymers of galacturonic acid), and hemicellulose (a polymer of a variety of sugars including xylose, arabinose, and mannose). Cellulose is the largest source of glucose for biofuels. Glucuronoarabinoxylan (GAX) hemicellulose complex is linked to lignin. Since lignin component of plant cell wall provides structure, it cannot be converted to carbohydrates and hence is recalcitrant to conversion protocols. Likewise, ash content also reduces biomass to biofuel conversion reaction. Certain constituents of cell wall are water soluble like sugars, proteins, amino acids, mixed-linkage glucans, and phenolic glycosides, whereas chlorophyll, lipids, and waxes are water-insoluble ingredients

Biomass for Bioenergy - Recent Trends and Future Challenges

Different studies have reported various approaches for compositional analysis of energy sorghum leaves and stem. In some sorghum genotypes, proportion of cellulose can vary between 27 and 52%, while the range of hemicellulose content is 17–23% and lignin content is 6.2–8.1% [31, 32]. Along with the biomass yield, low lignin, high cellulose, and hemicellulose contents are also the desirable selection attributes for energy sorghum genotypes [33]. Such sorghums exhibit wide variations in biomass composition [34]. Now a days, near-infrared spectroscopic (NIR) analysis is routinely used for high-throughput computation of biomass

Cellulosic bioethanol production requires three main steps: pretreatment, hydrolysis, and fermentation [35] (Figure 1). Pretreatment is performed to fractionate lignocellulose into different components via physical (boiling, steaming, and ultrasonication), chemical (acid, alkali, salts, etc.), physiochemical (ammonium fiber explosion or AFEX), and biological methods (bacteria and fungi). It increases porosity and surface area of the substrate. During hydrolysis, nonstructural carbohydrates are degraded in to sugars. Enzyme-based hydrolysis is preferred over acid

The process of fermentation proceeds under liquid or solid state in the presence of bacteria or yeast [36]. In a recent study, 24 sorghum genotypes (Table 1) were subjected to stem compositional analysis [37]. These genotypes had previously been

The dried stem samples of these genotypes were grinded and used for measuring crude protein (%), ash contents (%), neutral detergent fiber (NDF %), acid detergent fiber (ADF %), hemicellulose (%), cellulose (%), and acid detergent lignin

Crude protein % <sup>¼</sup> <sup>0</sup>:1NH2SO4 � <sup>100</sup> � <sup>6</sup>:<sup>25</sup> � ð Þ <sup>0</sup>:<sup>0014</sup> � total diluted volume

Weight of sample � diluted digested material ml ð Þ

(1)

hydrolysis being a mild and cost-effective process.

selected on the basis of morphological traits [38].

Flow chart of sorghum cellulosic ethanol production process.

(ADL %), using the respective formulae:

that need ethanol extraction.

composition [28].

Figure 1.

164

Sorghum genotypes used for stem compositional analysis.

$$\text{Ash\%} = \frac{\text{Weight of Ash} \times 100}{\text{Weight of sample}} \tag{2}$$

$$\text{NDF\%} = \frac{(\text{Weight of cruricible reside}) - \text{Weight of cruricible} \times 100}{\text{Weight of sample}} \tag{3}$$

$$\text{ADF\%} = \frac{(\text{Weight of cruricible} + \text{ADF residue}) - \text{Weight of cruricible} \times 100}{\text{Weight of sample}} \tag{4}$$

$$\text{HemiCellolose } \% = \frac{(\text{NDF} - \text{ADF}) - \text{Weight of cruricible} \times 100}{\text{Weight of sample}} \tag{5}$$

$$\text{Cellolose\%} = \tag{6}$$

$$(\text{Weight of cruricible} + \text{ADF residue}) - \text{Weight of cruricible} + \text{resistance after 24 NH}\_2\text{SO}\_4 \times 100$$

$$\text{Weight of sample} \tag{6}$$


(7)

Statistical analysis indicated highly significant variations among all sorghum genotypes for crude protein, ash contents, NDF, ADF, ADL, hemicellulose, and cellulose contents (Table 2).

PCA analysis of different biochemical traits indicated three principle components (PC1, PC2, and PC3) having Eigen values greater than 1 (Table 3). The cumulative variability of three PCs was 82.94% for the studied genotypes. The total variability in traits shared by three PCs was 37.48, 27.37, and 18.096%, respectively. Different biomass-related traits added more than 34% of variation factor in PC1 such as: ash contents (43.7%), ADL (47.6%), cellulose (45.5%), hemicelluloses (37.4%), and NDF (48.5%). PC1 showed weak and positive correlation with crude protein (0.000%) and ADF (0.012%). The PC2 contributed for 27.37% of total variability. PC2 showed positive and strong correlation with the traits such as ADL (38.4%), ADF (50.1%), and cellulose (46.8%). Weak and negative correlation was

### Biomass for Bioenergy - Recent Trends and Future Challenges


PC, principle component; SD, standard deviation; CV, coefficient of variation; AC, ash contents; ADL, acid detergent lignin; ADF, acid detergent fiber; C, cellulose; CP, crude protein; HC, hemicellulose; NDF, neutral detergent fiber.

#### Table 2.

Principle component analysis (PCA) related to biomass traits in sorghum.


SD, standard deviation; CV, coefficient of variation; AC, ash contents; ADL, acid detergent lignin; ADF, acid detergent fiber; C, cellulose; CP, crude protein; HC, hemicellulose; NDF, neutral detergent fiber.

#### Table 3.

Descriptive statistics for quantitative traits of sorghum germplasm.

observed for ash contents (21.8%), crude protein (18.8%), hemicellulose (48.1%), and NDF (26.1%). Crude protein and ash contents showed 77.7 and 37.3% of the factor variations in PC3, respectively.

analysis, NDF, ADL, and ADF are generally used as standard quality testing techniques [39], while lignin concentration markedly affects the efficiency

Traits CP AC NDF ADF ADL HC C CP 1 0.347\* 0.128 0.066 0.182 0.051 0.080 AC 0.347 1 0.473\* 0.362\* 0.173 0.431\* 0.311\* NDF 0.128 0.473 1 0.293\* 0.033 0.802\*\* 0.289\* ADF 0.066 0.362 0.293 1 0.153 0.067 0.955\*\* ADL 0.182 0.173 0.033 0.153 1 0.313 0.335\* HC 0.051 0.431 0.802 0.067 0.313 1 0.016 C 0.080 0.311 0.289 0.955 0.335 0.016 1

PCA grouping of 24 USDA sorghum genotypes using quantitative traits.

Sorghum an Important Annual Feedstock for Bioenergy DOI: http://dx.doi.org/10.5772/intechopen.86086

Correlation coefficients of various traits of sorghum genotypes.

Study reports that by increasing the level of lignin, cellulose and hemicellulose concentrations decreased. The genetic relationships among 24 genotypes were identified through construction of dendrogram on the basis of similarity matrix utilizing the UPGMA algorithm (Figure 3). The genotypes were grouped into two main clusters: only two genotypes (PI-583832-02-SD and PI-456415-03-SD) were present in subcluster-1, while the subcluster-2 was divided into smaller groups. The genotypes PI-570039-02-SD, PI-330022-01-SD, and NSL-54978 were grouped together and showed some distinctness from rest of the members of the group, whereas the maximum genetic relatedness was found among genotypes

PI-329569-01-SD and PI-303658-02-SD followed by genotypes PI-329733-01-SD, PI-525981-01-SD, PI-303656-01-SD, and PI-648187-01-SD. The genotypes

of hydrolysis [40].

Normal correlation. \*\*Strong correlation.

Figure 2.

\*

Table 4.

167

Biplot analysis described that variables were greatly obliged as vectors; comparative length of the vector was distinguished as the relative proportion of the variability in each variable. The traits like ADL and CP, which were plotted near the central point, showed more similarities, while cellulose, ADF, NDF, and HC displayed more variability (Figure 2). Significant characters such as ADL, ADF, and cellulose were located at positive and positive coordinate region in biplot. Traits like AC, NDF, and HC were allocated at negative coordinate (Figure 2). Variability in the traits explains the variations among genotypes, which can be used in sorghum breeding plan effectively. Correlation analysis among biofuel-related stem compositional traits indicated that concentration of protein and lignin contents showed negative interaction with cellulose and hemicelluloses (Table 4). It showed that significant genetic variability is present among 24 sorghum genotypes. In sorghum, cellulose and hemicellulose contents play significant role in biofuel quality. For fiber Sorghum an Important Annual Feedstock for Bioenergy DOI: http://dx.doi.org/10.5772/intechopen.86086

Figure 2. PCA grouping of 24 USDA sorghum genotypes using quantitative traits.


\*\*Strong correlation.

#### Table 4.

observed for ash contents (21.8%), crude protein (18.8%), hemicellulose (48.1%), and NDF (26.1%). Crude protein and ash contents showed 77.7 and 37.3% of the

detergent fiber; C, cellulose; CP, crude protein; HC, hemicellulose; NDF, neutral detergent fiber.

Eigen vectors PC1 PC2 PC3 AC 0.437 0.218 0.373 ADL 0.476 0.384 0.134 ADF 0.012 0.501 0.191 C 0.455 0.468 0.097 CP 0.000 0.188 0.777 HC 0.374 0.481 0.293 NDF 0.485 0.261 0.328 Eigen value 2.623 1.916 1.267 Variability % 37.476 27.371 18.096 Cumulative % 37.476 64.847 82.943

PC, principle component; SD, standard deviation; CV, coefficient of variation; AC, ash contents; ADL, acid detergent lignin; ADF, acid detergent fiber; C, cellulose; CP, crude protein; HC, hemicellulose; NDF, neutral detergent fiber.

Variables Minimum Maximum Mean SD CV (%) CP 4.927 10.927 7.808 1.414 1.37 AC 5.217 19.470 12.418 3.877 2.39 NDF 54.633 81.500 63.947 6.411 2.29 ADF 26.167 54.500 34.410 6.994 4.34 ADL 1.500 8.000 3.160 1.316 14.17 HC 22.087 44.150 31.419 5.981 1.64 C 29.000 57.167 39.250 7.331 2.03 SD, standard deviation; CV, coefficient of variation; AC, ash contents; ADL, acid detergent lignin; ADF, acid

Principle component analysis (PCA) related to biomass traits in sorghum.

Biomass for Bioenergy - Recent Trends and Future Challenges

Biplot analysis described that variables were greatly obliged as vectors; comparative length of the vector was distinguished as the relative proportion of the variability in each variable. The traits like ADL and CP, which were plotted near the central point, showed more similarities, while cellulose, ADF, NDF, and HC displayed more variability (Figure 2). Significant characters such as ADL, ADF, and cellulose were located at positive and positive coordinate region in biplot. Traits like AC, NDF, and HC were allocated at negative coordinate (Figure 2). Variability in the traits explains the variations among genotypes, which can be used in sorghum breeding plan effectively. Correlation analysis among biofuel-related stem compositional traits indicated that concentration of protein and lignin contents showed negative interaction with cellulose and hemicelluloses (Table 4). It showed that significant genetic variability is present among 24 sorghum genotypes. In sorghum, cellulose and hemicellulose contents play significant role in biofuel quality. For fiber

factor variations in PC3, respectively.

Descriptive statistics for quantitative traits of sorghum germplasm.

Table 2.

Table 3.

166

Correlation coefficients of various traits of sorghum genotypes.

analysis, NDF, ADL, and ADF are generally used as standard quality testing techniques [39], while lignin concentration markedly affects the efficiency of hydrolysis [40].

Study reports that by increasing the level of lignin, cellulose and hemicellulose concentrations decreased. The genetic relationships among 24 genotypes were identified through construction of dendrogram on the basis of similarity matrix utilizing the UPGMA algorithm (Figure 3). The genotypes were grouped into two main clusters: only two genotypes (PI-583832-02-SD and PI-456415-03-SD) were present in subcluster-1, while the subcluster-2 was divided into smaller groups. The genotypes PI-570039-02-SD, PI-330022-01-SD, and NSL-54978 were grouped together and showed some distinctness from rest of the members of the group, whereas the maximum genetic relatedness was found among genotypes PI-329569-01-SD and PI-303658-02-SD followed by genotypes PI-329733-01-SD, PI-525981-01-SD, PI-303656-01-SD, and PI-648187-01-SD. The genotypes

2.3 Transcriptional and translational analyses of sorghum biomass

Sorghum an Important Annual Feedstock for Bioenergy DOI: http://dx.doi.org/10.5772/intechopen.86086

by applying various biotechnological approaches such as proteomics,

and expression divergence between different sorghum varieties.

gene expression at different stages of sorghum plant development [45].

in sorghum. Four short stature genotypes were chosen as negative control

1. NSL-54978 Tall

5. PI-457393-02-SD Medium

9. PI-648187-01-SD Small

13. PI-643630-01-SD Negative control

2. PI-456441-03-SD 3. PI-525981-01-SD 4. PI-303656-01-SD

6. PI-583832-02-SD 7. PI-620625-01-SD 8. PI-456415-03-SD

10. PI-609239-01-SD 11. PI-330039-02-SD 12. PI-329733-01-SD

14. PI-643735-03-SD 15. PI-643581-01-SD 16. PI-642993-01-SD

Sorghum genotypes and their respective groups based on height.

Sr. # Genotypes Height-based groups

The in vitro-germinated, 15-day-old sorghum seedlings were used for protein extraction. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) revealed diverse banding pattern of proteins ranging in size from 14.9 to 124 kDa

with different expression levels in all studied genotypes (Table 6).

(Table 5).

Table 5.

169

The mysterious relationship between phenotype and genotype can be revealed

RNA-seq technology for expression profiling has been applied in sorghum to study different gene functions [44]. This technique gives a precise assessment of

Proteomics offers the set of the most efficient tools for recognition, assessment, and quantification of unique proteins. Our recent study [44] merged transcriptomic and proteomic approaches for screening sorghum germplasm best suited for bioenergy and for comparative analysis of protein expression of elite sorghum germplasm. The study was based on 24 USDA sorghum genotypes selected for biomass potential in the field experiments, which is already reported in this chapter [37]. For translational analysis, 12 out of 24 selected genotypes were divided into three groups based on stem height, since height is directly correlated with biomass

transcriptomics, and metabolomics [42]. In transcriptomics, a huge set of gene libraries can be established by employing different techniques of bioinformatics and next-generation sequencing [43]. Over the last decade, expression profiling experiments for genome-wide investigation in sorghum have been carried out to analyze responses to numerous abiotic and biotic stresses, to determine tissue-specific and genotype-specific gene expression motifs, and to disclose the genetic modification

#### Figure 4.

Cladogenesis studies using homology-based classification of 24 sorghum genotypes.

PI-583832-02-SD and PI-329733-01-SD were also found genetically distinct from rest of the genotypes used in the study (Figure 4). Variance decomposition for optimal classification showed that there were 23.41 and 76.59% variances present within and between classes, respectively.

The sorghum germplasm with less lignin and protein contents is desirable for biofuel production. Sorghum genotype PI-609239-01-SD had maximum value of NDF (83.5%) and ash contents (19.5%), while genotype PI-303658-02-SD exhibited the maximum value (57.5%) of cellulose content.

Though sorghum is viewed as a cheap source of biofuel being able to grow on marginal lands, few studies have indicated the lower biofuel potential of energy sorghums grown on marginal lands than the crop land [41]. Hence, screening of energy sorghum having stress tolerance, with efficient production technology and conservation tillage practices, is the key element of sustainable commercial production of energy sorghum [5].
