Natural Variation of Lignocellulosic Components in Miscanthus Biomass in China

Lignocellulose content is an important factor affecting the conversion efficiency of biomass energy plants. In this study, 179 Miscanthus accessions in China were used to determine the content of lignocellulose components in stems via acid hydrolysis and high-performance liquid chromatography. Results showed that the average lignocellulose content of wild Miscanthus germplasm resources was 80.27 ± 6.51%, and the average content of cellulose, hemicellulose, lignin, extracts, and total ash was 38.38 ± 3.52, 24.23 ± 4.21, 17.66 ± 1.56, 14.50 ± 5.60, and 2.53 ± 0.59%, respectively. The average lignocellulose content of M. sinensis, M. floridulus, M. nudipes, M. sacchariflorus, M. lutarioriparius, and the hybrids was 77.94 ± 6.06, 75.16 ± 4.98, 75.68 ± 3.02, 83.71 ± 4.78, 81.50 ± 5.23, and 74.72 ± 7.13%, respectively. In all the tested materials, the highest cellulose content was 48.52%, and the lowest was 29.79%. Hemicellulose had the maximum content of 34.23% and a minimum content of 15.71%. The highest lignin content was 23.75%, and the lowest was 13.01%. The lignocellulosic components of different ploidy materials were compared. The content of lignocellulosic components of diploid M. sacchariflorus was higher than that of tetraploid M. sacchariflorus, and the content of lignocellulosic components of diploid M. lutarioriparius was lower than that of tetraploid M. lutarioriparius. Analysis of the relationship between the changes in lignocellulosic components and geographical locations of Miscanthus showed that the holocellulose and hemicellulose content was significantly positive correlated with the latitude of the original growth location. Results indicated that the lignocellulosic components of Miscanthus resources in China are rich in genetic diversity.


INTRODUCTION
Miscanthus is a tall perennial herbaceous plant. It belongs to the subtribe Saccharinae, tribe Andropogoneae, subfamily Panicoideae, and family Poaceae. It originated in East Asia and Southeast Asia and is now widely distributed in China, Japan, and Pacific Islands (Hodkinson et al., 2002;Hastings et al., 2009;Jensen et al., 2011) China is an important origin and distribution center of Miscanthus (Clifton-Brown et al., 2001;Clifton et al., 2015;Li et al., 2019), with extensive wild germplasm resources and abundant genetic diversity (Hodkinson et al., 2002;Anzoua et al., 2011;Ge et al., 2019). Seven (Ge et al., 2017), and natural hybrids exist in nature (Lewandowski et al., 2003;Cichorz et al., 2015).
Miscanthus is a lignocellulosic crop with highly efficient C4 photosynthesis, high biomass production, strong stress resistance, and wide adaptability. Miscanthus has high water and fertilizer efficiency, excellent cellulose quality, extensive cultivation, eco-friendly environment, and low production costs (Beale and Long, 1995;Clifton-Brown et al., 2001;Lewandowski and Schmidt, 2005;Clifton et al., 2015). The biological yield of Miscanthus (3 × 104 kg/ha) is about three times higher than that of switchgrass . Compared with other grasses, the lignocellulose of Miscanthus is closer to that of wooden materials (Villaverde et al., 2010;Lygin et al., 2011), where the contents of cellulose, hemicellulose, and lignin are ∼30-50, 10-40, and 5-30%, respectively (McKendry, 2002;Yang et al., 2007;Kleinert and Barth, 2008). Miscanthus is a grassy lignocellulosic material used for converting heat, electricity, and liquid fuels (Cherubini, 2010), as well as for producing aromatic products (Pauly and Keegstra, 2008;Luo et al., 2016;Upton and Kasko, 2016). Miscanthus has a higher energy ratio than natural gas and coal, thus it has lower greenhouse gas emissions (Moukamnerd et al., 2010;McCalmont et al., 2017). Compared with high starch or high sugar crops such as sweet sorghum, Miscanthus has low moisture and sugar content during harvesting, making it more convenient to store and transport. In addition, unlike cereal crops, harvesting Miscanthus as a biofuel does not directly increase the price of cereals (Ziolkowska, 2014). These characteristics make it stand out among many energy crops and have made it one of the most promising non-grain energy plants. Hence, research on Miscanthus has sparked wide interest (Feltus and Vandenbrink, 2012;Cao et al., 2019).
With the increase in energy demand, the conversion of lignocellulosic biomass to fuels, such as ethanol, has been the focus of research in many countries. Plants can convert light energy into monosaccharides through photosynthesis and then use CO 2 to fix monosaccharides into high-energy polymers and generate composite cell walls composed of cellulose, hemicellulose, and lignin (Rubin, 2008). The important factors that make saccharifying lignocellulose raw materials difficult are the degree of polymerization and lignification and cellulose crystallinity (Abramson et al., 2010). Therefore, the main obstacle hindering the accurate determination of lignocellulose content is how to effectively decompose cell walls into fermentable sugars. Five methods are commonly used for determining lignocellulose, including washing cellulose analysis methods, which can measure neutral detergent fiber, acid detergent fiber, and acid detergent lignin. The Klason method is a classic technique for determining lignocellulose, but it overestimates the true lignin value of raw materials. The application of this method is limited because it cannot determine soluble fiber and remove farinaceous substance (Hatfield et al., 1994;Wang et al., 2020). New and state-of-the art technologies, such as near-infrared spectroscopy and nuclear magnetic resonance, have been widely used in determining lignocellulose. Near-infrared spectroscopy can determine the concentration of various plant components, such as fat, grease, protein, and total fiber. However, the detection result of this method is not sufficiently accurate because the spectral measurement value has no direct relation to lignin concentration, and the influence of comparison parameter on the measurement value is high (Li et al., 2015;Ramirez et al., 2015;Hayes et al., 2017;Jin et al., 2017;Elle et al., 2019). Nuclear magnetic resonance spectroscopy is an analytical technique for detecting the composition and structural characteristics of lignin. Considering its inability to obtain good and clear spectra from complex plant samples, this method is currently not widely used (Capanema et al., 2004;Balakshin et al., 2011). The National Renewable Energy Laboratory (NREL) of the United States proposed the NREL method (Sluiter et al., 2008(Sluiter et al., , 2012. Samples are hydrolyzed with sulfuric acid after extracting the extract of the sample to be tested. Glucose content is measured by highperformance liquid chromatography (HPLC). Cellulose content is quantified using the substitution ratio of glucose and cellulose. Lignin content is determined using the differential weight of the residue after hydrolysis of the sample. This method is not only experimentally operable but also provides accurate detection results.
Major breakthroughs in terms of unit biomass production and optimization of biomass conversion efficiency are needed to make the products of second-generation lignocellulosic energy crops economically competitive (Sims et al., 2010;Feltus and Vandenbrink, 2012). The biomass composition of energy plants affects the conversion efficiency. In addition, using Miscanthus as a feedstock for bioenergy requires that the biomass composition is adapted to various bioenergy conversion processes (Arnoult and Brancourt-Hulmel, 2015). The development of breeding programs also requires a clear understanding of the content of biomass composition. In order to investigated the biomass composition of Miscanthus, the components of lignocellulose in different ecological types of wild resources were determined by using the NREL method. Our research shows that the content of Miscanthus lignocellulose is affected by both genetic factors and environmental factors. The results are of great importance for the development and utilization of Miscanthus resources in China, genetic breeding of superior energy plants, and the conversion and utilization of biomass energy.

Methods
The experimental method followed the NREL method for determining lignocellulose (Thygesen et al., 2005;Sluiter et al., 2012;Kuchelmeister and Bauer, 2015). This method was modified and improved.

Sample Pretreatment
Stems were dried to constant weight, crushed, and passed through a 40-mesh sieve. The ground sample (m1 = 0.6 ± 0.010 g) was weighed, reflowed in a Soxhlet extractor (a traditional glass apparatus, Shandong, Hualu) containing water for 8 h, and then dried in a drying oven at 40 • C. Then, the sample was refluxed in a Soxhlet extractor containing absolute ethyl alcohol for 16 h and dried in a drying (DHG-9140A, Shanghai) oven at 40 • C. After extraction, the remaining solid material (m2) was mainly lignocellulose; the part lost during the process was the extract, and its content was calculated using the equation:

Acidolysis of Samples
The extracted sample (m0 = 0.3000 g) was weighed and placed in a pressure-resistant tube (89063-334, VWR). Exactly 3.00 mL 72% H 2 SO 4 was added, and the mixture was thoroughly stirred and mixed. Then, the sample was placed in a water bath (2321, Fisher Scientific) at 30 • C for 60 min. Thereafter, 84.00 mL ddH 2 O was added, and the sample was sterilized in an autoclave (GI80TR, ZEALWAY) (121 • C, 1 h).
After acidification of the sample, cellulose was degraded to glucose, whereas hemicellulose was degraded to xylan, arabinose, galactose, and mannose. Lignin was divided into acid-insoluble lignin (AIL) and acid-soluble lignin (ASL). The residue was used for the determination of AIL, whereas the filtrate was used for the determination of ASL and monosaccharides.

Determination of Lignin Content
Use a filter crucible (89038-050, VWR) with 15 µm Pore Diameter to filter the hydrolyzed sample to collect the filtrate and residue. AIL was determined via the ashing method. The residue was dried to a constant weight (m3) and then placed in a boxtype electrical resistance furnace (SX2-G/T, Shanghai Yuejin). The sample was turned to ash at 575 ± 25 • C for 10 h, cooled to room temperature, and weighed (m4). The percent AIL content was determined using the equation : ASL was determined using an UV-Vis spectrophotometer (Nanodrop2000c, Thermo). The absorbance of the filtrate was determined at λ = 205 nm. The percent ASL content was calculated using the equation: Where ε represents the absorption value, D represents dilution factor, V is the total liquid volume (87 mL), and K = 110 represents the absorption coefficient of acid-soluble lignin (Hayes, 2012). Lignin content was calculated using the equation:

Determination of Cellulose and Hemicellulose
Monosaccharide content was determined via HPLC [Chromatographic conditions: chromatographic column (Biorad Aminex HPX-87P), Deashing packed column, Detector (evaporative light scattering detector), Injection volume (35 µL), mobile phase (Ultrapure water), flow velocity (0.6 mL/min), Nitrogen pressure (30 psi); drift tube (heating mode, 80 ± 25 • C), Sprayer (60%); running time (20min)]. Exactly 4 mL of the filtrate was obtained, and the pH was adjusted to 5-6 with CaCO 3 . The supernatant was collected by centrifugation and filtered through a 0.22 µm filter membrane. Then, HPLC was used to determine the content of monosaccharides. Both monosaccharides and calcium carbonate are pure reagents (Sigma) for chromatographic analysis. The cellulose and hemicellulose contents were calculated from the monosaccharide content as follows: % Holocellulose = %Cellulose + %Hemicellulose (7) where Ac is the dehydration correction coefficient. The Ac values of pentose and hexose were 0.88 and 0.90, respectively. %Glu, %Xyl, %Ara, %Gal, and %Man represent the contents of the corresponding monosaccharides obtained by the regression curve method.

Determination of Total Ash
Weigh the mass of the empty crucible, record it as m5. Then weigh about 0.5 g of the sample, put it in a filter crucible and weigh it (m6), then put it into a box-type electric furnace, 575 ± 25 • C, 24 h, cool it to room temperature in a desiccator and weigh m7.

Data Processing
The determination results of each Miscanthus sample were expressed as the average of three replicates. Data statistics were completed and coefficient of variation was calculated using Excel. Maps, mapdata, and ggplot2 software packages in R (3.6.0) were used to draw the distribution of material sources. SPSS software (statistics 24.0) was used to perform single-factor ANOVA test and obtain the boxplot of component content. Pearson correlation analysis was used to determine the material geographic location and lignocellulosic component content.  (Figure 1).
Owing to ploidy changes, the cellulose and hemicellulose content of 4X M. sacchariflorus tended to diversify, whereas and its lignin content was more stable. The cellulose and lignin content of 4X M. lutarioriparius tended to diversify, whereas its hemicellulose content remained stable.

Comparative Analysis of Miscanthus Materials From Different Ecological Regions
The correlation between the differences in geographical factors (longitude, latitude, and altitude) and lignocellulosic components of M. sacchariflorus were analyzed. Results showed that the hemicellulose and holocellulose content increased with the increase in latitude, and a very significant positive correlation with latitude was observed. The correlation coefficients were 0.409 and 0.441 for hemicellulose and holocellulose, respectively. A significant positive correlation was found between total ash content and longitude and latitude, with correlation coefficients of 0.342 and 0.361, respectively. The extract content had a significant negative correlation with longitude and latitude, with coefficients of −0.346 and −0.523, respectively. In addition, no correlation was found between each lignocellulosic component of Miscanthus and altitude (Table 4). Further linear analysis was performed. Results showed that the difference in latitude would cause changes in hemicellulose and holocellulose content (the sum of hemicellulose and cellulose). The values of R 2 were 0.7924 and 0.7654, respectively, indicating that the linear fitting equation was credible, and the high credibility reflects the correlation between total cellulose and hemicellulose content and latitude (Figure 3). This result was consistent with that of Pearson correlation analysis.

Effects of Geographical Factors on Lignocellulose Content
China We found that related geographical factors, such as latitude, had an important selection effect on Miscanthus species, and the holocellulose and hemicellulose content increased with latitude. This result was consistent with that obtained by Zhao et al. (2014). Youngmi Kim also found similar patterns on the lignocellulose content of switchgrass (Kim et al., 2011). Low temperature increases the content of soluble sugar in plants to protect the stability of cell membranes (Jan et al., 2009;Pompeiano et al., 2015). Papini-Terzi found a correlation between some genes associated with cell wall metabolism and sugar content in plants. Plants under cold weather conditions increase their soluble sugar content, thereby indirectly inducing the synthesis of related lignocelluloses (Papini-Terzi et al., 2009;Vicentini et al., 2009;Waclawovsky et al., 2010). The holocellulose content of Miscanthus in high latitudes is relatively high. A statistical field phenotypic survey revealed significant differences in the flowering and maturity stages of Miscanthus plants transplanted to the Shandong experimental base due to the influence of photoperiod (Imaizumi and Kay, 2006). Light time increases with the increase in latitude, resulting in short flowering time. Analysis of M. sacchariflorus showed that in the Yangtze River Basin and south of the region, the flowering period is mainly concentrated in mid-to-late September and early October, but the flowering period in Shandong and Beijing mainly occurs in mid-to-late August and early September. Blooming in Heilongjiang, Jilin, and Liaoning mostly occurs in mid-to-late June. Hence, geographical location has an obvious selective effect on the genetic variation in Miscanthus, and this variation is closely related to phenological conditions corresponding to geographical location, such as photoperiod, accumulated temperature, and rainfall. Considering the vastness of China's geographical locations, the species of Miscanthus in China have rich diversity.

Comparative Analysis of the Determination Results of Lignocellulosic Components of Miscanthus
Jung compared the lignocellulose content of Miscanthus, switchgrass, sorghum, and reeds and found that the lignin content of Miscanthus was significantly lower than that of reeds. Therefore, Miscanthus is a more suitable energy plant than reeds. Moreover, M. sacchariflorus contains 14.12% lignin and 64.23% holocellulose (Heaton, 2008 47.52, 19.49 to 34.23, and 14.75 to 23.75%, respectively. The maximum cellulose, hemicellulose, and lignin content was 1.5 times or higher than it, indicating that Miscanthus plants have abundant diversity within each species. Obvious differences were observed among various species of Miscanthus. For example, the content of hemicellulose and other components of M. sacchariflorus was considerably different from that of M. sinensis and M. floridulus according to single-factor ANOVA analysis. Significant differences were also observed in cellulose and lignin content among species. However, the contents of major lignocellulosic components, such as cellulose, hemicellulose, and lignin, between M. sinensis and M. floridulus and between M. sacchariflorus and M. lutarioriparius were slightly different. This result arose because the evolutionary relationship between M. sinensis and M. floridulus is close, as well as the genetic evolution between M. sacchariflorus and M. lutarioriparius. Ge described this evolutionary relationship in detail (Ge et al., 2017).
The coefficient of variation was calculated to determine the difference and potential genetic diversity of the various lignocellulosic components of Miscanthus. Generally, the coefficient of variation of each species of Miscanthus was high, indicating that its lignocellulosic components in China have rich diversity. The coefficient of variation of the content of each component of the hybrids was high in all species, proving that selecting varieties with high cellulose and hemicellulose contents and low lignin content from hybrids is easier than from wild types. The coefficient of variation of each component (except total ash) of M. nudipes was relatively low among all species, and this observation may explain the relatively concentrated geographical distribution of this species.

Significance and Application of the Determination of Lignocellulose Content of Miscanthus
In a 3-year field trial in Illinois, USA, Jung found that Miscanthus × giganteus has a biotransformation efficiency about 2.7 times higher than that of corn (Jung et al., 2015). Therefore, Miscanthus is widely studied as a second-generation biomass energy source. Lignocellulosic biomass is mainly a complex structure composed of cellulose, hemicellulose, lignin, and some extractable components. The amount, proportion, and type of each ingredient largely depend on the type of raw material (Pauly and Keegstra, 2008;Zhang et al., 2012;Yu et al., 2018). Cellulose and hemicellulose belong to polysaccharides. Cellulose and hemicellulose are used for the conversion of biomass energy, and their content determines the efficiency of fuel conversion (Bosch and Hazen, 2013). Lignin is an amorphous high-molecular organic polymer with a three-dimensional network structure composed of carbon-oxygen and carbon-carbon bonds. It crosslinks cellulose and hemicellulose to provide good support for the stem. The complex chemical structure of lignin hinders the degradation of cellulose and hemicellulose and makes the conversion and use of biomass energy difficult (Boudet et al., 2003;Sticklen, 2006Sticklen, , 2008Chang, 2007;Chen and Dixon, 2007;Li et al., 2008). Therefore, choosing varieties with high cellulose and hemicellulose content and low lignin content is beneficial to improve energy conversion efficiency. In view of this analysis, we proposed the H/L index, which can reflect the difficulty in converting cellulose energy plants into energy substances to a certain extent. Miscanthus with a high H/L value is suitable for conversion to alcohols by fermentation. The H/L value of Miscanthus did not significantly differ among species, but obvious differences were observed among varieties. The H/L index also has an important reference value in breed selection.
M. giganteus is the most studied species in terms of production applications. M. giganteus has high cellulose content and strong adaptability. Hence, it is widely cultivated in European countries. We obtained 23 hybrids (M. sinensis × M. sacchaflorus, M. floridulus × M. sacchariflorus, and M. floridulus × M. lutarioriparius) through artificial crosses. The yield of hybrids has obvious advantages compared with other Miscanthus species. Moreover, the lignin content of the hybrids was higher than that of M. sinensis and M. floridulus. Therefore, selecting varieties with high total cellulose content and low lignin content from hybrids is easier than from wild types. Hence, artificial hybrid breeding is an effective way of selecting excellent energy plants. According to the determination results of lignocellulose combined with the growth adaptability characteristics of Miscanthus, and by making full use of the abundant resources of Miscanthus in China, selective artificial breeding was performed to select energy plants with optimized content of stem components and broad growth adaptability.

CONCLUSION
From the overall results, Miscanthus is a good bioenergy plant with high lignocellulose content. At present, the biological production of various platform chemicals has been realized, such as ethanol, butanol, lactic acid, levulinic acid, sorbitol, glycerol, 1,3-propanediol, itaconic acid, succinic acid, and 2,5-FDCA. Therefore, suitable accessions of Miscanthus can be selected based on their biomass composition and different transformation processes. Further analyses found that Miscanthus there is obvious differences among intra-and interspecies of Miscanthus, and the content of lignocellulosic components has a wide range of values, in consistent with rich genetic diversity. More interestingly, our research shows that the content of Miscanthus lignocellulose is not only affected by genetic factors (such as ploidy), but also by environmental factors, such as Miscanthus in high latitudes with higher hemicellulose content. Overall, this research laid a solid foundation for the efficient development of Miscanthus biomass conversion, genetic breeding, and utilization in the future.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
PX and SC analysis of experimental data and article writing. HL and DZ collection of experimental materials. YH and YW designed the experimental method. CC and GZ guidance of experimental ideas. All authors contributed to the article and approved the submitted version.