Induction, identication, and characterization of autoallo-dodecaploid barnyard grass (Echinochloa crus-galli L.) using in vitro colchicine treatment

Polyploidization is a major trend in plant evolution that has many advantages over diploid. Barnyard grass (Echinochloa crusgalli L.) has many good characteristics, but has not been fully utilized until now. In this study, we report for the rst time the in vitro induction of autoallo-dodecaploid E. crus-galli by colchicine treatment. Calli derived from young panicles were transferred to liquid medium containing different concentrations of colchicine (0.01, 0.05, or 0.1% w/v) and incubated for 24, 48, or 72 h. Treatment with 0.05% colchicine for 48 h was the most effective condition for producing polyploid plants, yielding 42.9% dodecaploids. The relative DNA content of the induced dodecaploids was twice that of wild-type hexaploids. The chromosome number of dodecaploids was 2n = 12x = 108, whereas that of hexaploids was 2n = 6x = 54. Compared with the hexaploids, the dodecaploids had larger individual stomata, but a lower stomatal density. There were signicant differences between dodecaploid and hexaploid plants in terms of morphological variables, such as plant height, leaf length, panicle length, and grain size. Dodecaploid plants showed the obvious “gigas” effects of polyploid organs, as well as signicantly reduced seed set. The nutritional concentrations of crude protein, crude fat, crude ash, and nitrogen-free extract in the dodecaploid were higher than those in the hexaploid, whereas the concentration of crude ber in the dodecaploid was lower. Compared with the hexaploid, the concentrations of calcium, iron and some free amino acids in dodecaploid plants were signicantly higher than in hexaploids. The dodecaploid E. crus-galli had been obtained successfully by treating calli with colchicine. And E. crus-galli has the potential to be developed as a new type of high quality forage crop for cultivation under stress conditions, especially the dodecaploid with its greater nutritional value. Echinochloa crus-galli L. was induced by in vitro colchicine treatment and some characteristics were identied. measurements were made on completely developed ag leaves of eld-grown plants. The lower epidermis was observed under a light microscope. The length and width of each stoma and the corresponding guard cells were measured with a micrometer. The stomata numbers within the grid squares were then counted to calculate the stomatal density. 30 stomata sizes and densities were measured randomly on three leaves for each population.


Introduction
Barnyard grasses (genus Echinochloa) belong to the subfamily Panicoideae (family Poaceae) (Giussani et al. 2001) and include about 50 annual or perennial species that are widely distributed throughout the world (Aoki and Yamaguchi 2008; Luo and Min 1990). These species are mainly regarded as weeds that compete with crops and cause reductions in crop yields (Chauhan and Johnson 2011;Rao et al. 2007; Xuan et al. 2006), but two species, Echinochloa utilis and Echinochloa frumentacea have been cultivated as food and forage crops for a long time (Luo and Min 1990;Yabuno 1987). Of the Echinochloa species, common barnyard grass (Echinochloa crus-galli) is the most widespread (Guo et al. 2017). Research has shown that E. crus-galli plants have a high nutritional value and good palatability to many herbivorous animals and sh. Moreover, E. crus-galli can grow on low fertility or saline-alkaline soils and produce high yields, unlike other cultivated grasses. But E. crus-galli has not been subjected to conscious or unconscious domestication or improvement until now. Making full use of E. crus-galli would allow the development of a novel crop for poor soils and improve the availability of human food and animal feed in inhospitable parts of the world (Sun et al. 1989).
Polyploidization, one of the most important evolutionary events in plants, can increase genetic diversity, introduce new genetic combinations, foster adaptation to different environments, and increase plant vigor (Comai 2005; Fang and Morrell 2016; Yu et al. 2021). Studies have shown that all angiosperm species may have experienced one or more polyploidization events during their evolution (Jiang et al. 2013;Jiao et al. 2011;Otto 2007). The development of polyploids is currently one of the most effective plant breeding techniques available for obtaining useful characteristics, such as vigorous growth, larger organs and biomass, more nutrients, and higher levels of resistance to abiotic stresses, pests, and diseases (Chen 2007 Previous studies had suggested that E. crus-galli is a natural allohexaploid, arising from hybridization and polyploidization between tetraploid Echinochloa oryzicola (the paternal donor) and an unknown diploid species (as the maternal donor) (Aoki and Yamaguchi 2008; Guo et al. 2017;Yabuno 1966). In the present study, autoallo-dodecaploid plants of E. crus-galli were successfully induced by treating calli with colchicine. The polyploidy was identi ed by stomatal characteristics, ow cytometry, and con rmed by chromosome counts, and morphological characters and plant nutrient concentrations were compared between dodecaploid and hexaploid plants.
Roots were prevalent after approximately 7-10 days, at which time the plantlets were acclimated for three days in the culture room, then washed in tap water, and transplanted to the open eld.

Ploidy analysis by ow cytometry
Flow cytometry analysis was performed as follows. Approximately 0.5 cm 2 of young leaf tissue from in vitro grown plantlets was chopped with a sharp razor blade in 0.5 mL of Partec HR-A buffer (Partec high-resolution nuclei extraction solution), in a plastic Petri dish. The sample was ltered through a 30-µm Partec CellTrics cell strainers directly into the sample tube, with 2 mL of HR-B buffer (Partec high-resolution DAPI staining solution). The sample was stained for 5 min at room temperature, and the relative uorescence of total DNA was measured by Sysmex CyFlow Ploidy Analyser II (Sysmex Partec GmbH, Münster, Germany) instrument.

Chromosome counts
After the plantlets had been transferred to the eld for about two weeks, root tips were excised from the plants and pretreated with 2 mM 8hydroxyquinoline for 2 h at room temperature and then xed in fresh Carnoy's xative [methanol/acetic acid, 3:1 (v/v)] overnight. Then, the root tips were rinsed in 75 mM KCl for 30 min at room temperature, digested in an enzyme mixture containing 2% cellulase and 2% pectinase at 28°C for 4 h, washed three times in distilled water, and incubated in distilled water at room temperature for 20 min. These root tips were then placed on precooled slides and squashed in the presence of the xative. The slides were heated over an alcohol ame to dry the xative, stained with 5% Giemsa in Sorensen's buffer at room temperature for 1 h, washed under a stream of tap water, and then dried at room temperature. The chromosomes were observed under an Olympus BX51 light microscope (Olympus Corporation, Tokyo, Japan) and photographed.

Stomatal observation
Stomatal measurements were made on completely developed ag leaves of eld-grown plants. The lower epidermis was observed under a light microscope. The length and width of each stoma and the corresponding guard cells were measured with a micrometer. The stomata numbers within the grid squares were then counted to calculate the stomatal density. 30 stomata sizes and densities were measured randomly on three leaves for each population.

Morphological observations
Morphological characteristics of dodecaploid and hexaploid plants, namely plant height, tiller number, ag leaf length and width, leaf thickness, panicle length, anther length, grain length and width, total grain number, lled grain number, and seed set were measured. Total grain number, lled grain number, and seed set were measured on 10 panicles, and the other characteristics were measured on 30 biological replicates for each population.

Determination of plant nutrient concentrations
The nutrient concentrations of the leaves of dodecaploid and hexaploid plants, such as concentrations of water, crude protein (CP), crude fat (ether extract, EE), crude ber (CF), crude ash (CA), and nitrogen-free extract (NFE), were determined in hexaploid and dodecaploid plants sampled two months after planting. In addition, the nutrient concentrations of total carotene (TCA), calcium, phosphorus, iron, and free amino acids were also determined. The samples were weighed (fresh weight) then dried at 105°C for 30 min, before being dried at 80°C for 8 h. The water contents of the samples were obtained from fresh weight minus dry weight, relative to fresh weight. The nutritional concentrations were determined according to Zhang (2016): CP was determined by the Kjeldahl method, EE was determined by the Soxhlet extraction method, CF was determined by the acid-alkali washing method, and CA was determined by the dry ash method. NFE was calculated using the equation: . TCA was determined by extraction with acetone and colorimetric (λ = 450 nm) determination. For elemental analysis, the samples were digested with concentrated nitric acid-perchloric acid at 200°C, and the concentrations of calcium, phosphorus and iron were determined using an inductively coupled plasma-optical emission spectrometer (ICP-OES) (Thermo Fisher Scienti c, USA). Free amino acids concentrations were determined by high-performance liquid chromatographic (HPLC) (Agilent, USA).
The concentration of each nutrient was determined on three biological replicates for each sample.

Data analysis
The data were analyzed using SPSS Statistics Version 24.0 (IBM, Armonk, NY, USA). Signi cant differences among the means of dodecaploid and hexaploid material were identi ed using the two-sample t-test.

Results
Survival of colchicine-treated calli The survival rates of calli decreased with increasing colchicine concentration or treatment time. Colchicine treatment at 0.1% for 72 h proved to be highly toxic, resulting in the death of all the calli ( Table 1). The colchicine concentration and treatment time also in uenced the frequency of plantlet regeneration, the number of plantlets decreasing with increases in either colchicine concentration or treatment time. The results showed that the highest polyploidy rate was 42.9%, in the 0.05% colchicine concentration treatment applied for 48 h ( Table 1). Identi cation of ploidy level by ow cytometry analysis The use of ow cytometry to identify ploidy level was faster than by conventional chromosome count methods. By comparison with the DNA content of G1 nuclei of hexaploid E. crus-galli, the ploidy level of each plant regenerated from colchicine-treated callus was determined by ow cytometry analysis. The results showed that 17 plants in all colchicine-induced populations exhibited a DNA content twice that of the control (hexaploid) plants (Fig. 1). The results con rmed the ploidy level of the colchicine-induced plants and showed that dodecaploid plants of E. crus-galli had been obtained successfully.

Determination of chromosome number
Chromosome counting is the most direct and accurate ploidy analysis method. The ploidy levels of the dodecaploid E. crus-galli determined by ow cytometry were con rmed by chromosome counts. Yabuno (1953) showed that the chromosome number for E. crus-galli was 2n = 6x = 54. Through chromosome counting in root tips, we found that the chromosome number of hexaploid E. crus-galli plants was indeed 54, while the chromosome number of the putatively dodecaploid plants was 2n = 12x = 108 (Fig. 2). This showed that the dodecaploid E. crus-galli had been successfully isolated and recovered by treating calli with colchicine in the present study.

Stomatal characteristics
Variations in stomata length and width, guard cell length and width, and stomatal density were recorded in both dodecaploids and hexaploids ( Table 2). The stomatal length and width of the dodecaploids (31.66 and 8.36 µm, respectively) were signi cantly larger (P < 0.01) than those of the hexaploids (18.16 and 6.24 µm, respectively). The guard cell length and width of the dodecaploids (46.63 and 9.86 µm, respectively) were also signi cantly greater (P < 0.01) than those of the hexaploids (31.83 and 6.99 µm, respectively), whereas the stomatal density in dodecaploid plants (115.95 mm − 2 ) was signi cantly lower (P < 0.01) than that in the hexaploid plants (175.64 mm − 2 ) (Table 2, Fig. 3). The results con rmed the tendency that stomata size increased but stomatal density decreased after polyploidization (Beaulieu et
2. Except for water content, the other nutrient concentrations were calculated based on dry matter.
3. Values represent the mean ± SD. Values within the same column followed by different uppercase and lowercase letters are signi cantly different at 0.01 and 0.05 probability levels, respectively.
Concentrations of total carotene, calcium, phosphorus, and iron. The result showed that the concentrations of TCA, calcium, phosphorus, and iron in dodecaploid E. crus-galli plants all increased (Table 4). Moreover, the concentration of calcium and iron in dodecaploid plants were signi cantly higher than those in hexaploid plants (P < 0.01 and P < 0.05, respectively).
Concentrations of free amino acids. The concentrations of free amino acids in hexaploid and dodecaploid plants are shown in Table 5. It can be seen the concentrations of all amino acids in dodecaploid plants increased with different extents except for aspartic acid (Asp). Especially, the concentrations of glycine (Gly), threonine (Thr), proline (Pro), methionine (Met), arginine (Arg) and total free amino acids were signi cantly higher than those in hexaploid plants (P < 0.01 for Arg, P < 0.05 for the others). Note: 1. The unit of data was µg/g.

2.
Values represent the mean ± SD. Values within the same column followed by different uppercase and lowercase letters are signi cantly different at 0.01 and 0.05 probability levels, respectively.
These results con rmed the trend of increased nutrient concentrations in plant polyploids (Dhawan and Lavania 1996;Li 2008), showing that polyploidization can be an important way to improve the feeding quality of crops.

Discussion
Key factors that affect the e ciency of induced polyploidy Colchicine is frequently used for polyploid induction in plants (Dhooghe et al. 2011). In this study, we demonstrated that autopolyploid plants of the allohexaploid wild-type E. crus-galli plant could be obtained by treating calli with colchicine. Colchicine concentration and exposure time are two key parameters for successful polyploidization, but there is an evident interaction between them. In the present study, low concentrations or short exposure times were not successful, while excessively high doses or long exposure times were lethal. Thus, every plant requires testing to nd the optimum balance between concentration and treatment duration (Dhooghe et al. 2011). In our study, the combination of 0.05% colchicine concentration for 48 h resulted in the highest rate of chromosome doubling ( Table 1).
The potential value of E. crus-galli as a forage crop E. crus-galli is the most widespread species in the genus Echinochloa (Guo et al. 2017), which re ects its broad and strong adaptability.
Research has shown that E. crus-galli leaves and seeds have high nutritive value. The crude protein concentration of the leaves is similar to that of rice and maize grains, while the cellulose concentration is about 34%, making it very suitable for the nutritional needs of herbivores. Feeding experiments showed that the palatability of E. crus-galli plants was high with respect to cattle, sheep, goat, rabbit, and sh (Sun et al. 1989).
In the present study, the autoallo-dodecaploid E. crus-galli was successfully obtained for the rst time from the allohexaploid E. crus-galli by in vitro chromosome doubling. The dodecaploid plant has a very high ploidy level (12x) and chromosome number (2n = 108) and higher plants with such high chromosome numbers are rare in nature. We found that the dodecaploid E. crus-galli could still grow normally and also had the common advantages associated with polyploidization. In terms of morphology, dodecaploid E. crus-galli showed the obvious "gigas" effects in polyploid organs, such as longer and thicker leaves, longer panicles, larger grains, larger stomata and longer stomatal guard cells. While the fertility of the dodecaploid plants was lowered, with a seed set of only 26.27% compared with 90.65% for the hexaploid; low fertility in the dodecaploid may allow for increased vegetative biomass by way of compensation and should not be a problem because barnyard grasses are not grown for their seed production. The dodecaploid plants had greater nutritional value than did the hexaploid plants. The concentrations of crude protein, crude fat, crude ash, and nitrogen-free extract in dodecaploid plants were higher, while the concentration of crude ber was lower than those in hexaploid plants, increasing the feed quality and palatability. Moreover, the concentrations of calcium, iron and some free amino acids in the dodecaploid plants were signi cantly higher than those in hexaploid plants.
The high nutrient concentrations of E. crus-galli determined in the present study were consistent with those which had already been reported (Sun et al. 1989;Yi and Peng 1993;Zhang et al. 1992). These results highlighted the good feeding value of E. crus-galli once again. Moreover, E. crus-galli had a well-developed root system, with strong tillering and abundant leaves, and tolerance to drought, waterlogging, and salinealkaline soils, as well as other harsh conditions, so that it can be planted on land where typical forage crops could not grow. In addition, E. crusgalli plants can be mowed several times once planted (Zhang et al. 1992). As a consequence of these valuable traits, attention is increasingly being paid to E. crus-galli by many researchers, who consider E. crus-galli to have the potential to be developed into a new type of high quality forage crop for stressful conditions. Such a development would be of great signi cance in improving the structure of animal husbandry (Zhang et al. 2018). The dodecaploid E. crus-galli created in this study could accelerate the development of such research because of its higher nutritional value. Although the seed set of the dodecaploid plants was approximately one-quarter that of the hexaploid, the high reproduction coe cient of E. crus-galli meant that one dodecaploid plant could still produce a large number of seeds, to meet the seed demand from livestock farmers for this potentially new forage crop. There is clear potential to develop hexaploid and dodecaploid E. crus-galli as a new type of forage crop in the future.

Conclusions
In the present study, autoallo-dodecaploid plants of E. crus-galli were successfully induced by treating calli with colchicine. The polyploidy was identi ed by stomatal characteristics, ow cytometry, and con rmed by chromosome counts, and morphological characters and plant nutrient concentrations were compared between dodecaploid and hexaploid plants. E. crus-galli has the potential to be developed as a new type of high quality forage crop for cultivation under stress conditions, especially the dodecaploid with its greater nutritional value. Comparison of stomata characteristics between hexaploid and dodecaploid E. crus-galli. a Hexaploid. b Dodecaploid