Mulberry (Morus alba
] is a woody plant native to China that is commercially valuable. The most important use of mulberry is as the sole food source of the domesticated silkworm (Bombyx mori
L.). However, mulberry is also used in animal fodder, pharmaceuticals, food production, and landscaping [2
]. Polyploidy is a heritable change in which the entire chromosome set is multiplied, and it plays an important role in plant evolution [4
]. Two forms of polyploidy are often considered: allopolyploidy, which originates from interspecies hybrids, and autopolyploidy, which originates from intraspecies genome duplication events. Polyploidy is particularly widespread in the flowering plants (angiosperms), including many major crops [5
]. Polyploid plants are often larger and have larger organs than their diploid relatives, including higher yield, larger leaves, larger fruit, more robustness, and some other agronomic characters [6
], which makes polyploids quite appealing for agricultural breeding.
How are the larger plants regulated by the polyploidization? The most naïve hypothesis was that increase in gene copy number increased the amount of protein, which in turn increased the cell volume [9
]. It was found that the ploidy-dependent increase in cell volume is genetically regulated in the experiment of investigating a wide range in cell size by tetraploidizing various mutants and transgenics of Arabidopsis thaliana [10
]. Early research reported that polyploidization increased the chloroplast number and photosynthesis per cell, which may be due to increasing size of cells [11
]. However, the mechanism behind the ploidy-related regulation of cell size, cell proliferation and expansion remains largely unclear. In recent years, plant breeders have worked with polyploids in mulberry and several artificially generated polyploids with “larger” mulberry characteristics have been reported [12
]. Hence, we sought to investigate the physiological and molecular mechanisms for the enlargement phenomenon in mulberry polyploids.
Transcriptome-wide gene expression analysis has been demonstrated in many bred and natural polyploid plants. Research on transcriptional analyses of autotetraploids and their related diploids show a significant divergence in species-specific traits even though a great deal of common characteristics also exist [5
]. For example, only ~1%–3% of genes are significantly differentially expressed between the autotetraploids and diploids of Arabidopsis (Arabidopsis thaliana
L.), rice (Olyza Sativa
L.), and Chinese woad (Isatis indigotica
], whereas ~10% of potato (Solanum phureja
L.), birch (Betula platyphylla
Suk.), and Paulownia
Hemsl.) genes are significantly differentially expressed [8
]. In addition, RNA profiling using different tissues may partly cause transcriptome divergence. Among woody plants, the biosynthesis and signal transduction of indole-3-acetic acid and ethylene have been altered by a genome duplication event in birch [8
], whereas differentially expressed transcripts are enriched in the energy metabolism pathway and in genetic information storage in Paulownia
]. Hence, it is necessary to investigate the changing expression in key genes after polyploidization in the mulberry.
RNA-Seq is a powerful tool for detailed transcriptomic studies that is cost-efficient and yields a far greater amount of information than traditional sequencing technology [21
]. In this study, we obtained a series of mulberry autotetraploids using a colchicine treatment and compared differences between the transcriptomes of diploid and autotetraploid mulberry plants using RNA-Seq technology. The results improved our understanding of the genetic regulation associated with mulberry autopolyploidization.
Polyploidy has played an important role in the evolution of angiosperms and was involved in the speciation of many important crops [28
]. Autopolyploidy is usually associated with increased plant, organ, and cell sizes, so polyploids generated through plant breeding have been used as tools to increase crop yields [29
]. By targeting relevant genes through transcriptomic techniques, the genetic basis of autopolyploidism has been investigated [19
]. RNA-Seq is a newly developed high-throughput sequencing technology that provides a powerful and cost-efficient research platform for transcriptional profile analyses [32
]. In this study, the transcriptomes of autopolyploid and diploid mulberry were investigated using Illumina RNA-Seq technology with mulberry genome sequences as the reference [22
]. We obtained 609 transcripts that are differentially expressed between autopolyploids and diploids, accounting for ~2.87% of the total genome sequences.
Transcriptomic analyses have been performed in the generated autopolyploids of several plants, including Arabidopsis, potato, rice, Chinese woad, Rangpur lime (Citrus limonia
Osbeck), birch, and Paulownia
]. The percentage of differentially expressed genes in generated autopolyploids varies from 1.08% in Rangpur lime [33
] to 12.6% in birch [8
]. In herbaceous plants, leaflet and root tip tissues of potato has ~10% differentially expressed genes among different ploidies, whereas seedlings of Arabidopsis and pollen of rice have 1.63% and 2.59% between autotetraploid and diploid, respectively [16
]. Compared to the other woody plants studied, mature leaves of mulberry had a higher percentage (2.87%) of differentially expressed genes than the 1.08% in mature leaves of Rangpur lime, but much lower than the 12.6% in shoot tips of birch or the 9.49% in young leaves from Paulownia
]. Different tissues source for RNA profiling maybe another reason for transcriptome divergence. The percentages of differentially expressed genes between autopolyploids and diploids show species-specific and tissue-specific features.
Among the genes demonstrating expression changes between autopolyploids and diploids, plant hormone-related genes are an important category [8
]. Ethylene- and auxin-related processes are controlled by highly regulated genes in autopolyploid Arabidopsis seedlings [16
], and the biosynthesis and signal transduction of the auxin and ethylene pathways are altered after genome duplication in birch [8
]. In Rangpur lime, GA- and auxin-related GO categories are over-expressed in autotetraploid plants [33
]. In this study, among the 609 regulated genes in the autotetraploids, 30 (4.9%) genes were associated with plant hormones (Table 3
). Cytokinin, GAs, and auxin—all of which are plant hormones that promote plant development and growth [34
]—were significantly affected in autotetraploid mulberry compared with in diploid mulberry (Table 3
There were six regulated genes related to cytokinin biosynthesis and signal transduction, and five regulated genes associated with GA signal transduction in the autotetraploids (Table 3
). Among them, two adenylate isopentenyltransferase genes—important enzymes catalyzed [23
] during cytokinin biosynthesis—were up-regulated, whereas three two-component response regulator ARR genes—which are negative regulators of cytokinin signal transduction [24
]—were down-regulated in autotetraploid mulberry (Table 3
). In the GA signal transduction pathway, a GA receptor, the GID1 gene, was up-regulated, whereas three genes encoding negative regulators of DELLA proteins [25
] were down-regulated in the autotetraploids (Table 3
). Levels of ethylene—a plant hormone suppressing development and growth [34
]—were significantly altered in autotetraploid mulberry (Table 3
). Of the 10 regulated genes related to ethylene, most of them were down-regulated in the autotetraploids (Table 3
). In summary, plant hormones—especially cytokinin, gibberellin, and ethylene—may play important roles in the phenotypic changes of autotetraploids.
The rate of photosynthesis and chloroplast numbers both increase in association with ploidy increases [11
]. In addition, photosynthesis-related genes are up-regulated in polyploidy plants [18
]. In Arabidopsis, photosynthesis- and chlorophyll-related GO categories are enriched in autotetraploids compared with diploids [16
]. In the present work, a series of differentially expressed genes that are involved in photosynthesis—including genes specifically expressed in chloroplasts, cytochrome genes, and photosystem-related genes—were up-regulated in autotetraploid mulberry (Table 4
). Several important genes were substantially up-regulated in autotetraploids, such as the photosystem I reaction center subunit XI gene and the homogentisate phytyltransferase gene (Table 4
). RNA for Illumina sequencing in this study were from mature leaves, the main tissues for photosynthesis, which may be another reason for so many different expressed genes related to photosynthesis. In brief, photosynthesis may be an important factor affecting phenotypic changes in autotetraploid plants.
The mechanism of polyploidization regulating larger organs has been studied for several years. Previous studies have reported that polyploidization increased the cell size, chloroplast number and photosynthesis per cell [10
]. In this study, we investigated the leaf cross-section of diploid and autotetraploid mulberry. We seemed to observe the cell size increased in autotetraploid compared with diploid but the cell number did not reveal an apparent difference (Figure 2
). On the transcriptome level, we found that gene expression of two important hormones, cytokinin and GAs—promoting plant growth and affecting cell size [34
]—were positively regulated in autotetraploid mulberry (Table 3
). Moreover, a series of photosynthesis related genes, including several chloroplast specifically expressed genes, were up-regulated in autotetraploid mulberry (Table 4
). It could be speculated further that mulberry autotetraploid could increase level of cytokinin and GAs, which thereby increased the cell size and photosynthesis, ultimately resulted in larger organs. Further research on the mechanism of larger organs regulated by polyploidization is needed.
4. Experimental Section
4.1. Plant Materials
Seeds of diploid mulberry (M. atropurpurea) were soaked in 0.1% colchicine for 48 h in the dark to induce autotetraploidy, and seeds soaked in distilled water under the same conditions acted as controls. The seeds were sown in a greenhouse after colchicine treatment, and 247 novel saplings and 39 control diploid saplings were transplanted into plastic pots in a greenhouse.
4.2. Ploidy Measurement
The DNA content of the leaves of two-month old seedlings was evaluated by flow cytometry using the methods of Galbraith et al.
] with some modifications. Three biological replicates performed of each sample. Briefly, ~0.5–1 cm2
young intact leaves were chopped in 1 mL of ice-cold extraction buffer (50 mM MgCl2
, 50 mM citric acid, 5 mM HEPES, 0.1% Triton X-100, and 1% PVP-40) using a new razor blade. The crude suspension was filtered through a 42-μm nylon filter to remove cell debris and then added to a propidium iodide staining solution to a final concentration of 50 μg/mL. After 1 h of incubation at room temperature, the fluorescence intensity was measured using a FACSAriaII (BD Biosciences, San Jose, CA, USA) flow cytometer with excited blue light at 488 nm and 5 × 108
J/s. The percentages of the cells that showed varied DNA contents were determined using BDFACSDiva software, and the DNA content of diploid leaves in the control group were measured in parallel.
Mitotic chromosomes were counted in young leaf bud of two-month old seedlings. Three biological replicates performed of each sample. Buds were collected and treated with a fixative buffer (concentrated hydrochloric acid/45% acetic acid/ethanol at 2:1:1 (v/v/v)) for 5 min. After flushing with water two or three times and immersing in water for 10 min, leaf buds were placed onto microscope slides and stained with one drop of carbolfuchsin and one drop of 45% acetic acid. Then, leaf buds were mashed with tweezers, the cytoplasmic residue was cleared, and the remains covered with glass. The chromosomes were visualized under a microscope (AxioScope A1, Carl Zeiss MicroImaging, Göttingen, Germany) using 1000× magnification. Approximately 10 metaphase cells were assessed for each leaf bud.
4.3. Phenotype Measurement
A series trees of autotetraploid (YY56) and diploid (TL) were adjacent planted in a plantation with the same parcel and climate condition. Ten adult stage (three-year old) mulberry trees of each cultivar were used in phenotype measurement. Leaf cross-sections were evaluated using scanning electron microscopy. An area ~1 cm2
from the center of mature leaves was fixed for 24 h in 2.5% glutaraldehyde fixation solution containing 2.5% glutaraldehyde and 0.1 M phosphate-buffered saline (PBS, pH 7.4) and dehydrated using a graded series of alcohol-isoamyl acetate concentrations, each for 15 min. The samples were dried using a critical point dryer (HCP-2, Hitachi, Tokyo, Japan), mounted on scanning electron microscopy stubs, sputter-coated with gold using an ion coater (Eiko IB-5, Hitachi), and observed under a scanning electron microscope (Philips XL30, Philips Electron Optics, FEI UK Ltd., Cambridge, UK). At least five leaf samples from each individual tree were viewed and measured at 20 kV using 2000× magnification. Leaf areas were measured from at least 10 healthy and fully expanded leaves collected at random from each tree. Height and breast-height diameters were measured from ten trees. The lengths, weights and diameters of the fruits were measured using a Vernier caliper, and the fruit maturation period were measured from end of the flowering to black fruit stage. At least 10 healthy fruits were selected at random from each tree and measured. SIGMASTAT from SPSS [39
] was used to analyze the morphological data. Student’s t
test were used to detect differences between autotetraploid and diploid at the usual probability level p
4.4. RNA Extraction, Illumina Sequencing, and Data Processing
For Illumina sequencing, three biological replicates from six independent adult stage trees of autotetraploid and diploid were used. Total RNA from each sample was extracted using the RNAiso Plus reagent (Takara BIO Inc., Otsu, Japan) and further purified using RNeasy Plant Mini kit reagents (Qiagen, Valencia, CA, USA). RNA quality was verified using a 2100 Bioanalyzer RNA Nanochip (Agilent, Santa Clara, CA, USA). All samples had an RNA integrity value of >7.5. RNA was then quantified using a NanoDrop ND-1000 Spectrophotometer (Nano-Drop, Wilmington, DE, USA). Total RNA (10 μg) was prepared for the cDNA library for each pool.
Illumina sequencing was performed at the Beijing Genomics Institute, Shenzhen, China using the HiSeq 2000 platform (Illumina, San Diego, CA, USA). First, poly-T oligo-attached magnetic beads (Illumina) were used to isolate poly(A) mRNA from total RNA. The purified mRNA was then fragmented into 200- to 700-nt pieces. The first strand of cDNA was synthesized using random hexamer primers, followed by synthesis of the second strand using SuperScript Double-Stranded cDNA Synthesis kit reagents (Invitrogen, Camarillo, CA, USA). The synthesized cDNA was subjected to end repair and phosphorylation using T4 DNA and Klenow DNA polymerases and T4 polynucleotide kinase, respectively. The repaired cDNA fragments were 3ʹ-adenylated using the Exo-Klenow fragment, and the Illumina paired-end adapters were ligated to the ends of these 3ʹ-adenylated cDNA fragments. To select templates for downstream enrichment, the products of the ligation reaction were purified by electrophoresis in a Tris-acetate-EDTA (2% w/v) agarose gel. cDNA fragments (200 ± 25 bp) were excised from the gel. Fifteen rounds of PCR were performed to enrich the purified cDNA templates using PCR primers PE 1.0 and PE 2.0 (Illumina) with Phusion DNA polymerase. Finally, after validation on an Agilent Technologies 2100 Bioanalyzer using Agilent DNA 1000 Chip kit reagents, the cDNA library was constructed by 50-bp, single-end RNA sequencing (RNA-Seq) in a PE flow cell using the Illumina Genome Analyzer HiSeq 2000.
Sequencing quality was evaluated and the data were summarized using the Illumina/Solexa pipeline software. Library saturation was also analyzed. For the raw data, adaptor sequences were eliminated, and distinct clean reads were identified. Subsequently, clean reads and distinct clean reads were classified based on their copy numbers within the library, and the percentages of total clean and distinct reads were calculated. The raw data have been deposited in the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) database [40
] under submission number GSE70428.
4.5. Annotation and Analysis of Sequence Data
For annotation, all sequences were mapped to the mulberry genome [22
]. The expression levels of each gene were estimated using the frequency of clean reads and then normalized to RPKM [41
]. Differential expression genes analysis was used the NOIseq metheod [42
]. Fold changes were assessed using the log2
ratio after expression abundances were normalized to RPKM. Differential expression genes were cutoff log2
) ≥ 1 and probability ≥ 0.8. Sequences were characterized using the GO database (http://www.geneontology.org/
]. Pathway assignments were determined with the KEGG pathway database [44
] using the blastx algorithm with an E
-value threshold of 1.0 × 10−5
. The differential gene expression analysis data have been submitted to the GEO database under submission number GSE70428.
4.6. qPCR and Statistical Analysis
Mature leaves of adult stage mulberry trees were used in qPCR. Primers were designed using Primer 5.0 software [45
]. Mulberry MaACT3
(GenBank accession number: HQ163775) gene was used as the reference gene. Expression levels for all of the candidate genes were computed based on the stable expression level of the reference gene. qPCR was performed in 96-well plates on a Roche LightCycler 480 system using SYBR-GREEN1 fluorescent reagents (Takara, Otsu, Shiga, Japan). Reactions were each carried out in 20 μL containing 0.4 μM (final concentration) of each primer (Table S3
). The qPCR thermal profile consisted of 95 °C for 30 s, followed by 40 cycles of 95 °C for 10 s, 58 °C for 10 s, and 72 °C for 10 s. Dissociation curves were obtained from a thermal melting profile generated under a final PCR cycle of 95 °C for 5 s followed by a constant increase in temperature from 65 to 97 °C. Threshold values were empirically determined based on the observed linear amplification phase of all of the primer sets. Sample cycle threshold (Ct
) values were standardized for each template based on the reference gene control primer reaction, and the 2−ΔΔCt
method was used to analyze relative changes in gene expression. Three biological replicates were used to ensure statistical credibility. SIGMASTAT from SPSS was used to analyze the qPCR data. Student’s t
tests were used to detect differences between autotetraploid and diploid at the usual probability level p