Abstract
To reveal genetic factors or pathways involved in the pod degreening,
we performed transcriptome and metabolome analyses using a yellow pod
cultivar of the common bean “golden hook” ecotype and its green pod
mutants yielded via gamma radiation. Transcriptional profiling showed
that expression levels of red chlorophyll catabolite reductase (RCCR,
Phvul.008G280300) involved in chlorophyll degradation was strongly
enhanced at an early stage (2 cm long) in wild type but not in green
pod mutants. The expression levels of genes involved in cellulose
synthesis was inhibited by the pod degreening. Metabolomic profiling
showed that the content of most flavonoid, flavones, and isoflavonoid
was decreased during pod development, but the content of afzelechin,
taxifolin, dihydrokaempferol, and cyanidin 3-O-rutinoside was
remarkably increased in both wild type and green pod mutant. This study
revealed that the pod degreening of the golden hook resulting from
chlorophyll degradation could trigger changes in cellulose and
flavonoids biosynthesis pathway, offering this cultivar a special color
appearance and good flavor.
Keywords: common bean (Phaseolus vulgaris L.), degreening, pod,
transcriptome, cellulose
Introduction
The common bean (Phaseolus vulgaris L.), as one of the most important
legume crops worldwide, provides a major source of dietary protein,
complex carbohydrates, dietary fiber, numerous vitamins, and trace
minerals ([41]Tharanathan and Mahadevamma, 2003). The common bean is
cultivated from 52°N to 32°S latitude and from near sea level to
elevations of more than 3,000 m. Among a large number of cultivated
varieties and landraces, there is a high level of diversification in
nutrient content, taste, color, and texture of pod and seed in the
common bean.
Plant pigments, including carotenoid, anthocyanin, and chlorophyll, are
in charge of fruit peel or flesh color ([42]Tanaka et al., 2008). For a
specific fruit or vegetable, the content and proportion of different
pigments are the decisive factors for the color appearance. Chlorophyll
is an essential photosynthetic pigment in chloroplast of higher plants
and performs complex processes of harvesting light energy and driving
electron transfer. Many fruit flesh or peel and leaves with green color
are caused by the high chlorophyll content ([43]Xu et al., 2020).
However, yellowing or senescence is often accompanied by chlorophyll
degradation. Chlorophyll and carotenoids are related to the color
variation from green to yellow. When carotenoids are masked by
excessive chlorophyll, the fruit flesh or peel and leaves appear green.
The yellow color of carotenoids is unmasked upon chlorophyll
degradation during ripening or senescence. Chlorophyll breakdown is a
complicated multistep enzymatic process. CLH is a crucial factor in the
regulation of the chlorophyll reduction in the pericarp
([44]Harpaz-Saad et al., 2007). In two pear cultivars, one turned
yellow during ripening due to loss of chlorophylls a/b and carotenoids,
while the other stayed green until fully ripen, which could be
accounted for the lower the expression levels of the genes for
chlorophyll degradation, including CLH, PAO, and NYC1-like
([45]Charoenchongsuk et al., 2015). Treatment with 1-MCP could maintain
intact chloroplasts with well-organized grana thylakoids and small
plastoglobuli and delay chlorophyll degradation by suppressing the
expression of PAO, NYC, NOL, and SGR1 ([46]Cheng et al., 2012). The SGR
gene and its homolog, SGR-like, had been detected in various plant
species; overexpression of SGRL reduced the chlorophyll content and
promoted chlorophyll breakdown ([47]Sakuraba et al., 2014). Synthesis
and degradation of chlorophyll are under control of the coordinate
regulatory cascade, in which the malfunction of key steps or genes
would affect total chlorophyll content and the color appearance.
The key factor that restricts the quality of fresh pods is the
cellulose content. The tender pods of common beans with low cellulose
content have a pleasant taste. In order to take full advantage of the
common bean in human diet, it is necessary to improve the quality and
reduce cellulose content. Cellulose biosynthesis is a complex
biochemical process, which includes various enzymes, such as CESA, Kor,
and SuSy. Uridine diphosphate-glucose (UDP-glucose) is regarded as the
immediate substrate for cellulose polymerization in higher plants.
Photosynthetically fixed CO[2] is the ultimate source of C for the
synthesis of nucleotide sugars, such as UDP-glucose, which are the
building blocks for synthesis of cell wall polysaccharides ([48]Nakai
et al., 1999). UDP-glucose can be derived from the cleavage of sucrose
catalyzed by SuSy yielding UDP-glucose and fructose, demonstrating that
SuSy had tight association with cellulose synthesis and the
availability of sucrose in the cell would affect the rate of cellulose
synthesis ([49]Coleman et al., 2009).
Flavonoids, including flavone, flavonol, flavanone, isoflavone, and
anthocyanin, constitute an important group of plant secondary
metabolite, which can enable plants to adjust to environmental
pressures ([50]Kovinich et al., 2014). Recent researches showed that
these compounds have physiological functions such as antioxidant,
bacteriostatic, and anti-inflammatory, which are beneficial to human
health. Especially isoflavonoid is predominantly synthesized in legumes
plants. Anthocyanin, a class of flavonoids, localized in vacuoles,
provided a wide range of colors ranging from orange/red to violet/blue.
The content and variety of anthocyanins are the primary determinants of
color in many fruit peel and flesh or flowers; the family of MYB and
WD40 transcription factors and DFR and CHS had significant regulatory
function on anthocyanin synthesis ([51]Wang et al., 2019; [52]Zhuang et
al., 2019).
Among abundant germplasms, including landraces, ecotypes, and
cultivars, there is a wide range diversification in pod and seed coat
colors and patterns in the common bean. By fine mapping of QTL, the
genes controlling the color of the seed coat, pod, stem, and flower
were mapped on chromosomes 1 and 3; UDP flavonoid glycosyl transferase
(UGT) was identified as the candidate gene for black seed coat in
Adzuki Bean ([53]Li et al., 2017). Through an Andean intragene pool
recombinant inbred line (RIL) population, 23 QTLs for 6 pod traits were
detected, and 4 QTLs for pod color were identified ([54]Yuste-Lisbona
et al., 2014). The R2R3 MYB transcription factor TcMYB113 regulated
green/red pod color in cacao ([55]Motamayor et al., 2013). The
characterization of the genetic variability of three stay-green common
bean cultivars indicated high initial chlorophyll a content and reduced
chlorophyll degradation throughout senescence. In the common bean, the
pod color, especially purple, is well documented; Lablab with purple
pods contained the pigments of anthocyanins and flavonol. The
expression patterns of LpPAL, LpF3H, LpF30H, LpDFR, LpANS, and LpPAP1
were significantly induced in purple pods compared to the green ones
([56]Cui et al., 2016). In purple kidney bean pod, PvMYB1, PvMYB2 (R2R3
MYB transcription factors), and PvTT8-1 [basic helix–loop–helix (bHLH)
transcription factors] might play a crucial role in transcriptional
activation of most anthocyanin biosynthetic genes ([57]Hu et al.,
2015). A locus, Prp (purple pod), having five alleles affecting
anthocyanin pigmentation of corolla and pod, was detailed in Phaseolus
vulgaris L. The allele Prp produced dark purple corolla ([58]Okonkwo
and Clayberg, 1984). Chlorophyll degradation and chloroplast breakdown
may be involved in degreening and be the foundation for the appearance
of other colors; however, the regulatory mechanism at the molecular
level has not been well studied in the common bean.
The construction of mutant libraries provides researchers and breeders
with good resources for fundamental research or breeding materials.
Dalong 1, having unique yellow and tender pods, is an important common
bean ecotype with superior nutrition and is a favorite vegetable in
northeast China. This cultivar with a determined growth habit does not
require trellises to climb. We constructed a gamma radiation mutant
library of Dalong 1. Among various phenotypic changes such as leaf
morphology, pod dimension, and sterility, a higher proportion of green
pod mutations were noted. In order to reveal the physiology and
molecules mechanism underlying the yellow to green mutation, we first
investigated the chlorophyll and cellulose content and performed
transcriptome and metabolome analyses at different pod developmental
stages. Key genes or metabolites triggered by the degreening of pod
were revealed, which would provide new insights into the detailed
regulatory mechanism of degreening process in the common bean.
Materials and Methods
Common Bean Ecotype of Golden Hook and Its Green Pod Mutation Lines
Common bean mutant population was constructed by exposing the seeds of
cultivar Dalong 1 of ecotype of the golden hook to ^60Co γ radiations
at a dosage of 200 Gy. Phenotype observation was performed at M[3]
generation. A total of 76 mutant lines with green pods were noted in
this mutant library. The wild-type Dalong 1 (yellow pod, YP) and the
green pod mutant M628 (green pod, GP) were cultivated in Harbin City,
Heilongjiang Province of China in 2018. After fertilization, the pods
were sampled at different growth stages based on pod length: 2, 5, and
10 cm. In total, six treatments, e.g., YP-2, YP-5, YP-10, GP-2, GP-5,
and GP-10, were sampled in triplicate. For further confirmation, we
selected another two green pod mutant lines (M693 and M756) and one
mutant line (M729) whose pod color was segregated (yellow pod named
M729Y and green pod named M729G). Three biological replicates were
collected per treatment. The samples were immediately frozen in liquid
nitrogen and then stored at −80°C for the following experiments.
Chlorophyll Content Measurement and Ultrastructure of Chloroplast
Chlorophyll Determination
Two hundred milligrams of plant sample was extracted in 25 ml of 80%
(v/v) ethyl alcohol in a dark at room temperature for 48 h. The
absorbance (A) of the supernatant was measured at 665, 649, and 470 nm,
recorded as A665, A649, and A470, respectively. The pigment
concentration was calculated using the following formulas:
[MATH: Ca(mg/L)=13.95A665<
/mn>−6.88A649
Cb(mg/L)=24.96A649<
/mn>−7.32A665
Ct(mg/L)=Ca+CbCxc(mg/L)=(1000A470−2.05Ca−114.8Cb)/245
:MATH]
The pigment content was calculated as pigment concentration multiplied
by extraction liquid product and divided by fresh weight.
Chloroplast Ultrastructure
The pod tissues, 2 mm^3 in volume, were fixed in 2.5% glutaraldehyde
fixation solution at 4°C for more than 2 h. The tissues were washed
with phosphate-buffered saline (PBS) at 4°C for 15 min, followed by
washing with acetone of different concentrations (50, 70, 90, and
100%). After washing, the resin was soaked overnight and polymerized at
a high temperature, sliced by a microtome (Leica, Germany), and was
stained by uranium acetate and lead citrate. The copper mesh was
covered on the sample drop for adsorbing for 5 min and stained with
phosphotungstic acid for 30 s. The samples were examined with a
transmission electron microscope (H-7650, Hitachi, Japan).
Determination of Cellulose Content
Cell Wall Extraction
Sample (0.3 g) was added with 1 mL 80% ethanol, rapidly homogenized at
room temperature, and then incubated at 90°C in a water bath for 20
min. After cooling, samples were centrifuged at 6,000 × g for 10 min,
and the supernatant was discarded. The precipitate was washed using 1.5
ml 80% ethanol and acetone, respectively. One milliliter reagent I was
added for soaking for 15 h to remove starch. After being centrifuged at
6,000 × g for 10 min, the supernatant was discarded; the precipitate
was dried, weighed, and regarded as the cell wall mass (CWM).
Extraction of Cellulose
Five milligrams of dried CWM was homogenized in 0.5 ml distilled water.
Then, 0.75 ml of concentrated sulfuric acid was slowly added to tubes
in an ice water bath, mixed well and then incubated in ice for 30 min.
The tubes were centrifuged at 8,000 × g at 4°C for 10 min; the
supernatants were diluted with distilled water 20 times.
Cellulose Determination
One hundred fifty microliters of diluted supernatant was mixed with 35
μl working solution and 315 μl sulfuric acid. A blank tube was added
with 150 μl distilled water, 35 μl working solution, and 315 μl
sulfuric acid. After mixing, it was placed in water bath at 95°C for 10
min and cooled to room temperature. The absorbance values of blank and
sample tubes were read at 620 nm: ΔA = A (sample tube) − A (blank
tube). W stands for the dry weight of the cell wall mass.
[MATH: Cellulosecontent(mg/gdryweight)=3.17×(ΔA+0.0043)/W. :MATH]
RNA Extraction, cDNA Library Construction, and RNA Sequencing
Eighteen samples were used for the transcriptome analysis. Total RNA
was extracted from three biological replicates of each sample using
TRIzol reagent, according to the manufacturer’s instructions. RNA
purity was checked using a NanoPhotometer spectrophotometer (IMPLEN,
CA, United States). RNA concentration was measured using a Qubit RNA
Assay Kit in Qubit 2.0 Flurometer (Life Technologies, CA, United
States). RNA integrity was assessed using the RNA Nano 6000 Assay Kit
of the Agilent Bioanalyzer 2100 system (Agilent Technologies, CA,
United States).
A total of 1.5 μg RNA per sample was used as input material for the RNA
sample preparations. Sequencing libraries were generated using NEBNext
Ultra RNA Library Prep Kit (New England Biolabs, United States), and
index codes were added to attribute sequences to each sample. The
library preparations were sequenced on Illumina Hiseq platform, and
paired-end reads were generated. Gene function annotation and DEGs were
analyzed.
Gene function was annotated using the following databases: NCBI
non-redundant protein sequences (Nr), Clusters of Orthologous Groups of
protein (KOG), GO, and KEGG. All the unigenes were searched against KO,
KOG, and KEGG databases using the BLASTX algorithm.
The clean reads were mapped back onto the assembled reference
transcriptome (EnsemblPlants^[59]1), and gene expression levels were
estimated by RSEM ([60]Li and Dewey, 2011) for each sample.
Differential expression analysis of two conditions/groups was performed
using the DESeq R package. DESeq provides statistical routines for
determining differential expression in digital gene expression data
using a model based on the negative binomial distribution. The
resulting P values were adjusted using the Benjamini and Hochberg’s
approach for controlling the false discovery rate. Genes with an
adjusted P < 0.05 found by DESeq were assigned as differentially
expressed. For pathway enrichment analysis, all DEGs of each comparison
group were mapped to pathways in the KEGG database by KOBAS software to
identify significantly enriched KEGG pathways ([61]Kanehisa et al.,
2008).
Real-Time Quantitative PCR Validation
For RT-qPCR, oligonucleotide primers were designed according to each
gene’s transcript sequence with Primer 3 and Beacon Designer 7
software. Actin11 (Phvul.008G011000) was used as the reference gene
([62]Borges et al., 2012). RT-qPCR was carried out using SYBR
Green-based PCR assay in LightCycler 96 (Roche, Switzerland). Each
reaction mix contained 1.0 μl of complementary DNAs (cDNAs), 7.5 μl
SYBR Premix ExTaq, 0.3 μl PCR forward primer (10 μmol L^–1), 0.3 μl PCR
reverse primer (10 μmol L^–1), and 5.9 μl ddH[2]O, to a final volume of
15 μl. The PCR conditions were 95°C for 30 s, followed by 40 cycles of
95°C for 10 s, and 60°C for 30 s. The melting curve conditions were
95°C for 15 s, 60°C for 60 s, and 95°C for 1 s. Each RT-qPCR analysis
was performed in triplicate, and the mean was used for RT-qPCR
analysis. The relative expression of the genes was calculated according
to the method of 2^–△△Ct, and SPSS was used to analyze the data.
Preparation of Metabolome Samples and Measurement by HPLC
The freeze-dried pod was homogenized using a mixer mill (MM 400,
Retsch) with a zirconia bead for 1.5 min at 30 Hz. One hundred
milligrams powder of pod was extracted overnight at 4°C with 1.0 ml 70%
aqueous methanol. Following centrifugation at 10,000 × g for 10 min,
the extracts were absorbed (CNWBOND Carbon-GCB SPE Cartridge, 250 mg, 3
mL; ANPEL, Shanghai, China^[63]2) and filtrated (SCAA-104, 0.22 μm pore
size; ANPEL, Shanghai, China^[64]3) before liquid chromatography–mass
spectrometry (LC-MS) analysis.
HPLC Conditions
The sample extracts were analyzed using an LC-electrospray ionization
(ESI)-MS/MS system (HPLC, Shim-pack UFLC SHIMADZU CBM30A system^[65]4 ;
MS, Applied Biosystems 6500 Q TRAP^[66]5). The analytical conditions
were as follows: HPLC, column, Waters ACQUITY UPLC HSS T3 C18 (1.8 μm,
2.1 mm × 100 mm); solvent system, water (0.04% acetic
acid):acetonitrile (0.04% acetic acid); gradient program, 100:0 V/V at
0 min, 5:95 V/V at 11.0 min, 5:95 V/V at 12.0 min, 95:5 V/V at 12.0
min, 95:5 V/V at 15.0 min; flowrate, 0.40 mL/min; temperature, 40°C;
and injection volume, 2 μl. The effluent was alternatively connected to
an ESI-triple quadrupole-linear ion trap (QTRAP)-MS.
ESI-Q TRAP-MS/MS
Linear ion trap (LIT) and triple quadrupole (QQQ) scans were acquired
on a QTRAP mass spectrometer, API 6500 QTRAP LC/MS/MS System, equipped
with an ESI turbo ion-spray interface, operating in a positive ion mode
and controlled by Analyst 1.6 software (AB Sciex). The ESI source
operation parameters were as follows: ion source, turbo spray; source
temperature, 500°C; ion spray voltage (IS), 5,500 V; ion source, gas I
(GSI), gas II (GSII), and curtain gas (CUR) were set at 55, 60, and
25.0 psi, respectively; the collision gas (CAD) was high. Instrument
tuning and mass calibration were performed with 10 and 100 μmol/L
polypropylene glycol solutions in QQQ and LIT modes, respectively. QQQ
scans were acquired as multiple reaction monitoring (MRM) experiments
with collision gas (nitrogen) set to 5 psi. DP and CE for individual
MRM transitions were done with further DP and CE optimization. A
specific set of MRM transitions was monitored for each period according
to the metabolites eluted within this period.
Statistical Analysis
Each experiment was repeated three times. All data were recorded as the
means ± standard deviation (SD) and analyzed using a one-way analysis
of variance (ANOVA) with SPSS software 19.0. Statistical difference
between samples (^∗P ≤ 0.05 was considered to be significant; ^∗∗P ≤
0.01 was considered to be extremely significant) was determined by
Duncan’s test.
Results
A High Rate of Yellow-to-Green Mutation Was Observed in a Gamma
Radiation-Induced Mutant Library
A mutant library was constructed by exposing seeds of Dalong 1 to ^60Co
γ radiations at a dosage of 200 Gy. Compared with the wild type, this
mutant library displayed a wide range of phenotypic variations in
traits, e.g., 100—seed weight, flower time, plant height, sterility,
pod width, pod, and seed coat color among 3,500 plants in M[3] lines.
In contrast to yellow pod of wild-type Dalong 1, there were 76 mutant
lines showing green pods or green and yellow segregation in a total of
607 M[3] lines, much higher mutation rate than other traits such as
sterility (33 mutant lines). Among the green pod mutant lines, the pod
color was unanimously green; however, the green and yellow segregation
displayed green and yellow pods segregation, approximately in a green
to yellow ratio of 3:1 or 15:1.
The pod color of wild-type Dalong 1 was green at the beginning of pod
development (2 cm long), which rapidly turned to yellow as pod extended
to approximately 10 cm long. The pod color of green pod mutant lines
stayed green until eventual maturation. The pods of Dalong 1 appeared a
little crooked, but green pod mutant was relatively straight
([67]Figure 1A). From the intersection, the tissues underneath the
epidermis of the pod also appeared green in consistent with pod
appearance ([68]Figures 1B,C).
FIGURE 1.
[69]FIGURE 1
[70]Open in a new tab
Pod color conversion, pigment content changes, and chloroplast
microstructure during pod development. (A) Phenotypes of common bean
pods of wild type (YP) and green pod mutant (GP). (B,C) phenotypes of
pod interior and cross-section of the wild type and green pod mutant.
(D) Phenotypes of green pod mutant with high cellulose content in the
pods. (E–G) Total chlorophyll and carotenoid contents and the Car/Chl
ratio at all developmental stages. (H,I) Structure of chloroplasts from
wild type (H) and mutant (I); t, thylakoids, sg, starch grain, p,
plastoglobuli. Scale bar: 1 μm and 500 nm.
Initially, a rapid colorimetric assay was employed to measure the
chlorophyll and carotenoid contents during pod development. As early as
the developmental stage of 2-cm-long pods, the chlorophyll and
carotenoid contents in the green pod mutants were higher than that in
the wild type. During pod development, the chlorophyll content in the
wild type decreased sharply. In contrast, the chlorophyll content in
green pod mutant lines only slightly decreased ([71]Figures 1E–G).
The chloroplast ultrastructure of pods at 10 cm long in green pod
mutant and wild type was compared using TEM. In contrast to the wild
type, the chloroplast of the mutant remained a well-organized grana
thylakoids and had more starch grain and less plastoglobuli
([72]Figures 1H,I), indicating that the chloroplasts were capable of
making chlorophyll.
Transcriptome Profiles
In order to reveal the transcriptional abundance of genes involved in
degreening process, transcriptome was performed using equal amounts of
RNA from pods. After removing low-quality reads, adaptor sequences, and
ribosomal RNA (rRNA) reads, we obtained 50407645 (YP-2), 48995896
(YP-5), and 54170305 (YP-10) and 59445277 (GP-2), 51527246 (GP-5), and
49269434 (GP-10) clean reads, respectively ([73]Supplementary Table
S1). Clean reads were then mapped to the common bean reference genome
using HISTAT software. Comparing YP-2 to YP-5, there were 4,916 genes
downregulated and 3,974 genes upregulated, while only 203 genes were
downregulated and 136 genes upregulated when YP-5 was compared with
YP-10. In the green pod mutant, the down- and upregulated genes were
780 and 406 for GP-2 vs. GP-5 and 3,228 and 1,700 for GP-5 vs. GP-10
([74]Figure 2A). Three compared combinations (YP-2 vs. YP-5, YP-5 vs.
YP-10, and YP-2 vs. YP-10 and GP-2 vs. GP-5, GP-5 vs. GP-10, and GP-2
vs. GP-10) shared 114 and 150 DEGs, respectively ([75]Figure 2C).
FIGURE 2.
[76]FIGURE 2
[77]Open in a new tab
The differential expression genes analysis between wild type and green
pod mutant by transcriptome. (A,B) The volcano plot showing the number
of differential expression genes between two comparative groups. (C)
Venn diagram showing common and specific DEGs numbers from different
combinations displayed in the overlapping and non-overlapping regions,
respectively. (D–F) Top 20 significantly enriched KEGG pathways in (D)
YP-2 vs. GP-2, (E) YP-5 vs. GP-5, and (F) YP-10 vs. GP-10. (G) The
statistics of DEGs in some metabolic pathway at two developmental
stages of wild type and green pod mutant.
To further determine the function of the DEGs during pod development,
KEGG analysis was performed. KEGG enrichment analysis revealed that
limonene and pinene degradation, histidine metabolism, fatty acid
elongation, and protein processing in the endoplasmic reticulum were
the prominently enriched for YP-2 vs. GP-2 ([78]Figure 2D). Comparing
YP-5 to GP-5, many pathways showed enrichment, such as fructose and
mannose metabolism, fatty acid metabolism, amino sugar, and nucleotide
sugar metabolism. Starch and sucrose metabolism and isoflavonoid
biosynthesis were also noticeably enriched ([79]Figure 2E). Neuroactive
ligand–receptor interaction and D-glutamine and D-glutamate metabolism
were the two most obvious enrichment pathways for YP-10 vs. GP-10;
other pathways, e.g., circadian rhythm plant, phenylpropanoid
biosynthesis, and carotenoid biosynthesis, were also enriched to a
certain extent ([80]Figure 2F).
As the pods extended, many marked changes occurred in numerous crucial
pathways, such as glycolysis, starch and sucrose metabolism, porphyrin
and chlorophyll metabolism, carotenoid biosynthesis, phenylpropanoid
biosynthesis, carbon metabolism, and plant hormone signal transduction.
In the wild type, radical changes in expression levels occurred from
YP-2 to YP-5, while in the green pod mutation line, contrasting changes
in expression levels were observed from GP-5 to Gp-10. Numerous
pathways were activated at the early stages in the wild type; there
were a large number of DEGs for YP-2 vs. YP-5 but fewer DEGs for YP-5
vs. YP-10. In contrast, more DEGs for GP-5 vs. GP-10 than GP-2 vs. GP-5
were detected ([81]Figure 2G).
Key Genes Involved in Chlorophyll Synthesis and Degradation Were Revealed by
Transcriptome
As the chlorophyll content decreased with the pods degreening
([82]Figure 1E), we focused on porphyrin and chlorophyll metabolism
pathway. It was evident from Supplementary the figure that the
expression levels of the genes for chlorophyll synthesis generally
showed a downward trend during pod development both in wild type and
green pod mutant ([83]Supplementary Figure S3). The genes for
chlorophyll synthesis, e.g., as HEMC (Phvul.002G034500), glutamyl
transfer RNA (tRNA) reductase (HEMA, Phvul.002G216100), CHLI
(Phvul.003G057600, Phvul.006G178400), and POR (Phvul.005G083700,
Phvul.011G148900), were decreasing throughout in the wild type
([84]Supplementary Figure S3A), whereas the genes for chlorophyll
synthesis in the green pod mutant increased slightly from GP-2 to GP-5
and thereafter decreased ([85]Supplementary Figure S3A). In particular,
the transcription of POR (Phvul.005G083700) showed the opposite pattern
when the pods extended from 2 to 5 cm, which was increased in the green
pod mutant but decreased steeply in wild type ([86]Supplementary Figure
S3A).
The expression levels of chlorophyll degradation-related genes started
to accumulate at early stage of wild type; notably, RCCR
(Phvul.008G280300) had high expression levels at YP-2 ([87]Figures
3A,B) and then gradually decreased. Similar trends were observed for
the genes of PAO (Phvul.011G086700) and SGR2 (Phvul.002G153100) in the
wild type ([88]Figure 3A). In the green pod mutant, the expression
levels of CLH (Phvul.007G278100) increased gradually, and SGR
(Phvul.009G132100) increased from GP-2 to GP-5 and then decreased
sharply ([89]Figure 3A). The expression levels of RCCR
(Phvul.008G280300) in GP were extremely low throughout the pod
development ([90]Figures 3A,B). The aggravated degradation of
chlorophyll together with the decreased synthesis resulted in the
chlorophyll content in the pods decreased sharply, leading to the
degreening phenotype.
FIGURE 3.
[91]FIGURE 3
[92]Open in a new tab
The differential expression genes analysis in porphyrin and chlorophyll
metabolism pathway of YP, GP, and the other green pod mutant lines. (A)
The changes in gene expression level of degradation in YP and GP,
respectively. (B) Expression levels of DEGs for chlorophyll degradation
between YP-2 and GP-2 and YP-5 and GP-5, respectively. (C) The
chlorophyll content in yellow pods (M729Y) and three green pod mutants
(M729G, M693, and M756), respectively. (D) Expression levels of DEGs
for chlorophyll degradation successively in the M729Y, M729G, M693, and
M756 mutant lines. Error bars represent the standard deviation of three
replicates. Data were analyzed using the t-test. *p < 0.05, **p < 0.01.
The other four mutant lines (M729Y, M729G, M693, and M756) were also
selected to analyze the variations of chlorophyll content and
expression levels of genes involved in chlorophyll metabolism pathway.
Similarly, the chlorophyll content declined throughout in these mutant
lines ([93]Figure 3C). The similar expression pattern of the genes for
chlorophyll synthesis was found in the other green pod mutant lines
([94]Supplementary Figure S3B). The genes involved in chlorophyll
degradation were highly expressed in the M729Y with yellow pods;
especially the expression levels of RCCR were high at the initial
stage. However, the expression levels of genes for chlorophyll
degradation in other green pod mutant lines were low (M729G, M693, and
M756) ([95]Figure 3D).
Cellulose Contents and Transcriptional Abundance of Genes Involved in
Cellulose Synthesis and Degradation
The tender pods of common bean are always consumed as vegetable,
especially in Chinese recipes. In some cultivars, fibrosis in the pod
leads to a hardened pod and loss of flavor. Therefore, eliminating
fibrosis is a main target for breeding. After measuring, the cellulose
content in the green pod mutant was higher than that in the wild type
([96]Figure 4A). This result was also verified in other green pod
mutant lines; the green pod mutant lines had higher cellulose content
([97]Figures 4B,C). Therefore, we focused on cellulose synthesis and
degradation and cellulose synthase genes. The genes of
Phvul.004G093300, Phvul.005G022100, and Phvul.009G242700 participating
in cellulose synthesis had distinctly higher expression in GP than in
YP samples ([98]Figures 4D–F). The cellulose in the green pod mutant
appeared to rapidly synthesized, leading to a higher cellulose content.
There were also more genes involved in cellulose degradation in GP than
in YP, which might be due to the degradation activity of cellulose
being triggered by a higher cellulose content ([99]Figures 4G–I).
FIGURE 4.
[100]FIGURE 4
[101]Open in a new tab
The cellulose content and the differential expression genes analysis in
starch and sucrose metabolism pathway of YP and GP. (A) The cellulose
content of wild type and green pod mutant. (B,C) The cellulose content
in the other green pod mutant lines, (B) M729Y and M729G and (C) M693
and M756. (D–F) Differential expression genes for cellulose synthesis
between YP-2 and GP-2, YP-5 and GP-5, and YP-10 and GP-10,
respectively. Phvul.004G093300 (cellulose synthase A catalytic subunit
2, CESA2), Phvul.005G022100 (CESA3), Phvul.002G188600, Phvul.009G090100
and Phvul.009G242700 (CESA4), Phvul.007G081700 (CESA6),
Phvul.003G154600, Phvul.009G205100 (CESA7), Phvul.003G290600 (cellulose
synthase-like protein G2, CSLG2), Phvul.008G279800, and
Phvul.006G058400 (cellulose synthase-like protein E6, CSLE6). (G–I)
Differential expression genes for cellulose degradation between YP-2
and GP-2, YP-5 and GP-5, and YP-10 and GP-10, respectively.
Phvul.011G145900, Phvul.011G200700, Phvul.011G202500, Phvul.001G112700,
Phvul.002G201800, Phvul.011G145900, Phvul.011G202000, Phvul.011G202500,
Phvul.007G250300, and Phvul.008G210100 (beta-glucosidase).
Phvul.002G297600, Phvul.002G318800, Phvul.006G133700, Phvul.007G218400,
Phvul.008G212700, and Phvul.009G016100 (endoglucanase). Error bars
represent the standard deviation of three replicates. Data were
analyzed using the t-test. *p < 0.05, **p < 0.01.
Real-Time Quantitative PCR Validation
To validate the key DEGs, the DEGs that showed significant differences
in expression levels were selected: 12 chlorophyll synthesis pathway
genes, 6 chlorophyll degradation genes, and 5 and 4 genes for cellulose
synthesis and degradation, respectively. To investigate the changes in
expression levels of the genes related to chlorophyll pathway, six
different development stages of pods were selected. The results of
RNA-seq were confirmed by RT-qPCR. Particularly, some genes, such as
HEMA2, HEME1, HEMF, CHLI2, CAO, and POR, had a high expression when the
pods were 7 cm ([102]Figure 5A). For chlorophyll degradation, PAO and
SGR played a significant role. The expression levels of RCCR in the
green pod mutant were low; however, they were very high in the early
stage in in the wild type ([103]Figure 5B). The expression levels of
cellulose synthesis and degradation genes showed a similar pattern to
the transcriptome data ([104]Figures 5C,D).
FIGURE 5.
[105]FIGURE 5
[106]Open in a new tab
Quantitative real-time PCR (RT-qPCR) validation of DEGs found
throughout transcriptome between the wild type and mutant. (A) DEGs
involved in chlorophyll synthesis. (B) DEGs involved in chlorophyll
degradation. (C) DEGs involved in cellulose synthesis. (D) DEGs
involved in cellulose degradation.
Meanwhile, the expression levels of the genes related to chlorophyll
pathway were studied in different tissues (root, stem, leaf, and
flower) using RT-qPCR ([107]Figure 6). CHLI2 and POR had a high
expression in leaf, flower, and stem ([108]Figures 6B,D,F); the
expression of HEMF was higher in root ([109]Figure 6H). PAO and SGR2
had a high expression in leaf and flower ([110]Figures 6C,E); the
expression of CLH2 was high in leaf and stem ([111]Figures 6C,G).
Compared to other genes, the expression levels of CLH1, RCCR, and SGR1
were lower. Based on the above results, the mechanism model of pod
degreening was speculated ([112]Figure 7).
FIGURE 6.
[113]FIGURE 6
[114]Open in a new tab
Transcription levels of 18 DEGs involving in chlorophyll synthesis and
degradation pathway in different tissues. (A) The pictures of different
organs of common bean. (B–I) The relative transcript levels of
chlorophyll synthesis and degradation, respectively, in four different
organs: (B,C) leaf, (D,E) flower, (F,G) stem, and (H,I) root.
FIGURE 7.
[115]FIGURE 7
[116]Open in a new tab
The proposed mechanism model of color conversion of the wild type and
green pod mutant.
Transcriptome and Metabolome Variations in the Flavonoid Pathway
To determine the contents and composition variations of flavonoids
during pod development, we performed metabolisms analysis by widely
targeted metabolomics. A total of 154 flavonoid metabolites were
identified from 18 samples (YP-2, YP-5, YP-10, GP-2, GP-5, and GP-10),
each with three biological replicates. Setting VIP ≥ 1 together with
fold change ≥ 2 or ≤ 0.5 as thresholds for significant differences,
there were 33, 42, and 33 differential metabolites in the three
comparison groups: YP-2 vs. GP-2, YP-5 vs. GP-5, and YP-10 vs. GP-10,
respectively. Comparing the three comparison groups, 11, 12, and 7
metabolites were upregulated, and 22, 30, and 36 metabolites were
downregulated ([117]Figure 8).
FIGURE 8.
[118]FIGURE 8
[119]Open in a new tab
Heat map visualization, PCA, and OPLS-DA of the relative difference in
flavonoids metabolites of YP and GP. (A) Heat map of all detected
metabolites. The content of each metabolite was normalized to complete
linkage hierarchical clustering. Each example is visualized in a single
column, and each metabolite is represented by a single row. Red
indicates high abundance; green showed low relative content. (B) Score
plots for principle components 1 and 2 within and between groups. (C–E)
OPLS-DA model plots for YP versus GP at different comparison groups in
the same period. (F) The differentially metabolism analysis between
wild type and green pod mutant at different comparable groups.
Hierarchical cluster analysis of the flavonoids showed three main
clusters based on the relative differences in accumulation patterns
([120]Figure 8A), which indicated the complex differences in the
metabolite levels. In the PCA plot, PCA on all metabolites indicated
that PC1, explaining 44.75% of the total variance, separated samples of
three different stages. PC2, accounting for 16.43% of the total
variance, separated samples of YP from those of GP ([121]Figure 8B). In
the Orthogonal Projections to Latent Structures Discriminant Analysis
(OPLS-DA) models, YP and GP samples were clearly separated
([122]Figures 8C–E), indicating a major distinction in the metabolic
profiles between the two different color groups.
In the comparison group, YP-2 vs. GP-2 ([123]Table 1), naringin,
vitexin 2″-O-beta-L-rhamnoside, and isotrifoliin were significantly
higher in the YP-2, especially vitexin 2″-O-beta-L-rhamnoside (fold
change = 0.0009, VIP = 3.26). Compared to YP-5 ([124]Table 1), the
content of three isoflavones (prunetin, biochanin A, and sissotrin) was
higher in GP-5 and was 4. 15-, 2. 88-, and 8.81-fold, respectively. In
YP-10, the content of 3,7-di-O-methylquercetin was significantly
higher; other two flavonols (astragalin and kaempferin) were also
higher in YP-10. Two flavonoids, luteolin and apigenin, were higher in
GP-10. Compared to the wild type, isoflavones were largely accumulated
in the green pod mutant, especially sissotrin. The content of five
detected anthocyanin (oenin, pelargonin, cyanin, cyanidin, and
keracyanin) was higher in the wild type at three different stages
([125]Table 1).
TABLE 1.
Differentially accumulated flavonoids compounds in the pod of wild type
and green pod mutant.
YP-2 vs. GP-2 Index Metabolite name Class Content
__________________________________________________________________
VIP Fold change Type
YP-2 GP-2
Flavonoid pme0331 Naringin Flavanone 79,200.00 38,166.67 1.03E + 00
4.82E – 01 Down
pme3393 Fustin Flavonol 18,166.67 43,833.33 1.16E + 00 2.41E + 00 Up
pme0088 Luteolin Flavone 219,333.33 593,666.67 1.16E + 00 2.71E + 00 Up
Flavone and flavonol pme0363 Chrysoeriol Flavone 265,333.33 633,333.33
1.09E + 00 2.39E + 00 Up
pme3227 Vitexin 2″-O-beta-L-rhamnoside Flavone 9,623.33 9.00 3.26E + 00
9.35E – 04 Down
pme3212 Quercetin 3-O-glucoside Flavonol 4,450,000.00 1,413,333.33
1.29E + 00 3.18E – 01 Down
Anthocyanin pme0444 Malvidin 3-O-glucoside Anthocyanins 196,333.33
90,933.33 1.04E + 00 4.63E – 01 Down
pme1793 Pelargonin Anthocyanins 330,666.67 89,533.33 1.38E +00 2.71E –
01 Down
pme1777 Cyanidin 3,5-O-diglucoside Anthocyanins 49,966.67 23,966.67
1.02E + 00 4.80E – 01 Down
YP-5 vs. GP-5 YP-5 GP-5
Flavonoid pme0199 Quercetin Flavonol 488,000.00 179,666.67 1.03E +00
3.68E – 01 Down
pme0088 Luteolin Flavone 31,800.00 105,400.00 1.02E + 00 3.31E + 00 Up
pme0196 Kaempferol Flavonol 19,866.67 3,439.33 2.21E + 00 1.73E – 01
Down
Flavone and flavonol pme3296 Kaempferin Flavonol 12,933.33 3,746.67
1.20E + 00 2.90E – 01 Down
pme1541 Acacetin Flavone 9,696.67 38,300.00 1.22E + 00 3.95E + 00 Up
Isoflavonoid pme3292 Prunetin Isoflavone 11,070.00 45,933.33 1.27E + 00
4.15E + 00 Up
pme3250 Biochanin A Isoflavone 849.33 6,273.33 2.07E + 00 7.39E + 00 Up
pme3399 Sissotrin Isoflavone 10,406.67 91,633.33 1.59E + 00 8.81E + 00
Up
Anthocyanin pme3609 Cyanidin Anthocyanins 160,000.00 64,733.33 1.02E +
00 4.05E – 01 Down
YP-10 vs. GP-10 YP-10 GP-10
Flavonoid pme0088 Luteolin Flavone 3,170.00 7,193.33 1.05E +00 2.27E
+00 Up
pme0379 Apigenin Flavone 4,439.33 10,163.33 2.39E + 00 2.29E + 00 Up
Flavone and flavonol pme3290 3,7-Di-O-methylquercetin Flavonol 639.33
9.00 1.16E + 00 1.41E – 02 Down
pmb0604 Astragalin Flavonol 4,780,000.00 1,863,333.33 1.29E +00 3.90E –
01 Down
pme3296 Kaempferin Flavonol 4,206.67 1,476.67 1.18E + 00 3.51E – 01
Down
Isoflavonoid pme3502 Ononin Isoflavone 27,300.00 10,116.67 1.30E + 00
3.71E – 01 Down
pme3399 Sissotrin Isoflavone 9.00 956.00 1.24E + 00 1.06E + 02 Up
Anthocyanin pme1773 Cyanidin 3-O-rutinoside Anthocyanins 327,333.33
134,266.67 1.11E + 00 4.10E – 01 Down
[126]Open in a new tab
Analysis of transcriptome data revealed that most DEGs involved in the
flavonoids pathway showed to be downregulated ([127]Supplementary
Figures S4A,B); there were more DEGs in flavones and isoflavones
pathways in the wild type. Many genes of YP and GP samples showed
similar expression trends, such as DFR2, ANR, LAR, DFR3, IF7MAT6, VR4
([128]Supplementary Figures S4A–D). There were more genes expressed in
the pods at 5 cm long (YP-5, GP-5) ([129]Supplementary Figure S5). By
KEGG enrichment analysis, the DEGs (YP-5/GP-5) of isoflavonoid
biosynthesis pathway were enriched. The genes involving in anthocyanin
synthesis were rarely detected.
Discussion
The Molecular Mechanism of Degreening Process
The loss of green color in senescent leaves and ripening fruits is a
noticeable natural phenomenon. It is well known that the color
conversion is generally accompanied by chlorophyll degradation; the
chlorophyll content constantly decreased during the green-peel fruit
ripening ([130]Gambi et al., 2018). During the pod development of the
golden hook, the green pod color rapidly turned yellow, showing an
inherent characteristic as an ecotype of the common bean. The yellow
pods of the golden hook are tender with less cellulose and easy to
cook.
In this study, the color transition from green to yellow resulted from
the rapid degradation of chlorophyll and the appearance of carotenoids.
The PAO degradation pathway of chlorophyll is a multistep enzymatic
process, in which CLH, PAO, RCCR, and stay-green (SGR) genes have been
documented ([131]Tsuchiya et al., 1999; [132]Pruzinska et al., 2003;
[133]Harpaz-Saad et al., 2007). CLH, the first step enzyme in the
chlorophyll catabolic pathway, acts as a rate-limiting enzyme
controlled via posttranslational regulation ([134]Harpaz-Saad et al.,
2007). In this study, CLH1 showed low expression in both YP and GP,
whereas CLH2 was upregulated in the mutant during pod development.
Previous researches had shown that PAO and RCCR participated in the
chlorophyll catabolic pathway. PAO is a key enzyme and is upregulated
during leaf senescence ([135]Ginsburg et al., 1994; [136]Hörtensteiner
et al., 1995; [137]Pruzinska et al., 2003). The PAO activity seemed to
be responsible for the stay-green genotype in pepper ([138]Roca and
Minguez-Mosquera, 2006). The expression of RCCR is constitutive
([139]Rodoni et al., 1997; [140]Tsuchiya et al., 1999). In this study,
the expression levels of PAO were high in both wild type and green pod
mutant but particularly were increased in the wild type during pod
development. The expression of RCCR was very high in small tender pod
of the wild type and then declined; this means that RCCR acted as key
enzyme for chlorophyll degradation in early stages in the wild type.
There was an extremely low expression of RCCR in the green pod mutant
lines. Some studies suggested that SGR may take part in the chloroplast
breakdown process ([141]Kusaba et al., 2007; [142]Park et al., 2007).
SGR, as a chloroplast protein, has an important role in chlorophyll
degradation by catabolic enzymes and proteases through inducing LHCPII
disassembly ([143]Aubry et al., 2008). Our results support this notion;
the expression levels of SGR2 showed significant changes, but the
expression levels of SGR1 were low. The phytohormone ethylene and
jasmonic acid could affect the chlorophyll degradation pathway; they
could increase the expression of Chl catabolic genes ([144]Fang et al.,
2020; [145]Lv et al., 2020). Compared to the green pod mutant, the
expression levels of chlorophyll degradation genes in the wild type was
earlier and higher. PAO, RCCR, and SGR might be the key genes in the
golden hook involved in chlorophyll degradation; especially RCCR played
a much more important role. Analysis of the resequencing data of the
wild type and green pod mutant revealed that there were some no-sense
mutations occurring in the genes involving in chlorophyll degradation
([146]Supplementary Table S3).
In this study, comparing wild type to green pod mutant, only the pods
turned yellow during pod development; there were no noticeable color
differences in the other tissues. The expression levels of the genes
for chlorophyll degradation were low in the leaf and stem. According to
the results of this study, we developed a mechanic model describing
degreening process ([147]Figure 7), in which the chlorophyll content
decreased, and the structure of chloroplast breakdown in the wild type
during pod development. The expression levels of chlorophyll
degradation genes were higher and earlier in the pods of the wild type
leading to rapid degreening, especially RCCR showed a significantly
differential expression between the wild type and green pod mutant.
The Internal Mechanism of Phenotypic Changes in Cellulose Content
A common feature for different wall types of plants is the prevalence
of cellulose, which consists of β-1,4 glucan chains that are
synthesized by CESA protein complexes (CSCs). The CSC typically
consists of different heterotrimeric CESA configurations. Different
CESA has respective roles; during primary wall synthesis, the CSC
contains CESA1, 3 and one CESA6-like subunit ([148]Desprez et al.,
2007; [149]Persson et al., 2007). By contrast, the secondary
wall-synthesizing CSCs contain CESA4, 7, and 8 ([150]Taylor et al.,
2003). In this study, some CESA showed differential expression. The
genes (Phvul.004G093300, Phvul.005G022100, and Phvul.009G242700)
participating in cellulose synthesis had distinct higher expression in
GP than in YP samples. There were also more genes involving in
cellulose degradation in GP; the cellulose in the green pod mutant
showed rapid synthesis. The high cellulose content in pods might be one
of the reasons that the green pod mutant was much harder.
Cellulose biosynthesis is a complex biochemical process, which cannot
be reproduced in vitro. Fiber content is a key index of common bean
quality; the common bean with low fiber content has a good taste.
Golden hook is a very popular ecotype of the common bean and has a very
low cellulose content in fresh pods. There were also some genes for
cellulose synthesis having relative low expression in the wild type.
Conversely, there was relatively high cellulose content in green pod
mutants at the same stage; MeJA treatment promoted peach fruit
chlorophyll degradation and affected fruit softening through increasing
the expression of cellulase ([151]Wei et al., 2017). The genes for
cellulose synthesis pathway might be downregulated by the genes
involved in degreening of pods.
Sucrose is the main form of assimilated carbon that is produced during
plant photosynthesis; UDP-glucose can be transformed from sucrose
catalyzed by SuSy. UDP-glucose was just the immediate substrate for
cellulose synthesis ([152]Nakai et al., 1999; [153]Verbancic et al.,
2018). It was speculated that photosynthesis of green pod mutants might
provide synthetic substrate for cellulose, but the intrinsic
interaction need to be further studied.
Transcriptome and Metabolome Analyses Involved in Flavonoid Pathway
Flavonoids, important secondary metabolites found in plants, contribute
to plant environmental adaptation, fruit development ([154]Petroni and
Tonelli, 2011; [155]Li et al., 2019), and human health ([156]Alipour et
al., 2016). These compounds are accumulated in various tissues and are
also the direct factors that cause color variations in many fruits or
flowers. Combined metabolome and transcriptome analyses were carried
out with Ficus carica L.; cyaniding O-malonylhexoside demonstrated a
3,992-fold increase; cyaniding 3-O-gglucoside, cyanidin-3, and
5-O-diglucoside were upregulated 100-fold, revealing the anthocyanins
underlying the purple mutation ([157]Wang et al., 2017). The pod color
and pattern of the common bean is colorful; some varieties having red
or purple patterns could be rich in anthocyanin. On the other hand, the
common bean is a special legume variety that is abundant in flavonoids,
especially isoflavones. In a purple kidney bean cultivar, malvidin
3,5-diglucoside was identified as the major anthocyanin in the pod skin
by HPLC-ESI-MS ([158]Hu et al., 2015). Naringin, vitexin
2″-O-beta-L-rhamnoside, and isotrifoliin demonstrated significantly
higher contents in the YP-2. Compared to YP-5, the content of three
isoflavonoid (prunetin, biochanin A, and sissotrin) was higher in GP-5.
Only five anthocyanin were detected (oenin, pelargonin, cyanin,
cyanidin, and keracyanin), and all these five anthocyanins were
significantly higher (VIP >1, fold change <0.5) in the wild type than
that in the green pod mutant. When the pod length reached to 10 cm,
cyanidin 3-O-rutinoside (keracyanin) was the sole anthocyanin with
datable difference between GP-10 and YP-10. Thus, it could be
speculated that anthocyanins content could be influenced by the
degreening to certain extent but not significantly.
The Whole Nutrition Value and Breeding Strategy in Common Bean
The common bean is a nutrient-rich food that contains nutrients
essential for humans, such as proteins, minerals, vitamins,
carbohydrates, and fiber. The protein quality in the common bean is
high, and many cultivars have sufficiently high levels of essential and
non-essential amino acids to meet daily nutritional needs ([159]Ribeiro
et al., 2007). Although the protein content of the common bean is
approximately 50% of that of soybean, the digestibility of protein is
higher (78.70%). In view of the high mineral content, common beans
benefit health and can be used to cure a number of mineral deficiencies
([160]Mesquita et al., 2007).
The yield potential of the common bean is determined by growth habit,
pod number, maturity features, and seed characteristics. In addition to
the yield, the external phenotype, taste, cook characteristic, and
disease and abiotic stress tolerance are prime goals in breeding. As a
high-quality variety of oil beans, Dalong 1 has high nutritional value,
less cellulose, and good taste. In this study, we studied the color
conversion of pods and flavonoids metabolism between the wild type and
green pod mutants; it was helpful to understand the metabolic process
of nutrients and provide a theoretical basis for breeding.
Data Availability Statement
All the information regarding our sequencing data we deposited at
National Genomics Data Center BIG Sub repository: Project No:
PRJCA002296, GSAs: CRA002791, Experiment acc: CRX119315–CRX119332,
Samples acc: SAMC138882–SAMC1388.
Author Contributions
BH designed the experiments, carried out the research, and wrote and
revised the manuscript. JY, DZ, HW, and NG prepared plant materials. JZ
and YG helped to collected samples in the field. KX, HZ, and ZX gave
important suggestions and assisted in data analysis. All authors
participated in this study and approved the final version of the
manuscript.
Conflict of Interest
The authors declare that the research was conducted in the absence of
any commercial or financial relationships that could be construed as a
potential conflict of interest.
Acknowledgments