Skip to main content
NCBI home page
Physiology and Molecular Biology of Plants logoLink to Physiology and Molecular Biology of Plants
. 2023 Mar 22;29(3):349–360. doi: 10.1007/s12298-023-01298-5

Phylogenetic relationship and sequence diversity of candidate genes involved in anthocyanin biosynthesis pathway in Carthamus species with contrasting seed coat colors

Soraya Karami 1,✉, Mohammad Reza Sabzalian 2, Tayebeh Basaki 3, Fariba Ghaderi 4, Kiarash Jamshidi Goharrizi 5
PMCID: PMC10073368  PMID: 37033761

Abstract

The morphological structure of seed such as coat color can be considered as effective parameters in the evaluation of resistance to pests. The present study is aimed at achieving these goals: first, to determine the phylogenetic relationship of different species of safflower with different seed coat colors based on three candidate genes in the anthocyanin biosynthesis pathway that encode the early steps (PAL: phenylalanine ammonia-lyase and CHS: chalcone synthase) and the final step (UFGT: flavonoid-3-O-glucosyltransferase); second, based on our previous study on the absence of cyanidin-3-O-glucoside (Cyd-3-glu) in white/brown-seeded genotypes, it can be determined whether the lack of production is related to the absence of genes or the lack of expression. In general, the detection of Cyd-3-glu upstream compounds in all studied safflower genotypes, regardless of the color of the seed coat, can be interpreted as the expression of genes responsible for the synthesis of these compounds in the anthocyanin synthesis pathway. In addition, these findings indicated that the accumulation pattern of the mentioned secondary metabolites could be varied in safflower genotypes according to the seed coat color pattern. Regarding the UFGT gene, the evidence showed that this gene is expressed in safflower genotypes with two different seed coat color patterns, but in each genotype the tendency to produce secondary metabolites is different. Consequently, it seems that UFGT may not only regulate Cyd-3-glu biosynthesis but also involved in biosynthesis of flavonol glucoside in black safflower. Additionally, UFGT only affected flavonol glycosides biosynthesis and had no effect on Cyd-3-glu biosynthesis in white- seeded safflower genotypes.

Keywords: Anthocyanin biosynthesis pathway, Safflower, Phylogenetic tree, Gene

Introduction

As one of the highly important oilseeds in the food industry, safflower as a plant of the family Asteraceae, genus Carthamus has specific characteristics in its oil such as suitable quality (14% oleic, 74% linoleic, 6.5% palmitic, and 2% stearic acid), light color, high iodine and particular favorable taste (Nykiforuk et al. 2012). In addition, safflower oil is widely used in pharmaceutical and medical applications due to its precursors for producing Eicosanoids (Simopoulos 2004). Despite the mentioned benefits, the global attention to safflower is less than other oilseed plants (Singh and Jauhar 2006). One reason for this neglect is the low grain yield of safflower, which is due to several pests, especially safflower fly (Acanthiophilus helianthi) (Smith et al. 2006). According to Sabzalian et al. (2010) this pest has resulted in a 30–70% reduction in safflower grain yield in Iran (Sabzalian et al. 2010). This pest is multi-host polyphagia, and because of their spawning ways, their chemical control is not very efficient (Ashri 1971; Weiss 2000); therefore, it seems that using resistant cultivars is the most suitable method to control it.

The mechanism of insect resistance in plants is divided into three groups: anti-xenosis, antibiosis, and tolerance (Painter 1951). The findings showed that morphological structure of seed coat including surface structure, hardness, and color of seed coat could be considered effective parameters in evaluating resistance to safflower fly, as the thick and colored seed coats in Carthamus spp. germplasm decreases the damage caused by the safflower fly; therefore, it is possible to use seed coat color and thickness as signs to recognize Carthamus spp. genotype with high resistance to the fly. So that, genotype A82 as a new safflower breeding line with specific characterization of the morphological structure of seed coat compared to its parents and other safflower species (Karami et al. 2017), was proposed as a desirable and better genotype for regions infected with the safflower fly. The black-seeded genotype A82 is a uniform and new line of safflower developed through interspecific hybridization of C111, as the female parent with white seed coat, and a black-seeded genotype of a wild safflower species (C. oxyacanthus) next to selfing and black crossing programs (Karami et al. 2017).

In addition to morphological structures, plants produce secondary metabolites such as terpenoids, quinones, alkaloids, glucosinolates, and flavonoids that negatively affect insects' development, reproduction, and survival (Karami et al. 2018). Also, some non-nourishing compounds in seeds, such as proteinase inhibitors, lectins, and alpha-amylase inhibitors, are effective in this type of mechanism. Therefore, another finding of our previous study (Karami et al. 2021) indicated that, in safflower, some flavonoid compounds of seed coat/seed could develop resistance to the safflower fly through antibiosis mechanism. There was a direct and inverse relationship in the concentration of the polyphenolic compound (including apigenin, ferulic acid, quercetin, and rutin) and cyanidin-3-O-glucoside (Cyd-3-glu) content with safflower fly resistance, respectively (Karami et al. 2021). However, based on our previous study, Cyd-3-glu was not observed in the seed coat extract of white-seeded genotypes such as C111 genotypes and also brown-seeded genotypes of C. oxyacanthus such as Azar; while Cyd-3-glu was found in black-seeded genotypes (A82, Lanatus, and Glaucus) (Karami et al. 2018, 2021).

As mentioned above, one of the special and distinctive features of genotype A82 compared to other cultivated and wild safflower genotypes are its seed coat color (glossy black) (Fig. 1). Anthocyanins belonging to the flavonoid family are water-soluble pigments that mostly take part in blue to red coloring. Several studies have been conducted recently on anthocyanins, which have shown their role in the typical coloring of plants. For instance, anthocyanins take part in determining skin color (Fang et al. 2019; Zhang et al. 2020), flower and petal blotches (Jiang et al. 2020; Zhang et al. 2020), and fruit (Fan et al. 2020; Huang et al. 2020). Moreover, studies have shown that the anthocyanins like Cyd-3-glu and peonidin-3-glucoside in rice (Hu et al. 2003; Zhang et al. 2006; Jang and Xu 2009) and Cyd-3-glu, delphinidin-3-O-glucoside, petunidin-3-O-glucoside, and pelargonidin-3-O-glucoside in soybean (Kovinich et al. 2010) are in charge of the black color of the grain.

Fig. 1.

Fig. 1

Open in a new tab

Black-seeded genotype A82 as a novel breeding line (a) White-seeded genotypes of C. tinctorius and C. palaestinus (C111 and Palaestinus, respectively) (b) Black-seeded genotypes of C. glaucus and C. lanatus (Glaucus and Lanatus, respectively) (c) White-seeded genotype of C. turkestanicus (Turkestanicus) (d) Brown-seeded genotypes of C. oxyacanthus (Azar and Sarab)

In the case of Cartamos, several genes have been identified in safflower flowers that participate in the first steps of anthocyanin biosynthesis and the general biosynthesis of flavonoids like PAL (phenylalanine ammonia-lyase) and CHS (chalcone synthase). (Guo et al. 2016; Qiang et al. 2020) however, the gene that encodes the final stage in anthocyanin biosynthesis including UDP-glucose: flavonoid-3-O-glucosyltransferase (UF3GT), is mostly not known and there are a few reports on them. Only three UGTs (CtUGT3, CtUGT16, and CtUGT25) were explored which had a distinctively key roles in the biosynthesis with diverse functional specifications in two safflower lines with white and yellow flowers (Guo et al. 2016). However, so far, no similar study has been performed on the function of anthocyanins or other polyphenolic compounds in seed coat color formation. From a biochemical point of view, we showed in our previous study that Cyd-3-glu might be in charge of the black color of the seed coat, and the absence of Cyd-3-glu in other genotypes with white and brown seed coat color is due to the absence of genes expression responsible for the synthesis of this compound (Karami et al. 2018).

Therefore, due to the role of anthocyanins in creating the dark color of the safflower seed coat and the importance of coat color of seed in developing safflower fly resistance, genetic and molecular investigation of seed coat color in safflower is highly important. So, this work aimed to achieve these goals: first, to indicate the phylogenetic relationship of diverse species of safflower with different seed coat colors based on the three candidate genes in the anthocyanin biosynthesis pathway that encode the early steps (PAL and CHS) and the final step (UFGT); second, based on our previous study on the lack of Cyd-3-glu in white/brown-seeded genotypes, it can be determined whether the lack of production is related to lack of gene expression.

Materials and methods

Plant material, DNA extraction, and primer design

Plant material included 8 genotypes of Carthamus spp. with different seed coat colors (white, brown, black). Six genotypes of five wild safflower species (two from C. oxyacanthus, one from C. lanatus, one from C. glaucus, one from C. palaestinus, and one from C. turkestanicus), and one genotype of cultivated safflower (C. tinctorius). In addition, a new breeding line of safflower (A82) was used as plant material. The new line is a new and uniform line developed through interspecific hybridization (C. tinctorius × C. oxyacanthus). Figure 1 illustrates more details of the genotypes. To determine the plants’ types, M.R. Sabzalian (Associate professor) helped us, and all the samples were confirmed using the Herbarium in Isfahan University of Technology.

In a growth chamber for DNA extraction, BC1F12 and F12 generation seeds of genotype A82 and other genotypes mentioned above were cultivated at 25/22 °C in a 16/8 h (day/night) cycle, respectively. The samples were collected when the plants developed six leaves. Each sample contained 80–110 mg of fresh tissues (leaves), flash-frozen using liquid nitrogen. In addition, the samples were lyophilized before DNA extraction was done using a DNeasy 96 plant kit following the producer’s guidelines (Qiagen, Hilden, Germany). Afterward, a Quant-it PicoGreen dsDNA assay kit was used to quantify genomic DNA (ThermoFisher Scientific, Waltham, MA). In addition, 0.8% agarose gel was used to assess the samples visually.

Sequence files of published PAL and CHS genes in safflower and other members of Asteraceae family such as Rudbeckia hirta, Cynara scolymus, Helianthus annuus, Gynara bicolor, and Silybum marianum were downloaded from NCBI (the National Center for Biotechnology Information) and afterward utilized to develop the isolating primer pair for CtPAL and CtCHS using Primer3 (Table 1). Similarly, the coding sequences of the UFGT gene in Stevia rebaudiana, Gynura bicolor, and Ixeris dentate were taken into account to develop the isolating primers for CtUFGT (Table 1).

Table 1.

Nucleotide sequences of designed primers utilized for isolation of PAL, CHS and UFGT genes in Carthamus spp

Primer name Primer nucleotide sequence Exp. product size (bp) Obs. Product size (bp)
PAL-F CTCCTCCAGGGTTACTCC 872 872
PAL-R CCTTTGAACCCGTAATCC
CHS-F AAACGCTTCATGATGTACCA 599 599
CHS-R GCCGACTTCTTCCTCATCTC
UFGT-F CTTCAGGAGTCATCTGGAACTC 622 622
UFGT-R TCTCCTTCTTCATCCACCATAAC
Open in a new tab

PAL phenylalanine ammonia-lyase, CHS chalcone synthase, UFGT flavonoid-3-O-glucosyltransferase, F forward, R reverse primers

PCR optimization and Validation of candidate genes

PCR amplifications were conducted on 15 μL T- Gradient thermocycler (Biometra) including 1.5 mM MgCl2, 1 × PCR buffer, 0.8 mM dNTPs, 5 pmol each reverse and forward primers, 0.01 mg/mL purified bovine serum albumin, 1U Taq DNA polymerase, and 30 ng genomic DNA. The amplification program contained a starting denaturation step at 95 °C for 5 min along with denaturation at 92 °C for 1 min (40 cycles), primer annealing at 51–61 °C (based on the temperature at the melting of primer combination for each gene) for 45 s, extension at 72 °C for 1 min and a final extension step at 72 °C for 5 min. Every PCR contained a negative control with no DNA template. Products of PCR were resolved on 1% agarose gel verify amplicon specificity. With amplification confirmed, EXoSAP was used to purify the rest of PCR products, according to Mendes et al. (2014). Afterward, the PCR products were sequenced at least twice by the Macrogen company, Seoul, South Korea, using the Sanger sequencing method.

Sequence alignment and phylogenetic analysis

Multiple alignments (BLAST) and initial sequence analysis were carried out using MAFFT v. 7 (Katoh and Standley 2013), the studied sequences, and others available in databases. Nucleotide sequences of partial coding regions of the candidate genes for each species were deposited in GenBank.

The phylogenetic relationships of the candidate genes in the present study from different species of Carthamus and other members of the Asteraceae family were inferred using a phylogenetic tree created through the maximum likelihood method (ML). One thousand replicates were used in Bootstrap analysis to examine the statistics of branches in the ML method.

Chromatographic separation of polyphenolic components and cyanidin-3-O-glucoside content

To measure the major polyphenolic compounds (phenolic and flavonoid) and Cyd-3-glu content of seed coat extracts, the plants were harvested at the maturity stage and then seeds were de-coated. Since the classification of all of the species in the present work, except for Turkestanicus (C. turkestanicus) and Sarab genotypes (C. oxyacanthus), was previously done (Karami et al. 2018) based on data from Cyd-3-glu content and chromatographic separation of phenolic and flavonoid components, the measurements were performed based on these criteria for two Turkestanicus and Sarab genotypes as well as genotype A82 (BC1F12 generation). Afterward, charts were re-drawn based on these criteria for all species (genotypes) (Fig. 2). Cyd-3-glu quantification and chromatographic separation of polyphenolic components were determined using the pH-differential procedure (Giusti and Wrolstad 2001) and HPLC analysis (Karami et al. 2018), respectively.

Fig. 2.

Fig. 2

Open in a new tab

Major polyphenolic compounds (phenolic and flavonoid) and Cyanidin-3-O-glucoside (Cyd-3-glu) content of seed coat extracts in eight Carthamus genotypes based on HPLC analysis and the pH-differential procedure, respectively. For Major polyphenolic compounds and Cyanidin-3-O-glucoside content, the values are expressed in mg/100 g of sample dry weight and mg cyd-3-glu/g of sample dry weight, respectively

Results

Characterization and phylogenetic analysis of the candidate genes

Generally, to amplify and screen possible PAL, CHS, and UFGT genes in all the aforementioned Carthamus species, specific primer pairs were used. Target genes amplification with specific primers led to a specific amplicon with a specific size (Table 1). Complete details of the candidate genes and phylogenetic relationships related to Carthamus species are described below:

PAL gene

The PAL primers amplified a specific fragment size of ~ 872 bp in all the genotypes employed. In addition, based on the phylogenetic analysis, two different clades were determined based on the partial sequence of the PAL gene. The cultivated species (e.g., C. tinctorius), C. palaestinus, C. oxyacanthus, and genotype A82 formed a single group (first clade). In this cluster, genotype A82 indicated the highest similarity with C.tinctorius. In addition, C. palaestinus showed the highest similarity with C. tinctorius compared to other wild species. In contrast, wild species of C.glaucus, C.lanatus, and C. turkestanicus formed the other distinct cluster (second clade) (Fig. 3A).

Fig. 3.

Fig. 3

Open in a new tab

Phylogenetic tree drawn based on the PAL (A) and CHS (B) gene for six species of safflower and interspecific hybrid A82 by ML method. The numbers above each branch indicate the bootstrap value of 10,000 replications

CHS gene

Regarding the CHS gene, the specific primer of CHS amplified an expected fragment size of ~ 599 bp in the genotypes. Similar to PAL phylogenetic tree, C. palaestinus for the CHS gene indicated the highest similarity with C. tinctorius compared to the rest of the wild species. Still, C. oxyacanthus was in the same clade as a distinct sub-cluster located with C. tinctorius. In addition, C. turkestanicus and C. lanatus again belonged to the same branch and clustered with C. glaucus in the same clade as supported by a 100% bootstrap value (Fig. 3B).

Afterward, the phylogenic tree was formed based on the current sequences and the coding sequences of the PAL and CHS genes in members of the Asteraceae family (Fig. 4). The results showed that, exactly all the PAL genes of Carthamus species in the present study (i.e., MG816782.1: A82; MG816785.1: C111; MG816786.1: palaestinus; MG816783.1: azar; MG816784.1: sarab; MG816788.1: glaucus; MG816789.1: turkestanicus; MG816790.1: lanatus) and the previously established accession of C. tinctorius (JN998609.2, pal) belonged to the same cluster. On the other hand, the accession AM418588.1 of Cynara scolymus (PAL2) was placed in the same Carthamus cluster but in the neighboring and separated branch. The rest of the PAL genes belonging to the Asteraceae family were grouped in separate clusters, sub-clusters, or separate clades (Fig. 4A).

Fig. 4.

Fig. 4

Open in a new tab

Maximum likelihood-based phylogenetic tree obtained from the deduced nucleotide sequences of PAL (A), CHS (B) and UFGT (C) gene families of Carthamus species used in current study followed by some PAL (A), CHS (B) and UFGT (C) gene families previously established in GenBank belonging to the Asteraceae family (http://www.ncbi.nlm.nih.gov/). Note nucleotide sequences of the current study were enclosed by Purple outlines. Length of branches represented the relative sequence differences: the shorter branch lengths, the more similar sequences; Numbers above the branches indicate the bootstrap values

Similar to PAL phylogenetic tree, all the CHS genes of Carthamus species (i.e. MG759480.1: A82; MG759483.1: C111; MG759484.1: Palaestinus; MG759481.1: azar; MG759482.1: sarab; MG759485.1: glaucus; MG816781.1: turkestanicus; MG816780.1: lanatus) together to two accessions LC128420.1 (CHS1), KY471385.1 (CHS1) and JQ425086.1 (CHS) of C. tinctorius were placed in the same cluster. In addition, one accession of Centaurea jacea (EF112474.1, CHS1) was placed in the same Carthamus cluster but in the neighboring and separated branch. All of the above-mentioned accession numbers were grouped with the accession of LC128421.1 (C. tinctorius, CHS2) in the same clade. Other CHS genes belonging to the Asteraceae family were placed in the separate clade (Fig. 4B).

UFGT gene

Amplification and screening of the UFGT gene were carried out only on two genotypes, namely A82 and C111 belonging to the C. tinctorius with a distinct pattern of seed coat color. The results showed that the specific primer of UFGT improved an expected fragment size of ~ 622 bp in the two mentioned genotypes. Evaluation of the phylogenetic tree using the current sequences and the available coding sequences of the UFGT gene in the Asteraceae family showed that the black-seeded genotype A82 (MG816791.1) and its crossable relative, C. tinctorius var. C111 (MG816790.1) together with XM_023915054.2 (Lactuca sativa), XM_025137648.1 (Cynara cardunculus), XM_022150639.2 (Helianthus annus) and XM_035983501.1 (H. annus) with high value of bootstrap of 91% were grouped in the same clade. The rest of the UFGT genes were placed in the next separated clades (Fig. 4C).

Polyphenolic components and cyanidin-3-O-glucoside content

Chromatographic separation of polyphenolic components (i.e., chlorogenic, caffeic, p-coumaric, ferulic acids, rutin, quercetin, and apigenin) was carried out by HPLC. All of the above-mentioned components were identified in the selected genotypes (Fig. 2). Regarding Cyd-3-glu, pH-differential procedure with two buffer systems revealed that Cyd-3-glu was not identified either in seed coat extracts of cultivated (C. tinctorius) and wild species with white and brown color of seed coat (i.e., C. palaestinus, C. turkestanicus, and C. oxyacanthus), while it was detected in seed coat extracts of black-seeded genotypes (genotype A82 and genotypes belonging to wild species of C. lanatus and C. glaucus) (Fig. 2).

Discussion

Considering the role of anthocyanins in dark color of the safflower seed coat and developing resistance to the safflower fly based on previous studies, therefore from a plant breeding and nutritional point of view, identifying the phylogenetic relationship of diverse species of safflower with different seed coat color based on the three candidate genes (PAL, CHS and UFGT) in the anthocyanin biosynthesis pathway, which encode the initial and final steps, is of great importance.

The results of phylogenetic tree analyses and branch length of the tree suggest that diversification in the PAL and CHS encoding genes sequence in the genus Carthamus could reflect taxonomic relationships (Fig. 3). The genus Carthamus is divided into four sections: Sect. 1 features three annual species, including C. palaestinus, C. tinctorius, and C. oxyacanthus (Ashri 1971). Four species are classified in Sect. 2, namely C. glaucus, C. dentatus, C. alexandrines, and C. boissieri. Section 3 only represents one species (C. lanatus). Eventually, Sect. 4 introduces two species, including C. turkestanicus and C. baeticus. Study of chloroplast DNAs indicated that C. palaestinus and C. oxycanthus are the wild ancestors of C. tinctorius (cultivated safflower) (Sehgal et al. 2008). However, a phylogenetic study based on molecular analysis and the sequencing of seven genes revealed cultivated safflower strongly related to C. palaestinus (Chapman and Burke 2007; Majidi and Zadhoush 2014). Here, C. palaestinus was grouped with C. tinctorius; however, C. oxyacanthus differed from cultivated safflower (based on the PAL and CHS phylogenetic tree). The findings confirm that C. palaestinus has the closest relationship with C. tinctorius and confirm changes in the PAL and CHS genes sequence with taxonomic classification. In addition, as noted earlier, the genotype A82 is a novel breeding line with black seed coat developed through interspecific hybridization of C111 (C. tinctorius), as the female parent, and a black-seeded genotype of a wild safflower species (C. oxyacanthus). So, a grouping of genotype A82 with C. tinctorius var. C111 may suggest that PAL and CHS genes were transferred from C. tinctorius to genotype A82.

Regarding other wild species in the second clade, evidence suggests that C. turkestanicus is obtained from a cross between C. lanatus and C. glaucus (McPherson et al. 2004); therefore grouping of C. lanatus and C. turkestanicus is primarily because of ancestral relationship between the two species and may suggest that PAL gene was transferred from C. lanatus to C. turkestanicus. In addition, the length of branches in the phylogenetic tree indicates sequences that are relatively different so that with shorter branches, the sequence becomes more identical (Shafiei-Koij et al. 2020). Therefore, based on the branch length, it can be assumed that the species belonging to the first and the second clade underwent rapid diversification in the PAL and CHS genes, respectively.

On the other hand, the phylogenic tree formed based on the current sequences and the coding sequences of the PAL, CHS and UFGT genes in members of the Asteraceae family confirms that, despite the differences between the sequences, there is a high level of homology.

Afterwards, to better understand the result obtained (phylogenetic tree) and determine the role of the candidate genes in the anthocyanin synthesis pathway, the evaluation of polyphenolic compounds (flavonoids) content that are products of the candidates genes was considered. Plant pigments such as flavonoids in higher plants give a variety of colors from pale-yellow to blue (Tsao et al. 2003). For example, anthocyanins, chalcones, aurones, and some flavonols act as the primary pigments; in contrast, flavones and flavonols function as copigment substances. On the other hand, anthocyanins such as pelargonidin, delphinidin, petunidin, malvidin, and cyanidin confer different colors to plants from orange, purple, blue to black. At the same time, flavones and flavonols like apigenin, quercetin, rutin, kaempferol, and luteolin are colorless or extremely pale yellow (Ferrer et al. 2008). In plants, the flavonoid biosynthesis pathway was widely evaluated, and most of the genes and enzymes in this pathway have been identified (Ferrer et al. 2008).

The initial key enzyme in the biosynthesis pathway of phenylpropanoid is the phenylalanine ammonia-lyase (PAL) (Liu et al. 2006); however, because of the key role of PAL at the point of the branch of phenylpropanoid derivative metabolism, the enzyme is the key in the biosynthesis of flavonoids/anthocyanins (Capell and Christou 2004). The sequences of PAL are typically encoded using a small gene family in plants for various reasons (Dehghan et al. 2014). As mentioned above, the PAL region isolated of Carthamus species in the present study was grouped with one previously established accession of C. tinctorius (JN998609.2), which showed high similarity (96%) to PAL (PAL1) of Cynara scolymus according to the study of Dehghan et al. (2014). Based on De Paolis et al. (2008), the PAL1 gene can have a role in the color pigments synthesis through the flavonoid metabolic pathway in Cynara scolymus, and the PAL1 was strongly overexpressed in the external bracts of C. scolymus with violet-greenish color (De Paolis et al. 2008). In the present study, the grouping of colored and non-colored genotypes of Carthamus spp with PAL1 could confirm the above hypothesis about the role of the PAL enzyme in the pathway of anthocyanin biosynthesis.

In reconfirming the above mentioned hypothesis and looking at the division of phenylpropanoids compounds, it was found that phenylpropanoids are phenolic compounds that can be divided into two categories: (1): derivatives of benzoic acid such as gallic acid, and (2): derivatives of cinnamic acid-like coumaric, caffeic, chlorogenic, and ferulic acids (Fig. 5). In the present study, the four derivatives of cinnamic acid were evaluated, and all of these components were detected in Carthamus species; however, the amount of identified phenolic compounds was almost similar (except for ferulic acid) in selected species (Fig. 2). Therefore, this stipulates that PAL may modulate the biosynthesis of phenylpropanoids and act as a primary enzyme in the anthocyanin biosynthesis pathway.

Fig. 5.

Fig. 5

Open in a new tab

The proposed anthocyanin biosynthesis pathway in seed coat color of safflower (genotype A82)

As to the CHS gene, CHS plays a crucial role in the phenylpropanoid cascade, the starting point of the flavonoid pathway (Fig. 5). The CHS region isolated from the Carthamus species in the present study was grouped with previously established accessions of C. tinctorius, i.e., LC128420.1 (CHS1) and JQ425086.1 (CHS), introduced by Shinozaki et al. (2016) and Francini et al. (2008), respectively (Francini et al. 2008; Shinozaki et al. 2016). Besides, Shinozaki et al. (2016) published for the first time a report of cloning and functional analysis using the three chalcone synthase genes, i.e., CHS1, CHS2, and CHS3 from C. tinctorius, and found that the CtCHS genes encode a standard CHS in the flavonoid and/or QCG pathway in C. tinctorius. CCGs or quinochalcone C-glucosides are pigments that only some of them such as safflor yellow B, hydroxysafflor yellow A, precarthamin, anhydrosafflor yellow B, and carthamin and other related compounds such as naringenin and p-coumaric acid, are especially accumulated in the fresh flowers. In addition, it was shown that all the CtCHSs could recognize products of the PAL gene such as caffeoyl-CoA, p-coumaroyl-CoA, feruloyl-CoA, and sinapoyl-CoA as initial substrates to form chalcone (Shinozaki et al. 2016). Additionally, in the anthocyanin biosynthesis, chalcone (naringenin chalcone) is a direct product of the CHS catalyzed reaction, and naringenin as the first stable intermediate product is transformed from chalcone by the action of CHS and chalcone isomerase (CHI) in the pathway of flavonoid synthesis, which is the formost downstream flavonoids. Using flavone synthase (FNS), naringenin is converted into the flavone apigenin (FNS). In addition, it is possible to convert naringenin into dihydro flavonols such as dihydroquercetin using flavanone-3-hydroxylase (F3H) activity (Fig. 5). This reaction is a major step in the biosynthesis of flavonols and anthocyanidins. So, it is possible to reduce dihydroflavonols to respective flavonols (such as dihydroquercetin to quercetin) by flavonol synthase (FLS) enzyme and finally glycosylated, such as quercetin to quercetin-3-O-rutinoside (rutin) by UFGT activity (Fig. 5) (Ueyama et al. 2002).

In evaluating the above mentioned pathway, in our previus study (Karami et al. 2018) and the present study chalcone compound was not detected in the studied   Carthamus spp. It seems that its immediate conversion to downstream products such as apigenin, quercetin, and rutin could justify this. In agreement with this hypothesis, three compounds, i.e., apigenin, quercetin, and rutin, were detected in all evaluated Carthamus species by HPLC analysis of seed coat extracts (Fig. 2). Also, it is indicated that the most abundant flavonoid was rutin in the evaluated Carthamus spp, however brown and white-seeded genotypes (i.e., genotype C111, Azar, Sarab, Palaestinus, and Turkestanicus belonged to the C. tinctorius, and C. palaestinus, C. oxyacanthus, and C. turkestanicus, respectively) had 2.2- to 16.3-fold more rutin compared to that recorded in black-seeded ones (i.e., genotype A82, Lanatus, and Glaucus belonged to the C. tinctorius, C. lanatus, and C. glaucus, respectively) (Fig. 2). As a result, similar to the PAL gene, it can be assumed that regardless of the seed coat color, CHS gene as early biosynthetic genes in phenylpropanoid cascade/anthocyanin biosynthesis pathway is expressed in all studied Carthamus species.

Regarding the third candidate gene studied (UFGT), based on phylogenetic and grouping analysis, it can be stated that the detected sequences in two safflower genotypes with two different patterns of seed coat color (i.e., black-seeded genotype A82 and white-seeded genotype C111) belong to the UFGT gene as the late gene in anthocyanin biosynthesis pathway (Fig. 4C). The presence of the UFGT gene was previously reported by Guo et al. (2016) in the safflower plant (Guo et al. 2016). Three CtUGTs genes were identified as CtUGT3 (accession number: KT947113.1), CtUGT16 (accession number: KT947114.1), and CtUGT25 (accession number: KT947116.1), which affected flavonoids/flavonol biosynthesis by providing different functional characterization in yellow and white-flowered safflower genotypes. In addition to phylogenetic and grouping analysis, the detection of the Cyd-3-glu compound (Fig. 2) could confirm the expression of this gene (UFGT) in the seed coat of black safflower and the possible function of UFGT in anthocyanin biosynthesis in this tissue. These findings are consistent with reports by Lee et al. (2009) and Kovinich et al. (2010), in which black soybean exclusively collects high volumes of anthocyanins as 3-O-glucosides in the seed coat. In addition, Kovinich et al. (2010) showed that UFGT (UGT78K1) is involved in the biosynthesis of anthocyanin and flavonol glycoside by catalyzing the galactose transfer to the 3-position of cyanidin or delphinidin in vivo (Kovinich et al. 2010).

One of the other challenges of the present study was whether the non-detection of Cyd-3-glu in white-seeded safflower genotypes (especially C111) compared to black-seeded genotypes (especially A82) (Fig. 2) is because of the lack of genes synthesizing this compound in the pathway of anthocyanin synthesis or due to the lack of related genes expression. Two hypotheses can probably be put forward to explain this finding. First: undetected Cyd-3-glu in genotype C111, despite the amplification of the UFGT partial coding sequences and the partial coding sequences of upstream genes (especially PAL and CHS gene) in the anthocyanin synthesis pathway, is due to the lack of UFGT gene expression. However, proof of this hypothesis requires further investigation, such as gene expression studies by suppressing the UFGT gene.

The second hypothesis, which we believe is more plausible, is that UFGT gene expression occurs in both black and white-seeded genotypes (A82 and C111, respectively) because the UFGT gene can recognize multiple compounds as starter substrates for the synthesis of different secondary metabolites in different branches of the anthocyanin biosynthesis pathway. In other words, in the anthocyanin biosynthesis pathway, the transfer of glucose is catalyzed by UFGT so that corresponding anthocyanins are formed from uridine diphosphate (UDP)-glucose to the 3-position of anthocyanidins. However, some UFGTs that catalyze the transfer of glucose to the 3-position of anthocyanidins also catalyze glucosylate flavonol, dihydroflavonol, flavone, isoflavone (Owens and McIntosh 2009), flavanone, and coumestan substrates in vitro (Modolo et al. 2007). For example, it is possible to reduce dihydroflavonols such as dihydroquercetin to respective flavonols using FLS (flavonol synthase) enzyme and finally glycosylated, such as the conversion of quercetin (flavonol) to rutin (flavonol glycoside), by flavonoid-3-O-glucosyltransferase (UFGT) activity (Fig. 5) (Ueyama et al. 2002). In the present study, two flavonol compounds, quercetin and quercetin-3-O-rutinoside (rutin), were observed in both black and white-seeded genotypes (Fig. 2). In addition, the amount of rutin in white-seeded genotypes was 2.2- to 16.3-fold higher than black-seeded ones (Fig. 2). Therefore, according to the rutin biosynthesis pathway, it may be concluded that, firstly, the UFGT gene is expressed in C111 (white-seeded genotype). Second, in the white genotype mentioned, a higher amount of rutin (than quercetin) may be considered a stronger tendency of the UFGT enzyme to synthesize flavonols glycoside than the corresponding anthocyanins, Cyd-3-glu. On the other hand, in the black-seeded genotype (A82), the detection of three compounds, quercetin-3-O-rutinoside, quercetin, and cyanidin-3-O-glucoside, could be related to the UFGT gene' tendency to synthesize anthocyanin and flavonol glycoside. These findings were consistent with Kovinich et al. (2010) reports that UF3GT is involved in the anthocyanin and flavonol glycoside biosynthesis. Besides, it is theoretically assumed that CHI increases rutin production (Kovinich et al. 2010); however, Najid et al. (2016) proved that other enzymes such as FLS and UFGT at downstream branches of the rutin pathway might be more active in rutin production than CHI (Najid et al. 2016). Guo et al. (2016) also showed that UFGTs genes play a role in the flavonol glucosides biosynthesis process in safflower flowers. As, CtUGT3 and CtUGT25 showed a positive correlation with the production of kaempferol-3-O-β-D-glucoside in yellow-flowered safflower, where there was a positive relationship to quercetin-3-O-β-D-glucoside production in white-flowered safflower (Guo et al. 2016).

Conclusion

In general, the detection of Cyd-3-glu upstream compounds (i.e., chlorogenic, caffeic, ferulic, and p-coumaric acids along with apigenin, quercetin, and rutin) in all studied safflower genotypes, regardless of the color of the seed coat, can be interpreted as the expression of genes responsible for the synthesis of these compounds in the anthocyanin synthesis pathway. Moreover, the findings mean that the accumulation pattern of the mentioned secondary metabolites can be varied in safflower genotypes according to the seed coat color pattern.

Regarding the UFGT gene as the gene responsible for Cyd-3-glu encoding and production of the corresponding black seed coat color in safflower, the evidence showed that this gene is possibly expressed in safflower genotypes with two different seed coat color patterns. However, in each genotype, its tendency to produce secondary metabolites is possibly different. So it seems that UFGT may not only regulate Cyd-3-glu biosynthesis but can also be involved in biosynthesizing flavonol glucoside in black-seeded safflower. Additionally, it seems that UFGT only influences flavonol glycosides biosynthesis, without any effect on Cyd-3-glu biosynthesis in white-seeded safflower genotypes. In short, the findings indicated that the UFGT may have diverse functional specifications in flavonoid/anthocyanin biosynthesis of diverse safflower lines. There is a need for further studies to understand these genes’ physiological roles.

Acknowledgements

We express our gratitude to the anonymous reviewers for helpful comments to improve the manuscript.

Author contributions

SK and MRS contributed to the study conception and design. Experimental researches were performed by SK and TB. Data were analyzed by SK, FGh and KJG. The first draft of the manuscript was written by SK and the initial and final manuscript version were edited by KJG, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors have no financial or proprietary interests in any material discussed in this article.

Declarations

Conflict of interest

The authors have no relevant financial or nonfinancial interests to disclose. The authors have no competing interests to declare that are relevant to the content of this article. All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Soraya Karami, Email: soraya.karami@pnu.ac.ir.

Mohammad Reza Sabzalian, Email: sabzalian@cc.iut.ac.ir.

Tayebeh Basaki, Email: tbasaki2@pnu.ac.ir.

Fariba Ghaderi, Email: f.ghaderi@yu.ac.ir.

Kiarash Jamshidi Goharrizi, Email: jamshidi_kiarash@yahoo.com.

References

  1. Ashri A. Evaluation of the world collection of safflower, Carthamus tinctorius LI reaction to several diseases and associations with morphological characters in Israel 1. Crop Sci. 1971;11(2):253–257. doi: 10.2135/cropsci1971.0011183X001100020026x. [DOI] [Google Scholar]
  2. Capell T, Christou P. Progress in plant metabolic engineering. Curr Opin Biotechnol. 2004;15:148–154. doi: 10.1016/j.copbio.2004.01.009. [DOI] [PubMed] [Google Scholar]
  3. Chapman MA, Burke JM. DNA sequence diversity and the origin of cultivated safflower. BMC Plant Biol. 2007;7:60–68. doi: 10.1186/1471-2229-7-60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. De Paolis A, Pignone D, Morgese A, Sonnante G. Characterization and differential expression analysis of artichoke phenylalanine ammonia-lyase-coding sequences. Physiol Plant. 2008;132(1):33–43. doi: 10.1111/j.1399-3054.2007.00996.x. [DOI] [PubMed] [Google Scholar]
  5. Dehghan S, Sadeghi M, Pöppel A, Fischer R, Lakes-Harlan R, Kavousi HR, Rahnamaeian M. 2014. Differential inductions of phenylalanine ammonia-lyase and chalcone synthase during wounding, salicylic acid treatment, and salinity stress in safflower, Carthamus tinctorius. Biosci Rep. [DOI] [PMC free article] [PubMed]
  6. Fan R, Sun Q, Zeng J, Zhang X. Contribution of anthocyanin pathways to fruit flesh coloration in pitayas. BMC Plant Biol. 2020;20:361. doi: 10.1186/s12870-020-02566-2. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  7. Fang H, Dong Y, Yue X, Chen X, He N, Hu J, Chen X. MdCOL4 interaction mediates crosstalk between UV-B and high temperature to control fruit coloration in apple. Plant Cell Physiol. 2019;60:1055–1066. doi: 10.1093/pcp/pcz023. [DOI] [PubMed] [Google Scholar]
  8. Ferrer JL, Austin MB, JrC S, Noelb JP. Structure and function of enzymes involved in the biosynthesis of phenylpropanoids. Plant Physiol Biochem. 2008;46:356–370. doi: 10.1016/j.plaphy.2007.12.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Francini A, Nali C, Pellegrini E, Lorenzini G. Characterization and isolation of some genes of the shikimate pathway in sensitive and resistant Centaurea jacea plants after ozone exposure. Environ Pollut. 2008;151(2):272–279. doi: 10.1016/j.envpol.2007.07.007. [DOI] [PubMed] [Google Scholar]
  10. Giusti MM, Wrolstad RE. Anthocyanins. Characterization and measurement with UV-visible spectroscopy. Curr Protoc Food Anal Chem. 2001;1:1–13. [Google Scholar]
  11. Guo DD, Liu F, Tu YH, He BX, Gao Y, Guo ML. Expression patterns of three UGT genes in different chemotype safflower lines and under MeJA stimulus revealed their potential role in flavonoid biosynthesis. PLoS ONE. 2016;11(7):e0158159. doi: 10.1371/journal.pone.0158159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hu C, Zawistowski J, Ling W, Kitts DD. Black rice (Oryza sativa L. indica) pigmented fraction suppresses both reactive oxygen species and nitric oxide in chemical and biological model systems. J Agric Food Chem. 2003;51:5271–5277. doi: 10.1021/jf034466n. [DOI] [PubMed] [Google Scholar]
  13. Huang G, Zeng Y, Wei L, Yao Y, Dai J, Liu G, Gui Z. Comparative transcriptome analysis of mulberry reveals anthocyanin biosynthesis mechanisms in black (Morus atropurpurea Roxb.) and white (Morus alba L.) fruit genotypes. BMC Plant Biol. 2020;20:279. doi: 10.1186/s12870-020-02486-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Jang S, Xu Z. Lipophilic and hydrophilic antioxidants and their antioxidant activities in purple rice bran. J Agric Food Chem. 2009;57:858–862. doi: 10.1021/jf803113c. [DOI] [PubMed] [Google Scholar]
  15. Jiang T, Zhang M, Wen C, Xie X, Tian W, Wen S, Liu L. Integrated metabolomic and transcriptomic analysis of the anthocyanin regulatory networks in Salvia miltiorrhiza Bge. flowers. BMC Plant Biol. 2020 doi: 10.1186/s12870-020-02553-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Katoh S, Standley MR. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol. 2013;30(4):772–780. doi: 10.1093/molbev/mst010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Karami S, Sabzalian MR, Rahimmalek M, Saeidi G, Ghasemi S. Interaction of seed coat color and seed hardness: an effective relationship which can be exploited to enhance resistance to the safflower fly (Acanthiophilus helianthi) in Carthamus spp. Crop Protect. 2017;98:267–275. doi: 10.1016/j.cropro.2017.04.006. [DOI] [Google Scholar]
  18. Karami S, Sabzalian MR, Rahimmalek M. Seed poly phenolic profile, antioxidative activity, and fatty acids compo sition of wild and cultivated Carthamus species. Chem Biodivers. 2018;15:e1700562. doi: 10.1002/cbdv.201700562. [DOI] [PubMed] [Google Scholar]
  19. Karami S, Basaki T, Mousavi Khaneghah A. The inhibitory effects of polyphenolic compounds on the damage caused by safflower fly (Acanthiophilus helianthi) in Carthamus spp. Qual Assur Saf Crop Foods. 2021;13(2):87–93. [Google Scholar]
  20. Kovinich N, Saleem A, Arnason JT, Miki B. Functional characterization of a UDP-glucose: flavonoid 3-O-glucosyltransferase from the seed coat of black soybean (Glycine max (L.) Merr.) Phytochemistry. 2010;71:1253–1263. doi: 10.1016/j.phytochem.2010.05.009. [DOI] [PubMed] [Google Scholar]
  21. Lee JH, Kang NS, Shin SO, Shin SH, Lim SG, Suh DY, Baek IY, Park KY, Ha TJ. Characterization of anthocyanins in the black soybean (Glycine max L.) by HPLC-DAD-ESI/MS analysis. Food Chem. 2009;112:226–231. doi: 10.1016/j.foodchem.2008.05.056. [DOI] [Google Scholar]
  22. Liu RR, Xu SH, Li JL, Hu YL, Lin ZP. Expression profile of a PAL gene from Astragalus membranaceus var. Mongholicus and its crucial role in flux into flavonoid biosynthesis. Plant Cell Rep. 2006;25:705–710. doi: 10.1007/s00299-005-0072-7. [DOI] [PubMed] [Google Scholar]
  23. Majidi MM, Zadhoush S. Molecular and morphological variation in a world-wide collection of safflower. Crop Sci. 2014;54(5):2109–2119. doi: 10.2135/cropsci2013.12.0850. [DOI] [Google Scholar]
  24. McPherson MA, Good AG, Topinka AKC, Hall LM. Theoretical hybridization potential of transgenic safflower (Carthamus tinctorius L.) with weedy relatives in the New World. Can J Plant Sci. 2004;84:923–934. doi: 10.4141/P03-150. [DOI] [Google Scholar]
  25. Mendes MD, Barroso JG, Oliveira MM, Trindade H. Identification and characterization of a second isogene encoding γ-terpinene synthase in Thymus caespititius. J Plant Physiol. 2014;171(12):1017–1027. doi: 10.1016/j.jplph.2014.04.001. [DOI] [PubMed] [Google Scholar]
  26. Modolo LV, Blount JW, Achnine L, Naoumkina MA, Wang X, Dixon RA. A functional genomics approach to (iso) flavonoid glycosylation in the model legume Medicago truncatula. Plant Mol Biol. 2007;64:499–518. doi: 10.1007/s11103-007-9167-6. [DOI] [PubMed] [Google Scholar]
  27. Najid NM, Zain CRCM, Zainal Z (2016) Analysis of the relationship between Chalcone Isomerase gene expression level and rutin production in Ficus deltoidea var. deltoidea and F. deltoidea var. angustifolia. In: AIP Conference Proceedings, AIP Publishing LLC 1748 (1)
  28. Nykiforuk CL, Shewmaker C, Harry I, Yurchenko OP, Zhang M, Reed C, Moloney MM. High level accumulation of gamma linolenic acid (C18: 3Δ6. 9, 12 cis) in transgenic safflower (Carthamus tinctorius) seeds. Transgenic Res. 2012;21(2):367–381. doi: 10.1007/s11248-011-9543-5. [DOI] [PubMed] [Google Scholar]
  29. Owens DK, McIntosh CA. Identification, recombinant expression, and biochemical characterization of a flavonol 3-O-glucosyltransferase clone from Citrus paradisi. Phytochemistry. 2009;70:1382–1391. doi: 10.1016/j.phytochem.2009.07.027. [DOI] [PubMed] [Google Scholar]
  30. Painter RH. Insect resistance in crop plants. LWW. 1951;72(6):481. [Google Scholar]
  31. Qiang T, Liu J, Dong Y, Ma Y, Zhang B, Wei X, Xiao P. Transcriptome sequencing and chemical analysis reveal the formation mechanism of white florets in Carthamus tinctorius L. Plants. 2020;9:847. doi: 10.3390/plants9070847. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Sabzalian MR, Saeidi G, Mirlohi A, Hatami B. Wild safflower species (Carthamus oxyacanthus): a possible source of resistance to the safflower fly (Acanthiophilus helianthi) Crop Protec. 2010;29(6):550–555. doi: 10.1016/j.cropro.2009.12.013. [DOI] [Google Scholar]
  33. Sehgal D, Rajpal VR, Raina SN. Chloroplast DNA diversity reveals the contribution of two wild species to the origin and evolution of diploid safflower (Carthamus tinctorius L.) Genome. 2008;51:638–643. doi: 10.1139/G08-049. [DOI] [PubMed] [Google Scholar]
  34. Shafiei-Koij F, Ravichandran S, Barthet VJ, Rodrigue N, Mirlohi A, Majidi MM, Cloutier S. Evolution of Carthamus species revealed through sequence analyses of the fad2 gene family. Physiol Mol Biol Plants. 2020;26(3):419–432. doi: 10.1007/s12298-019-00739-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Shinozaki J, Kenmoku H, Nihei K, Masuda K, Noji M, Konno K, Kazuma K. Cloning and functional analysis of three chalcone synthases from the flowers of safflowers Carthamus tinctorius. Nat Prod Commun. 2016;11(6):1934578X1601100621. [PubMed] [Google Scholar]
  36. Simopoulos AP. Omega-3 fatty acids and antioxidants in edible wild plants. Biol Res. 2004 doi: 10.4067/S0716-97602004000200013. [DOI] [PubMed] [Google Scholar]
  37. Singh RJ, Jauhar PP. Genetic resources, chromosome engineering, and crop improvement: cereals. CRC Press; 2006. [Google Scholar]
  38. Smith L, Hayat R, Cristofaro M, Tronci C, Tozlu G, Lecce F. Assessment of risk of attack to safflower by Ceratapion basicorne (Coleoptera: Apionidae), a prospective biological control agent of Centaurea solstitialis (Asteraceae) Biol Control. 2006;36(3):337–344. doi: 10.1016/j.biocontrol.2005.11.001. [DOI] [Google Scholar]
  39. Tsao R, Yang R, Young JC, Zhu H. Polyphenolic profiles in eight apple cultivars using high-performance liquid chromatography (HPLC) J Agric Food Chem. 2003;51:6347–6353. doi: 10.1021/jf0346298. [DOI] [PubMed] [Google Scholar]
  40. Ueyama Y, Suzuki KI, Fukuchi-Mizutani M, Fukui Y, Miyazaki K, Ohkawa H, Kusumi T, Tanaka Y. Molecular and biochemical characterization of torenia flavonoid 3′-hydroxylase and flavone synthase II and modification of flower color by modulating the expression of these genes. Plant Sci. 2002;163(2):253–263. doi: 10.1016/S0168-9452(02)00098-5. [DOI] [Google Scholar]
  41. Weiss EA. Oilseed crops. Oxford, UK: Blackwell Science; 2000. [Google Scholar]
  42. Zhang M, Guo B, Zhang R, Chi J, Wei Z, Xu Z, Zhang Y, Tang X. Separation, purification and identification of antioxidant composition in black rice. Agric Sci China. 2006;5:431–440. doi: 10.1016/S1671-2927(06)60073-4. [DOI] [Google Scholar]
  43. Zhang Q, Wang L, Liu Z, Zhao Z, Zhao J, Wang Z, Liu M. Transcriptome and metabolome profiling unveil the mechanisms of Ziziphus jujube. Mill peel coloration. Food Chem. 2020;312:125903. doi: 10.1016/j.foodchem.2019.125903. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Citations

  1. Dehghan S, Sadeghi M, Pöppel A, Fischer R, Lakes-Harlan R, Kavousi HR, Rahnamaeian M. 2014. Differential inductions of phenylalanine ammonia-lyase and chalcone synthase during wounding, salicylic acid treatment, and salinity stress in safflower, Carthamus tinctorius. Biosci Rep. [DOI] [PMC free article] [PubMed]

Articles from Physiology and Molecular Biology of Plants are provided here courtesy of Springer

RESOURCES