2019 年 69 巻 3 号 p. 536-544
The chrysanthemum (Chrysanthemum morifolium) is one of the most popular ornamental plants in the world. Genetic transformation is a promising tool for improving traits, editing genomes, and studying plant physiology. Promoters are vital components for efficient transformation, determining the level, location, and timing of transgene expression. The cauliflower mosaic virus (CaMV) 35S promoter is most frequently used in dicotyledonous plants but is less efficient in chrysanthemums than in tobacco or torenia plants. Previously, we used the parsley ubiquitin (PcUbi) promoter in chrysanthemums for the first time and analyzed its activity in transgenic calli. To expand the variety of constitutive promoters in chrysanthemums, we cloned the upstream region of the actin 2 (CmACT2) gene and compared its promoter activity with the 35S and PcUbi promoters in several organs, as well as its durability for long-term cultivation. The CmACT2 promoter has higher activity than the 35S promoter in calli but is less durable. The PcUbi promoter has the highest activity not only in calli but also in leaves, ray florets, and disk florets, and retains its activity after long-term cultivation. In conclusion, we have provided useful information and an additional type of promoter available for transgene expression in chrysanthemums.
Genetic transformation is an important method in plant breeding that is used to directly modify traits and to analyze gene function, and complements other research, including genetic and bioinformatics studies. Promoters are essential components of transgenes and determine the level, location, timing, and long-term durability of transgene expression. Efficient promoters which help to express transgenes have been established for the production of transformants, particularly in model plants such as Arabidopsis thaliana. However, the production of transgenic varieties has proved to be challenging for many ornamental plants, including Chrysanthemum morifolium (chrysanthemums).
Chrysanthemums are one of the most economically valuable flowering plants, particularly in east Asian countries. Given the economic importance of this plant, new traits are highly sought after, including flower color, flower shape, architectural variants, long-lasting freshness, and pathogen resistance. Most chrysanthemums are self-incompatible (Wang et al. 2014) and are primarily hexaploid (2n = 6× = 54), often with additional aneuploidy (Shibata and Kawata 1986). The chrysanthemum genome is enormous (12.4–24.8 Gbp, http://www.etnobiofic.cat/gsad_v2/); therefore, the exploitation of genome information for crossbreeding is difficult. The production of current commercial chrysanthemum variants involves classical crossbreeding and mutation breeding techniques with large-scale selection (Shibata 2008). It is quite laborious to generate novel variants that are suitable for field cultivation, however, genetic transformation now offers a valuable method to modify the traits of chrysanthemums. For example, plants with purple- and blue-colored flowers have been successfully developed, in which anthocyanin biosynthetic gene promoters combined with a translational enhancer to allow enough accumulation of pigment and co-pigment in flower petals (Noda et al. 2013, 2017). Furthermore, genome editing has become the latest approach to breeding technology for many plant species (Kumar and Jain 2015), and could be applied to improve several traits in chrysanthemums. To obtain information about gene function in chrysanthemums, and to improve chrysanthemum traits, it is important to understand the promoters involved. Promoters control the location and level of expression of transgenes. However, there is limited knowledge around the variety of useful promoters that may be present in chrysanthemums.
Constitutive promoters, which lead to transgene expression in whole plants to a certain level, are generally used to analyze gene function and to modify overall traits. Of the promoters that have been evaluated, the 35S promoter from the cauliflower mosaic virus (CaMV 35S) has been the one most frequently used to express transgenes in chrysanthemums (Boase et al. 1998, Shinoyama et al. 2002, Takatsu et al. 2000, Urban et al. 1994, Yepes et al. 1995). However, the activity of the 35S promoter in chrysanthemums is lower than that in tobacco (Outchkourov et al. 2003) or torenia plants (Aida et al. 2008). Using the 5′-untranslated region (UTR) of the tobacco alcohol dehydrogenase gene (NtADH-5′UTR) with the 35S promoter can enhance translational efficiency in the expression of β-glucuronidase (GUS) in chrysanthemums (Aida et al. 2008). However, the percentage of GUS-positive transformants in all regenerated transformants was still not high, at 56.3% (9 of 16 transgenic lines) (Aida et al. 2008).
Transgene expression driven by the dual promoter of mannopine synthase 1′ and 2′ (mas 1′–2′) genes from Agrobacterium tumefaciens was detected in transgenic chrysanthemum plants (Shinoyama et al. 2008). However, there was not abundant overexpression of genes that led to the desired phenotype (4.5% and 21.8% of transgenic lines (Shinoyama et al. 2012a, 2012b). Several other plant-genome-derived promoters, namely the potato apoprotein 2 of the light-harvesting complex of photosystem I (LHca3.St1) promoter (Annadana et al. 2001), the chrysanthemum chlorophyll-a/b-binding protein (cab) promoter (Aida et al. 2004), and the chrysanthemum ribulose bisphosphate carboxylase small subunit 1 (rbcS1) promoter-terminator (Outchkourov et al. 2003) showed higher promoter activity than the 35S promoter in chrysanthemum leaves. However, these promoters were intended to be expressed in photosynthetic tissue, and there is no information on the activity of these promoters in calli, nor their durability. The percentage of transformants that showed detectable transgene activity from all regenerated transformants was 58.7% for the LHca3.St1 promoter (Annadana et al. 2001), 15.4%–50.0% for the cab promoter (Aida et al. 2004), and 65.5% for the rbcS1 promoter (Outchkourov et al. 2003).
It would be beneficial to assess promoter activities for successful chrysanthemum transformation. In this study, we aimed to evaluate promoters in terms of the ratio of transformants that showed substantial transgene activity, activity in a number of organs, and durability for long-term culture. To this end, we analyzed the expression of chrysanthemum actin genes. We found that CmACT2 was stably expressed in leaves, floral organs, and calli. We cloned the 2.5 kb upstream region of the actin gene and analyzed its activity as a constitutive promoter. In addition to the CmACT2 promoter, we also analyzed the activity of the parsley ubiquitin (PcUbi) promoter in plant organs, which exhibited greater activity than the 35S promoter in calli (Kishi-Kaboshi et al. 2017). The PcUbi promoter was shown to lead transgene expression in buds, roots, stems, flowers, and seeds of transgenic A. thaliana (Plesch and Ebneth 2011), but its activity in chrysanthemum plants has not been analyzed. We compared the promoter activity of 35S, CmACT2 and PcUbi in several organs, and assessed the recovery rate and durability of transformants, to assess the suitability of these promoters for successful transformation.
The Chrysanthemum morifolium ‘Sei-Marine’ cultivar (Inochio Seikoen Co., Ltd. http://www.seikoen-kiku.co.jp/) was used for this analysis. Chrysanthemum plants were cultivated in a greenhouse under natural daylight conditions.
PrimersAll primers and oligonucleotides used in this study were obtained either from Thermo Fisher Scientific (https://www.thermofisher.com/jp/ja/home.html) or Eurofins Genomics (https://www.eurofinsgenomics.jp/jp/home/); they are listed in Supplemental Table 1. Primers were designed according to information from the National Center for Biotechnology Information (NCBI) database (https://www.ncbi.nlm.nih.gov/), isolated cDNA sequences, isolated gene sequences, and the chrysanthemum EST database (Sasaki et al. 2017).
RNA extraction and reverse transcription PCR (RT-PCR) for actin gene expression analysisWe obtained cells from the following plant organs for RNA analysis: mature corolla of ray florets, mature corolla of disk florets, flower buds >4 mm in length, leaves, pistils of ray florets, pistils of disk florets, and androecia of disk florets.
Total RNA was extracted using TRIzol (Thermo Fisher Scientific) and further purified using an RNeasy mini-kit (QIAGEN; https://www.qiagen.com/jp/). RNA concentration was estimated using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific). cDNAs were synthesized from 500 ng of total RNA using a ReverTra Ace cDNA synthesis kit (TOYOBO, http://www.toyobo-global.com/seihin/xr/lifescience/), and RT-PCR was performed using PrimeSTAR GXL DNA polymerase (Takara, http://www.takara-bio.co.jp/research.htm) with specific primers (Supplemental Table 1). The same amount of cDNA was used for RT-PCR (0.4 μl cDNA per 10 μl RT-PCR reaction mixture). The house keeping gene, TATA-box binding protein 2 (CmTBP2, LC482052), was used as a reference for RT-PCR. The RT-PCR program was set to initial denaturation at 94°C for 2 min followed by 28 cycles (CmACT1, CmACT2) or 30 cycles (CmTBP2) of a three-step cycle: denaturation at 98°C for 10 sec, annealing at 60°C (CmACT1), 62°C (CmACT2) or 66°C (CmTBP2) for 15 sec, and extension at 68°C for 1 min.
Cloning of a promoter for the chrysanthemum CmACT2 geneTwo genes and a promoter, as described below, were isolated from the C. morifolium cultivar ‘Sei-Marine’. Total RNA prepared from leaves was used for gene cloning followed by RT-PCR.
We isolated two actin genes, CmACT1 (LC195830) and CmACT2 (LC195831), using gene-specific primers (Supplemental Table 1), which were synthesized according to information from other plant actin sequence data in the NCBI and chrysanthemum EST databases (Sasaki et al. 2017). The promoter regions of the CmACT2 gene were isolated using a GenomeWalker™ kit (Clontech Laboratories; https://www.clontech.com/), according to the manufacturer’s instructions. The first primer sequences were designed based on the sequence of the CmACT2 cDNA. Subsequently, promoter sequences were isolated in a step-by-step manner. Chrysanthemum genomic DNA was prepared from leaves using an ISOPLANT II kit (Nippon Gene Co., Ltd.; https://www.nippongene.com/english/). The isolated promoter region of the CmACT2 gene (2492 bp) was cloned using a TOPO-TA Cloning Kit (Thermo Fisher Scientific). The nucleotide sequence of the cloned 2.5 kb CmACT2 promoter region was registered in the DNA Data Bank of Japan (DDBJ; http://www.ddbj.nig.ac.jp/Welcome-j.html), and received the accession number LC381917.
Construction of plasmids for promoter analysisWe used the GUS gene, which encodes a hydrolase that catalyzes the cleavage of various β-glucuronides, as a reporter to quantify promoter activity. Before construction of CmACT2 pro::GUS binary vectors, two HindIII sites in the CmACT2 promoter were mutated with a one-base deletion using the KOD-Plus-Mutagenesis Kit (TOYOBO), and cloned to generate a pTOPO-CmACT2pro-plasmid. The mutated CmACT2 promoter region was amplified, digested with HindIII and BamHI, and exchanged with the corresponding site of p35S pro::GUS-HSPT (Kishi-Kaboshi et al. 2017) to produce pCmACT2_pro::GUS-HSPT. HSPT is a DNA fragment comprising a terminator from the heat shock protein 18.2 gene of A. thaliana (Nagaya et al. 2010). The expression cassette harboring the CmACT2 promoter was transferred into the plant binary vector pBCKK (Mitsuda et al. 2005) using the Gateway system (Invitrogen) to produce pBCKK-CmACT2 pro::GUS-HSPT (Fig. 1B).
Construction and activity of 35S pro::GUS, CmACT2 pro::GUS, and PcUbi pro::GUS transformants. (A) Expression level of actin genes in different organs. Expression of CmACT1 (upper) and CmACT2 (middle) were confirmed by RT-PCR. A house keeping gene CmTBP2 (lower) was used as a reference. (B) Schematic diagram of GUS-expression constructs to generate transformants. GUS: β-glucuronidase. HSPT: terminator of heat shock protein. (C) GUS activity was measured in calli at 3-months post-transformation. GUS activity was expressed as pmol 4MU mg−1 protein min−1 and shown transformed as a common logarithm. Circles indicate the value of GUS activity from independent lines; “mean” indicates mean GUS activity in each sample; “n” indicates the number of transgenic calli examined in this study.
The plant binary vector pBCKK containing the 35S (pBCKK-35S pro::GUS-HSPT) or PcUbi (pBCKK-PcUbi pro::GUS-HSPT) promoters were previously generated (Kishi-Kaboshi et al. 2017).
Generation of transgenic chrysanthemum calli and plantsThe binary vector was introduced into Agrobacterium tumefaciens strain EHA105 via electroporation. Transgenic chrysanthemum calli (from leaf tissue) and regenerant plants were generated as previously described (Aida et al. 2004). For the generation of the promoter::GUS-chrysanthemum calli, transformed calli were selected for the presence of npt II using 25 mg L−1 paromomycin on solid medium, and transformation was confirmed by screening for GUS activity. Shoot formation on the explants was observed within 2–3 months following transformation. These shoots were detached from the callus and grown on half-strength Murashige and Skoog (MS) salt medium without hormones and containing an MS vitamin source, 3% sucrose, and 0.25% gellum gum at 20°C for 2–3 months. Surviving shoots were transferred to soil, habituated for 1 month at 25°C, grown at 4°C for an additional month, and then at 25°C in a greenhouse (dedicated to genetically modified plants) with short-day conditions (8 h light and 16 h dark per day). Once they had flowered, the plants were cut back, with their stems cut at 10 cm above the soil, and then maintained at 25°C under long-day conditions (16 h light and 8 h dark per day) in a greenhouse. When the plants were transplanted, 5 cm long shoots were cut from the stock plants, transferred to fresh soil and habituated for 3 weeks at 25°C under long-day conditions (16 h light and 8 h dark per day).
Histochemical and fluorometric GUS assaysGUS activity was analyzed both histochemically and fluorometrically according to the method of Kosugi et al. (1990), with the following modifications. For histochemical semi-quantitative GUS staining, organs were detached from greenhouse-grown plants, immediately transferred to GUS reaction mixture (containing 1 mM 5-bromo-4-chloro-3-indolyl-β-D-glucuronide, 50 mM potassium phosphate buffer [pH 7.0], 10%–20% [v/v] methanol and 1 mM dithiothreitol [DTT]), vacuum infiltrated for 30 min at room temperature, and incubated for approximately 16–24 h at 37°C. The reaction was stopped by replacing the GUS reaction buffer with 70% ethanol, and the pigments and chlorophyll-derived colors were removed using repeated 70% ethanol washing.
Quantitative GUS analysis was carried out using fluorometric analysis. For the fluorometric analyses, each plant sample was immediately frozen with liquid nitrogen and stored at −80°C until use. Frozen samples were homogenized in GUS assay buffer (50 mM potassium phosphate [pH 7.0], 10 mM EDTA, 0.1% [v/v] Triton X-100, 0.1% [v/v] Sarkosyl and 2 mM DTT) and centrifuged at 3,000 g for 5 min. Samples (2–10 μl) of recovered supernatants were made up to 120 μl with GUS assay buffer and incubated with 80 μl 2.5 mM 4-methylumbelliferyl-β-D-glucuronide (4MU) as a substrate at 37°C for 30 min. The amount of 4-methylumbelliferone formed in each GUS reaction was determined using a fluorescence spectrophotometer (VersaFluorTM fluorometer; Bio-Rad; http://www.bio-rad.com/), with a 360 nm excitation wavelength and a 460 nm emission wavelength. For non-transgenic samples, the fluorometric GUS activity was determined using a multiplate reader (Synergy HT; BioTek; https://www.biotek.com/) with the same excitation and emission wavelengths as the VersaFluorTM. The protein content was determined using the Bradford Coomassie Brilliant Blue G-250 protein assay kit (Bio-Rad), with bovine serum albumin as the standard.
In order to expand the choice of promoters that support the constitutive expression of transgenes in chrysanthemums at a high level and over a long time period, we screened chrysanthemum actin genes. We cloned two actin genes, CmACT1 and CmACT2, and analyzed their expression levels; CmACT2 was stably expressed in leaves, ray florets, disk florets, androecia, pistils, flower buds, and calli (Fig. 1A). Disk florets are located at the center of the chrysanthemum flower, while ray florets are located at the perimeter. The expression levels of CmACT1, however, were not consistent among the organs of chrysanthemums we analyzed. Therefore, we chose the CmACT2 gene as the source of promoter used for this investigation, cloned the 2.5 kb region immediately upstream of its start codon and analyzed its ability to express a transgene.
The PcUbi promoter showed high activity in transgenic chrysanthemum calliWe generated three different transgenes containing either the 35S, the CmACT2, or the PcUbi promoter located upstream of the GUS gene (Fig. 1B; 35S pro::GUS, CmACT2 pro::GUS, and PcUbi pro::GUS). There were no obvious differences in calli formation rate among these transformants (data not shown). We examined GUS activity from randomly selected transgenic calli (Fig. 1C). Untransformed plant calli and other organs occasionally display faint background levels of GUS activity when assessed using a fluorometric assay (Table 1). Compared with untransformed plant calli, all transgenic calli clearly expressed GUS activity. When comparing promoter activity among each of the transgenic calli, CmACT2-promoter activity was higher than that of the 35S promoter, while the PcUbi-promoter activity was the highest among the three promoters.
| Calli | Mature leaf | Ray floret | Disk floret | Long-term cultivated tip | |
|---|---|---|---|---|---|
| Range | 0.61–1.16 | bdl–0.57 | bdl–0.18 | bdl–0.16 | bdl |
| Mean | 0.88 | 0.00023 | bdl | bdl | bdl |
GUS activity in non-transgenic chrysanthemums was analyzed using 4 replicates. GUS activity is expressed as pmol 4MU mg−1 protein min−1 and shown in real numbers. bdl: below detection limit.
Transgenic shoots were regenerated from each callus. The regenerant shoots were detached from the calli and grown independently. We found that CmACT2 pro::GUS and PcUbi pro::GUS shoots were more frequently regenerated than 35S pro::GUS shoots (Table 2). Because explants were sometimes cut into pieces while culturing, we assessed the rates of the regenerated shoots from explants as the regeneration rate and compared those. The regeneration rate of the 35S pro::GUS plants were about half those compared with the other transgenic plants (Table 2). We performed the transformation again to obtain additional 35S pro::GUS shoots. The shoot-regeneration rate was also low in 35S pro::GUS transformants.
| Promoter | No. of explants used for transformation | No. of regenerant plantsa | Regeneration rate (%)b |
|---|---|---|---|
| 35S | 846 | 12 | 1.4 |
| 420 | 5 | 1.2 | |
| CmACT2 | 777 | 22 | 2.8 |
| PcUbi | 746 | 24 | 3.2 |
To analyze promoter activity during the growth and development of transformed regenerants, 12 regenerated transformant shoots from each promoter system were transferred to soil to continue growing. We compared promoter activity among transgenic plants at the flowering stage (11–12 months following transformation), once the top flower was completely open. Plants which had suffered severe insect damage were eliminated from the study and the remaining ten of each type of plant were used for analysis. The upper leaves, which are located immediately below the top flower, were histochemically stained to analyze the localization of promoter-active sites and semi-quantitatively assess the staining intensity (and hence promoter activity) (Fig. 2A). The stain intensity and staining patterns varied among the plants (representative images are shown in Fig. 2A). Dark precipitates were seen in seven PcUbi pro::GUS plants and two 35S pro::GUS plants, suggesting higher levels of GUS activity (Fig. 2B). Conversely, no CmACT2 pro::GUS plants exhibited visible GUS staining. To quantitatively compare promoter activity, we conducted fluorometric assays for GUS activity using mature leaves obtained from locations toward the middle of each plant (Fig. 2B). We considered plants that expressed >100 pmol 4MU mg−1 protein min−1 GUS activity to be GUS-positive, which implies substantial expression of the transgene. PcUbi pro::GUS plants exhibited the most abundant GUS-positive leaves (9 of 10, Fig. 2B). The mean level of GUS activity in mature leaves of plants transformed by 35S pro::GUS or CmACT2 pro::GUS were similar, whereas GUS-positive leaves were more abundant in CmACT2 pro::GUS plants (7 of 10) than in 35S pro::GUS plants (4 of 10). The level in PcUbi pro::GUS-transformed mature leaves appeared to be higher compared with the others.
GUS activity in leaf tissue of 35S pro::GUS, CmACT2 pro::GUS, and PcUbi pro::GUS transformants. GUS staining in upper leaves (A) and GUS activity in mature leaves (B) at 10–11 months post-transformation was analyzed. For histochemical analysis, stain intensity and pattern varied among transformants, and representative samples are shown (A). GUS activity was measured in the mature leaves of each transformant line and compared with the staining results of upper leaves of the same line (B). GUS activity is expressed as pmol 4MU mg−1 protein min−1. Upper leaf staining results are indicated below the graph. Vertical bars in the graph show mean values. Highlighted samples expressed >100 pmol 4 MU mg−1 protein min−1 GUS activity and are considered GUS-positive.
Next, we compared promoter activities in disk florets and ray florets from the top flower of the three transformants (Fig. 3). Similar to the results from the top leaves, histochemical analysis of PcUbi pro::GUS plants showed obvious staining, whereas 35S pro::GUS plants showed faint staining. In PcUbi pro::GUS plants, disk florets from eight plants and ray florets from six plants showed dark staining. In the 35S pro::GUS plants, disk florets and ray florets from three plants showed faint staining. In CmACT2 pro::GUS plants, no staining was detected.
GUS activity in the flower tissue of 35S pro::GUS, CmACT2 pro::GUS, and PcUbi pro::GUS transformants. Histochemical staining of ray florets and disk florets (A). Stain intensity and pattern varied among transformants, and representative samples are shown. GUS activity of disk florets and ray florets at 10–11-months post-transformation is shown (B). GUS activity is expressed as pmol 4MU mg−1 protein min−1. Black bars: ray florets. Gray bars: disk florets. Vertical bars in the graph and following numbers are mean values of ray florets (black) and disk florets (gray). Highlighted samples expressed >100 pmol 4MU mg−1 protein min−1 GUS activity and are considered GUS-positive.
We next quantitatively assayed GUS activity in the florets of these transgenic plants. GUS-positive florets were most abundantly observed in PcUbi pro::GUS plants. Mean GUS activity in the disk florets and ray florets of PcUbi pro::GUS plants was clearly the highest (Fig. 3B). Comparing 35S pro::GUS with CmACT2 pro::GUS plants, mean GUS activity was slightly higher in CmACT2 pro::GUS plants; however, the number of GUS-positive florets was higher in 35S pro::GUS plants. Combining the quantitative and staining results, it appears that CmACT2 promoter activity is more scattered than that of the 35S promoter, therefore no staining was visible in CmACT2 pro::GUS florets. It also appears that 35S-promoter activity is highly localized, even within the same compound flower.
Durability of promoter activity in long-term-cultured chrysanthemum plantsChrysanthemum plants for commercial production are propagated asexually, therefore it is necessary that any transgene present in a parent plant should also be expressed in cuttings, in order to preserve the novel trait for several years. To determine the durability of promoter activity during cultivation, we analyzed GUS activity in two types of shoot tips following long-term cultivation (Fig. 4). The first type was tips from plants whose main shoots were cut down after flowering and the remaining plant material was maintained under long-day conditions (i.e., in a vegetative state) for 2 years in the same pots without fertilizer (long-term-cultivated tips, Fig. 4A). These plants exhibited slow growth because of the loss of soil nutrients and their pot-bound nature. The second type was planted tips, which were also maintained for 2 years but after they were cut they were transferred from their original pot to fresh soil and grown for an additional 3–4 weeks under long-day conditions (planted tips, Fig. 4A). In both the long-term-cultivated and the planted tips, the percentage of GUS-positive lines in each transformant and mean GUS activity was highest in PcUbi pro::GUS plants (Fig. 4B). A comparison of the data from long-term-cultivated tips and planted tips from the same parent, in three 35S pro::GUS plants, and six PcUbi pro::GUS plants, revealed that GUS activity was apparently higher (over ten-fold higher) in the planted tips, suggesting that although GUS activities of these lines had once declined, these activities recovered after transplantation. Conversely, the CmACT2 pro::GUS plants clearly showed the lowest GUS activity for both long-term-cultivated tips and planted tips. Comparison of the data from long-term-cultivated tips and planted tips in the CmACT2 pro::GUS plants revealed that GUS activities were not clearly changed after transplantation, indicating that transgene expression apparently did not recover following asexual propagation of these transformants.
GUS activity in long-term-cultivated 35S pro::GUS, CmACT2 pro::GUS, and PcUbi pro::GUS transformants. Positions of sample collection are schematically illustrated (A). Long-term-cultivated tips were obtained from plants that flowered once under short-day conditions, after which the main stem was cut and grown under long-day conditions for 2 years. The planted tips were prepared as for the long-term-cultivated tips, and other parts of the same plant were cut and transferred to fresh soil. These cuttings were harvested after 3–4 weeks of long-day conditions. GUS activity of the shoot tips in the 35S pro::GUS, CmACT2 pro::GUS and PcUbi pro::GUS transformants (B). Black bars: long-term-cultivated tips. Gray bars: planted tips. GUS activity is expressed as pmol 4MU mg−1 protein min−1. Vertical lines and following numbers show the mean of long-term-cultivated tips (black) and planted tips (gray). Highlighted samples expressed >100 pmol 4MU mg−1 protein min−1 GUS activity and are considered GUS-positive.
In the present study, we aimed to widen the variety of constitutive promoters that could lead to the high level and durable expression of a transgene in several organs of chrysanthemums. We showed that PcUbi promoters exhibit high activity in planta and retain this activity even after long-term cultivation. We isolated the 2.5 kb CmACT2 promoter and showed that this promoter has higher activity in calli compared with the 35S promoter (Fig. 1C), and comparable levels in mature leaves (Fig. 2) and flowers (Fig. 3). In addition, the shoot-regeneration rate, an important factor to consider for the efficient production of transgenic chrysanthemums, was higher in CmACT2 pro::GUS transformants than in 35S pro::GUS transformants (Table 2). In the transgenic plants in which transgene expression driven by the 35S promoter gradually declined, antibiotic resistance driven by the nos promoter also decreased (Weinhold et al. 2013). We suspect that the 35S promoter affected the expression of the surrounding nptII and lead to a decline of the regeneration rate in this study.
Contrary to our initial expectations, CmACT2-promoter activity was low in shoot tips following two years of cultivation and did not increase after transplantation, suggesting a lack of durability of expression (Fig. 4). Although CmACT2 expression levels appeared to be higher in leaf blades and in florets than in calli (Fig. 1A), GUS activity in leaves and flowers of CmACT2 pro::GUS plants (Figs. 2B, 3B) was lower than that in calli (Fig. 1C). This suggested that the CmACT2-promoter length used here was inadequate and resulted in decreased expression, presumably caused by unexpected gene silencing. Currently, we could not determine the CmACT2 promoter length appropriate for constitutive and strong expression of transgenes as originally expected. Evaluation of promoter constructs of different lengths is required to analyze the availability of CmACT2 promoters for constitutive expression.
While we were performing our analysis, a report on the activity of another chrysanthemum actin promoter (CmActin, −1372 to +60) was published by another research group (Hong et al. 2016). We compared the CmActin and CmACT2 promoters in the genome of Chrysanthemum seticuspe (Hirakawa et al. 2019), which is a wild relative of Chrysanthemum morifolium. The CmActin and CmACT2 sequences had two different counterparts in C. seticuspe gene sequences (Supplemental Fig. 1). Moreover, when we aligned the sequences of CmActin and CmACT2 promoters only part of the sequences adjacent to the start codon were aligned. Therefore, we think that the origin of these two actin promoters is different. Using selected transformants, the CmActin promoter showed higher activity than the 35S promoter (Hong et al. 2016). However, promoter activity also differed substantially among individual transformants (Figs. 2–4 in this paper; Aida et al. 2008). It is necessary to observe a number of transformants and the distribution of activity to assess the suitability of the CmActin promoter. Then, the CmActin promoter could be another choice of promoter for chrysanthemum transformation.
We showed that several 35S pro::GUS transformants exhibited highly localized GUS staining in their flowers (Fig. 3). It has previously been reported that phloem tissue exhibited the highest GUS staining, whereas parenchyma tissue did not exhibit clear GUS staining, in transgenic tobacco plants expressing GUS under the control of the 35S promoter (Jefferson et al. 1987). Conversely, in our study the GUS activity levels of CmACT2 pro::GUS and 35S pro::GUS were very similar in both mature leaves and florets; however, no CmACT2 pro::GUS transformants exhibited GUS staining (Figs. 3, 4). We suspect that the CmACT2 promoter has rather uniform activity in different tissues, resulting in dispersed GUS-staining precipitate, which may not be visible.
We demonstrated that PcUbi promoters have high activity in calli, leaves, and flowers. In contrast to the 35S pro::GUS plants, PcUbi pro::GUS plants with high GUS activity, as determined by fluorometric assay, also showed a high proportion of stained disk florets (Fig. 3B). Disk florets of PcUbi pro::GUS line 18 did not exhibit GUS staining, but GUS activity by fluorometry in this line was still higher than that of any other disk florets of 35S pro::GUS lines, which show staining. These results indicate that the PcUbi promoter has stronger and relatively more uniform activity than the 35S promoter in chrysanthemums. In addition to its high activity, the PcUbi promoter showed high durability after two years of cultivation, with more than half of the transplanted PcUbi pro::GUS tips clearly recovering their GUS activity (Fig. 4B). We also observed that the percentage of GUS-positive transformants in all regenerated transformants was high in the upper leaves of mature plants (Fig. 2B). In addition, all planted tips of PcUbi pro::GUS transformants were considered to be GUS-positive plants (Fig. 4B). As far as we know, there are no other promoters which show clear transgene activity at such high proportions in chrysanthemums. These results indicate that the PcUbi promoter is more useful than the 35S and the current 2.5 kb CmACT2 promoters in terms of transgene expression strength and durability, and with respect to the yield ratio of transformants in chrysanthemums.
Ubiquitin is a highly conserved protein in eukaryotes and is encoded by polyubiquitin genes and ubiquitin extension protein genes. In many plants, polyubiquitin genes are expressed in various plant tissues (Binet et al. 1991, Burke et al. 1988, Kawalleck et al. 1993). To express a transgene in a constitutive manner, various ubiquitin promoters were isolated, and their activities were analyzed in both monocot and dicot plants (Christensen et al. 1992, Hirai et al. 2011, Norris et al. 1993). Arabidopsis thaliana genome carries five polyubiquitin genes, whose mRNA levels are independently modulated (Sun and Callis 1997). Here, we successfully used the PcUbi promoter for the constitutive transgene expression in chrysanthemum and we intend to seek additional chrysanthemum ubiquitin genes in a future study to widen our choice of promoters in chrysanthemum.
To our knowledge, with the exception of our analysis using chrysanthemum calli (Kishi-Kaboshi et al. 2017), there has only been one other previously published report, a patent application, that describes PcUbi promoter activity in plants (Plesch and Ebneth 2011). In this patent application, the PcUbi promoter was shown to lead transgene expression in buds, roots, stems, flowers, and seeds of transgenic A. thaliana, based on histochemical analysis (Plesch and Ebneth 2011). Here, we showed that the PcUbi promoter has high activity in calli, leaves, flowers and tips of chrysanthemums, using a quantitative analysis. We believe that the PcUbi promoter is currently the most suitable promoter for constitutive and high-level transgene expression in chrysanthemums, and that it can accelerate chrysanthemum breeding and the analysis of gene functions.
M. K. and K. S. planned the study. M. K. mainly performed analyses and prepared first draft of the manuscript; R. A. and K. S. critically reviewed and improved the final manuscript.
We thank Inochio Seikoen Co., Ltd. (http://www.seikoen-kiku.co.jp/) for providing us with Chrysanthemum morifolium cultivar ‘Sei-Marine’ and Ms. Satoko Ohtawa, Ms. Yukiko Tsuruoka, Ms. Yuko Namekawa, and Ms. Yoshiko Kashiwagi for generating and maintaining transgenic chrysanthemum plants and calli. This work was supported by Cabinet Office, Government of Japan, Crossministerial Strategic Innovation Promotion Program (SIP), “Technologies for creating next-generation agriculture, forestry and fisheries” (funding agency: Bio-oriented Technology Research Advancement Institution, NARO), and the research fund of NARO Institute of Vegetable and Floriculture Science (NIVFS).
