External dsRNA Downregulates Anthocyanin Biosynthesis-Related Genes and Affects Anthocyanin Accumulation in Arabidopsis thaliana

2.1. AtCHS-Specific Exogenous dsRNAs Downregulate AtCHS mRNA Levels and Anthocyanin Content in Arabidopsis

We used PCR and in vitro transcription protocol to produce dsRNA molecules of the AtCHS gene, encoding the key enzyme in the flavonoid/anthocyanin biosynthesis pathway (Figure 1), and the non-related NPTII bacterial gene, encoding the bacterial neomycin phosphotransferase II enzyme. We synthesized the NPTII-dsRNAs and treated wild-type A. thaliana to verify whether any observed effects on AtCHS mRNA levels were sequence-specific. To analyze the effect of exogenous dsRNAs on the expression of AtCHS, large fragments of AtCHS and NPTII cDNAs were amplified (Figure 2a,d). Then, the obtained PCR products, containing T7 promoters at both ends, were used as templates for in vitro transcription. For external application, the synthesized dsRNAs were diluted in water to a final concentration of 0.35 µg/µL. The dsRNAs (100 µL of each dsRNA per individual plant, i.e., 35 µg) were applied on the leaf surface (on both the adaxial and abaxial sides) of four-week-old A. thaliana rosettes by spreading with sterile individual soft brushes [31] (Supplementary Video S1). Importantly, we treated the four-week-old rosettes of A. thaliana at a late day time (21:00–21:30) under low soil moisture conditions in all experiments, since the appropriate plant age, late day time, and low soil moisture at the time of dsRNA application were important parameters for successful NPTII suppression in transgenic A. thaliana according to our recent analysis [31]. An analysis of the dsRNA concentration effect on transgene silencing in A. thaliana was performed previously [30], indicating that 35 µg of a transgene-encoding dsRNA resulted in the highest transgene silencing efficiency as compared to other dsRNA concentrations.

Then, we studied whether exogenous application of the naked AtCHS and NPTII-dsRNAs on the foliar surface of wild-type A. thaliana could led to any changes in the mRNA transcript levels of AtCHS gene and anthocyanin levels in comparison with the control water treatment (Figure 3). Since under standard cultivation conditions anthocyanin production and AtCHS expression were low, we divided the treated A. thaliana rosettes into two groups for post-treatment incubation either under control conditions (+22 °C, 16 h light) or anthocyanin-inducing conditions (+7 °C, and 23 h light) for two and seven days in order to induce AtCHS expression and anthocyanin biosynthesis (Figure 3a). qRT-PCR revealed that cultivation of A. thaliana under the anthocyanin-inducing conditions resulted in a dramatically higher AtCHS mRNA levels in the control plants treated with water and NPTII-dsRNAs than cultivation under control conditions (Figure 3b). Importantly, this AtCHS-induction effect was not observed for plants treated with AtCHS-dsRNAs under the anthocyanin-inducing conditions, at both two days and seven days post-treatment (Figure 3b). Notably, AtCHS mRNA levels were considerably lower in the AtCHS-dsRNA-treated A. thaliana than in the water- and NPTII-treated controls under the anthocyanin-inducing conditions, at both two days and seven days post-treatment.

Both HPLC and spectrophotometric analysis of total anthocyanins revealed that plants treated with AtCHS-dsRNAs, but not with NPTII-dsRNAs, exhibited a markedly lower total content of anthocyanins under the anthocyanin-inducing conditions than the water-treated plants (Figure 3c and Figure S1a). Using HPLC with high-resolution mass spectrometry (HPLC-MS), we detected eight anthocyanin compounds in the water- and dsRNA-treated leaves of A. thaliana (Figure 4 and Figure S2a; Table S1). It is possible that other anthocyanins were present in the analyzed tissues of A. thaliana but in trace amounts. The content of most individual anthocyanins was higher in the plants grown under the anthocyanin-inducing conditions than at +22 °C (Figure 4a and Figure S2a). Plant treatment with exogenous AtCHS-dsRNAs reduced the content of all individual anthocyanins, and the changes were statistically significant for A7, A9, and A11* (Figure 4b and Figure S2a).

2.2. AtCHS-Specific Exogenous siRNAs Downregulate AtCHS mRNA Levels and Anthocyanin Content in Arabidopsis

Two 21-nt long complimentary single-stranded RNAs (ssCHS-s and ssCHS-a) designed to target the AtCHS mRNAs were in vitro synthesized and HPLC purified (Table S2). In addition, two 21-nt long complimentary single-stranded RNAs (NPTII R3-s-Me and R3-a-Me) designed to target the NPTII mRNAs [34] were in vitro synthesized and HPLC purified (Table S2). The complimentary ssRNAs contained a phosphate group at 5′ end, a 2ʹ-O-methyl at 3′ end, and 2-nt 3′ overhangs at both ends (Figure 2a,d; Table S2). The ssRNAs were combined and annealed to form siRNAs, i.e., siCHS and siNPTII (Figure 2a,d). The NPTII-specific siRNA was included in the study to verify whether any observed effects of AtCHS-siRNA were sequence-specific. In our earlier study, 50 pmol/μL was chosen as the optimal concentration of the NPTII-siRNAs for plant foliar treatments due to the combination of effectiveness and lower cost of ssRNA synthesis [34].

qRT-PCR analysis revealed that the level of AtCHS mRNAs was considerably lower in siCHS-treated plants than in the water- and siNPTII-treated controls grown under the anthocyanin-inducing conditions for two days after the treatments (Figure 5a,b). We noted that the effect became less evident seven days after the treatments. Both HPLC and spectrophotometric analysis of total anthocyanins in the plants cultivated for seven days after the treatments revealed that the siCHS treatment led to a markedly lower anthocyanin content under the anthocyanin-inducing conditions than the water and siNPTII treatments (Figure 5a,c and Figure S1b). siNPTII resulted in the anthocyanin level comparable to that in the water-treated plants. Using HPLC-MS, we also detected eight anthocyanin compounds in the A. thaliana treated with water or siRNAs (Figure S2b; Table S1). The siCHS treatment considerably lowered content of most individual anthocyanins.

2.3. Targeting AtMYBL2 and ANAC032 Repressors by Foliar dsRNA Treatments Downregulates Their mRNA Levels and Leads to a Pronounced AtCHS Upregulation

We used PCR and in vitro transcription protocol to produce dsRNA molecules of the AtMYBL2 and AtANAC032 genes, encoding transcriptional repressors negatively regulating anthocyanin biosynthesis in A. thaliana [50,51]. Full-length coding cDNAs of the AtMYBL2 and AtANAC032 genes were amplified (Figure 2b,c). The obtained PCR products, containing T7 promoters at both ends, were used as templates for in vitro transcription. For external plant treatments, the synthesized dsRNAs were diluted and applied on the leaf surface (on both the adaxial and abaxial sides) of four-week-old wild type Arabidopsis, as described above for AtCHS-dsRNA. At the same time, we treated A. thaliana with NPTII-dsRNA to verify whether any observed effects of the dsRNAs were sequence-specific.

Then, we studied whether simple exogenous application of the AtMYBL2-, AtANAC032-, and NPTII-dsRNAs to the foliar surface of four-week-old A. thaliana could lead to any changes in the mRNA transcript levels of AtMYBL2, AtANAC032, and AtCHS genes in comparison with the water-treated controls two and seven days post-treatments (Figure 6a). For this purpose, we also divided the treated A. thaliana rosettes into two groups for incubation under control conditions (+22 °C, 16 h light) and anthocyanin-inducing conditions (+7 °C, 23 h light). Foliar plant treatment with the AtMYBL2-dsRNAs triggered considerable downregulation of AtMYBL2 mRNA levels both under standard and anthocyanin-inducing conditions (Figure 6b). AtMYBL2 transcript levels were significantly reduced two and seven days post-treatment at +22 °C and seven days post-treatment at +7 °C. Exogenous application of AtANAC032-dsRNA also resulted in a pronounced inhibition of AtANAC032 mRNA levels, especially under the anthocyanin-inducing conditions (Figure 6c). Further analysis revealed a dramatic upregulation of AtCHS expression after application of both AtMYBL2- and AtANAC032-dsRNAs (Figure 6d). However, the dsRNA-induced effect on anthocyanin accumulation was not as evident as that for gene transcript levels (Figure 6e and Figure S1c). While the AtMYBL2- and AtANAC032-dsRNA treatments led to a 3.3-6.4-fold upregulation of AtCHSs, the total content of anthocyanins was considerably increased only after foliar treatment with AtMYBL2 and by 1.7-fold (Figure 6d,e). HPLC analysis of individual anthocyanins revealed a statistically considerable increase in the content of A8, A11, and A9 after treatment with AtMYBL2-dsRNA (Figure S2c). Importantly, exogenous application of NPTII-dsRNA did not have a marked effect on the AtMYBL2, AtANAC032, or AtCHS mRNA levels.

2.4. Detection of AtCHS-dsRNA in A. thaliana Leaves by Laser Scanning Microscopy

To further investigate localization and transport of the exogenous dsRNAs, we labeled the in vitro synthesized AtCHS-dsRNAs with the Cy3 dye and applied to the adaxial and abaxial foliar surface of four-week-old A. thaliana rosettes at 21:00 at a concentration of 0.35 µg/µL [31] by spreading with sterile individual soft brushes. Linear unmixing of the λ-stacks revealed two peaks of fluorescence, including a peak at ≈570 nm–580 nm and a peak at ≈690 nm (Figure S3). The peak at 570 nm–580 nm corresponds to Cy3 (Silencer™ siRNA Labeling Kit with Cy™3 dye), and the peak at 690 nm—chlorophyll fluorescence [52]. Inspection of the adaxial and abaxial leaf surface by laser scanning microscopy 13–15 h post-treatment detected the presence of the Cy3-labeled dsRNA in the leaf veins, parenchyma cells, and stomata of the dsRNA-treated A. thaliana leaves (Figure 7).

4.1. Plant Material and Growth Conditions

The seeds of wild-type A. thaliana (cv. Columbia) were vapor-phase sterilized as described [34] and plated on solid ½ Murashige and Skoog (MS) medium for two days at 4 °C. Then, the plates were kept at 22 °C for 1 week in a growth chamber (Sanyo MLR-352, Panasonic, Osaka, Japan) at a light intensity of ~120 μmol m⁻² s⁻¹ over a 16 h daily light period. One-week-old A. thaliana seedlings were planted to pots (7 cm × 7 cm) containing 100 g of commercially available rich soil (the soil was well-irrigated by filtered water applied at the bottom of the pots). Then, the plants were grown in the chamber at 22 °C under plastic wrap for additional three weeks without additional irrigation before RNA treatments of four-week-old plants. After the RNA treatments, the A. thaliana was incubated for additional seven days either under control (+22 °C, 16 h daily light period) or anthocyanin-inducing (+7 °C, 23 h daily light period) conditions in a growth chamber (KS-200, Smolenskoye SKTB SPU, Smolensk, Russia) without further irrigation to induce AtCHS expression and anthocyanin accumulation.

4.2. Isolation and Sequencing of AtCHS, AtMybL2, and AtANACO323 Transcripts

Full-length coding cDNA sequences of AtCHS (AT5G13930.1, 1188 bp) AtMYBL2 (AT1G71030.1, 588 bp), and AtANAC032 (AT1G77450, 762 bp) were amplified by RT-PCR using RNA samples extracted from the adult leaves of A. thaliana. The RT-PCRs were performed in a Bis-M1105 Thermal Cycler (Bis-N, Novosibirsk, Russia). The primers are listed in Table S3. The RT-PCR products were subcloned into pJET1.2/blunt and sequenced as described previously [54].

4.3. dsRNA Synthesis and Application

All dsRNAs were synthesized using the T7 RiboMAX™ Express RNAi System (Promega, Madison, WI, USA). For this purpose, the cloned full-length cDNAs of AtMYBL2 and AtANAC032 and a large cDNA fragment of AtCHS (736 bp out of 1188 bp) were amplified by PCR for in vitro transcription and dsRNA production. We also amplified a large fragment of NPTII (GenBank AJ414108, 599 bp out of 798 bp) using pZP-RCS2-nptII plasmid [55]. The T7 promoter sequence was introduced into both the 5′ and 3′ ends of the amplified AtCHS, AtMYBL2, AtANAC032, or NPTII in a single PCR for each gene using primers listed in Table S3. The PCRs were performed in the Bis-M1105 Thermal Cycler programmed according to T7 RiboMAX™ Express RNAi System instructions. Then, the obtained PCR products were used as templates for in vitro transcription and dsRNA synthesis following the manufacturer’s protocol. The resultant dsRNAs were analyzed by gel electrophoresis and spectrophotometry to estimate dsRNA purity, integrity, and amount.

The AtCHS-, AtMYBL2-, AtANAC032-, and NPTII-dsRNAs were applied to individual four-week-old rosettes of wild-type A. thaliana by spreading with individual soft brushes (natural pony hair) sterilized by autoclaving [31] (Supplementary Video S1). For each dsRNA treatment, 35 µg of the dsRNA were diluted in 100 µL of nuclease-free water and applied to the foliar surface (all leaves of one rosette for each type of condition were treated on both the adaxial (upper) and abaxial (lower) sides). One plant of A. thaliana was treated with the dsRNA of each type (100 µL) and one plant—with sterile filtered water (100 µL) in each independent experiment. The dsRNAs in all experiments were applied to four-week-old rosettes of A. thaliana at a late day time (21:00–21:30) under low soil moisture conditions, since the appropriate plant age, late day time, and low soil moisture at the time of dsRNA application were important parameters for successful NPTII suppression in transgenic A. thaliana according to our recent analysis [31]. Soil water content before dsRNA treatment was 50–60%.

4.4. siRNA Synthesis and Application

The AtCHS- and NPTII-encoding ssRNAs were in vitro-synthesized, modified, and HPLC purified by Syntol (Moscow, Russia). The RNA oligonucleotide sequences are presented in Table S2 and Figure 2a,d. The synthesized single-stranded oligonucleotides were prepared to form siRNA as described [34]. Briefly, to form the siRNA duplexes, equal volumes of the ssRNAs diluted to a concentration of 100 pmol/µL were combined and annealed at 90 °C for 1 min. Then, the mixture was slowly cooled to room temperature. The final concentration of annealed oligonucleotides was 50 pmol/µL. 100 µL of each siRNA duplex or 100 µL of nuclease-free water were applied onto the leaf surface of four-week-old A. thaliana by spreading with individual soft brushes as described above for dsRNA.

4.5. RNA Isolation and Reverse Transcription

For RNA isolations, a typical adult leaf of A. thaliana was collected from the same individual plant before treatment, two days and seven days post-treatment for each type of treatment in an independent experiment. Total RNA was isolated using the cetyltrimethylammonium bromide (CTAB)-based protocol [56] and complementary DNAs were synthesized as described [57].

4.6. Gene Expression Analysis by qRT-PCR

The reverse transcription products were amplified by PCR and verified for the absence of DNA contamination, using primers listed in Table S3. The qRT-PCRs were performed with SYBR Green I Real-time PCR dye and a real-time PCR kit (Evrogen, Moscow, Russia) as described [58] using two internal controls (GAPDH and UBQ) selected in previous studies as relevant reference genes for qRT-PCRs on Arabidopsis [59]. The expression was calculated by the 2−ΔΔCT method [60]. All gene identification numbers and used primers are listed in Table S3.

4.7. Quantification of Anthocyanins

Anthocyanin content in A. thaliana rosettes was determined using a SPECTROstar Nano spectrophotometer (BMG Labtech, Ortenberg, Germany) as described in Teng et al. [61]. Treated A. thaliana rosettes were frozen at −20 °C and subsequently homogenized using a mortar and a pestle. Shredded tissue was weighed and extracted for 1 d at 4 °C in 1 mL of 1% (v/v) hydrochloric acid in methanol. Then, the mixture was centrifuged at 13,200 rpm for 15 min and the absorbance of the supernatant was measured at 530 and 657 nm. Relative anthocyanin concentrations were calculated as described [61].

For HPLC–MS analysis, the samples were filtered through a 0.45-um nylon filter. Identification of all anthocyanins was performed using a 1260 Infinity analytical HPLC system (Agilent Technologies, Santa Clara, CA, USA) coupled to Bruker HCT ultra PTM Discovery System (Bruker Daltonik GmbH, Bremen, Germany) equipped with an electrospray ionization (ESI) source. The data for anthocyanins were acquired in positive ion mode under the operating conditions as described [62]. The MS spectra were recorded across an m/z range of 100–1500, and the individual anthocyanins were identified as described [63].

HPLC with diode array detection (HPLC–DAD) for quantification of all anthocyanins was performed using a HPLC LC-20AD XR analytical system (Shimadzu, Kyoto, Japan). DAD data were recorded in the 200–800 nm range, and chromatograms for quantification were acquired at 530 nm. The chromatographic separation was performed on Shim-pack GIST C18 column (150 mm, 2.1-nm i.d., 3-µm part size; Shimadzu, Japan). Anthocyanins were separated using 0.1% formic acid and acetonitrile as mobile phases A and B, respectively, with the following elution profile: 0 to 35 min 0% of B; 35 to 40 min 40% of B; 40 to 50 min 50% of B; 50 to 65 min 100% of B. 5 μL of the sample extract was injected with a constant column temperature maintained at 40 °C and a flow rate maintained at 0.2 mL/min. All solvents were of HPLC grade. The contents of anthocyanins were determined by external standard method using the four-point regression calibration curves built with the available standards. The commercial standard cyanidin chloride was obtained from Sigma-Aldrich (St. Louis, MO, USA) and used as the control.

4.8. Fluorescent dsRNA Labeling and Laser Scanning Microscopy

Fluorescent labeling of the in vitro synthesized AtCHS-dsRNA was performed using the Silencer™ siRNA Labeling Kit with Cy™3 dye (Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer’s instructions. A total of 35 µg of the labeled dsRNA (100 µL) was applied on the adaxial and abaxial leaf surface of four-week A. thaliana as described above. Fluorescent signals were analyzed 13–15 h after the foliar plant treatments by laser scanning microscopy. The whole leaves were mounted in distilled water in a Petri dish and were observed under a Zeiss LSM 780 laser scanning microscope operated in λ-mode equipped with a Plan-Apochromat 20×/0.8 and a Plan-Neofluar 40×/0.6 objectives. The excitation wavelength of an argon laser was set at 488 nm and the emission signal was registered at 20 evenly spaced wavelengths (8.9 nm apart) in a range from 500 to 693 nm by using a QUASAR detector. Finally, the resulted λ-stacks were linearly unmixed with the Zeiss Zen 2.1 SP3 (Black Edition) software. The laser scanning microscopy was carried out at the Far Eastern Center of Electron Microscopy (A.V. Zhirmunsky National Scientific Center of Marine Biology, FEB RAS, Vladivostok, Russia).

4.9. Statistical Analysis

The data are presented as mean ± standard error (SE) and were tested by paired Student’s t-test. The p < 0.05 level was selected as the point of minimal statistical significance in all analyses. At least three independent experiments were performed for each type of analysis.