Two cultivars of Polish spring wheat, Kontesa, and Torka (Triticum aestivum L), and one cultivar of spring triticale, Wanad (x Triticosecale Wittmack), were used in the experiments. Wild type, donor plants for transformation experiments as well as transgenic plants were grown under controlled conditions of long day (16 h light) as described by Gasparis et al. (2013).

Artificial microRNA cassettes for silencing Pina and Pinb genes in wheat, and Sina and Sinb genes in triticale, were constructed based on precursor microRNA of wheat Tae-miR164 (Mirbase.org: MI0006173). Primers for amplification and sequencing of precursor Tae-miR164 were designed based on the EST sequence (access number CA704421). The sequences of the primers were: forward 5′-AGGTGGAGAAGCAGGGCACGT-3′ and reverse 5′- TATCACGCAGAGCTGACACCACAA-3′. The primers were used to amplify specific precursor Tae-miR164 in Kontesa and Torka cultivars. In the next step, native fragments of microRNA of 21 bp in precursor Tae-miR164 were replaced with the sequence complementary to the Pin and Sin genes. The 21 bp sequences of microRNA for these genes were designed based on the Web MicroRNA Designer platform (WMD3¹) (Ossowski et al., 2008). The sequences were: (1) for Pina-derived amiR (ami-Pa): 5′-UGAAAUCCGAAGAUGCCATCG-3′ complementary to the sequence of allele Pina-D1a in the region of 321–341 and (2) for Pinb-derived amiR (ami-Pb): 5′-UGUUUGAAUACCUCACCUAGC-3′ complementary to the sequence of allele Pinb-D1c in the region of 348–368. Precursor microRNA was cloned into the pGEM-T (Promega) vector and used as a template for multiple PCR amplification to insert designed amiRNA into the precursor. Primers used in these reactions are as follows:

The order of the amplification procedure was: (1) pgt_A and ami_pa2 primers; (2) ami_pa3 and pgt_B primers; (3) ami_pa1 and ami_pa4 primers, and (4) products of amplification from 1 to 3 reactions plus pgt_A, pgt_B primers.

The cassettes containing ami-Pa and ami-Pb were cloned into the pBract214 overexpression vector². The vector is compatible with the Gateway cloning system. In the first step, the cassettes were amplified using: ami_pa5 5′-GGTGAAATCCGAAGATGCCATCGTT-3′, ami_pa6 5′-GATTGGTGAAATCCGAA-3′, and ami_pb5 with: 5′-GGTGTTTGAATACCTCACCTAGCTT-3′, ami_pb6 5′-GATTGGTGTTTGAATAC-3′, respectively and cloned into the entry vector pCR8/GW/TOPO (Invitrogen). In the next step, the cassettes were cloned to the destination pBract214 vector in the Gateway reaction. The presence of the amiRNA in the vectors was verified by restriction analysis and sequencing. The vectors were electroporated into the AGL1 strain of Agrobacterium tumefaciens and used for transformation.

Agrobacterium-mediated transformation experiments were performed according to our previously described protocols for wheat (Przetakiewicz et al., 2003, 2004) and for triticale (Nadolska-Orczyk et al., 2005). Putative transgenic plants were regenerated and selected on modified MS media containing 20 mg l^-1 of hygromycin as a selectable agent.

Genomic DNA was isolated from young leaves of 5 days seedlings according to the modified CTAB procedure (Murray and Thompson, 1980). The PCR was carried out in a 25 μl reaction mixture using Platinum Taq DNA Polymerase (Invitrogen) and 120 ng of template DNA. The reaction was run using the following program: initial denaturation step at 94°C for 2 min, 35 cycles of amplification at 94°C for 30 s, 65°C for 30 s, 72°C for 30 s with a final extension step at 72°C for 5 min.

The wheat and triticale plants transformed with Pina-derived amiR and Pinb-derived amiR vectors were tested with two pairs of specific primers amplifying a fragment of the hpt selection gene. The sequences of the primers for the first pair were: pBr_hyg-F2 5′-GACGGCAATTTCGATGATG-3′ and pBr_hyg-R2 5′-CCGGTCGGCATCTACTCTAT-3′, and amplified fragment of 205 bp. The sequences of the primers for the second pair were: pBr_hygF4 5′-ATGACGCACAATCCCACTATCCT-3′ and pBr_hygR4 5′-AGTTCGGTTTCAGGCAGGTCTT-3′, and amplified fragment of 405 bp.

Putative transgenic T₀ plants were tested with each of the two pairs of specific primers. In case if one PCR test was positive and the second one was negative the third one was applied with one of these two pairs of primers. Only PCR positive (PCR+) plants were grown for further testing. The same PCR tests were applied for eight randomly selected T₂ progeny from one T₁ transgenic parent. Segregation rates were estimated based on PCR+/PCR- plants. Non-transgenic, in vitro plants as well as null segregants were used as a control.

Total RNA was isolated from immature kernels at 26 DAP (days after pollination) using the modified SDS extraction method of Prescott and Martin (1986). An additional extraction step was performed using TRI-Reagent (Sigma-Aldrich) according to the manufacturer’s protocol. The RNA samples were digested with DNAse I recombinant (Roche). The first-strand cDNA was synthesized from 1 μg of RNA using the RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific) according to the manufacturer’s instructions.

The relative expression level of Pin genes in wheat and Sin genes in triticale was analyzed in 26 DAP samples of transgenic and control plants. The sequences of the primers were as follows: for Pina and Sina, qPinA1 (forward) 5′-CTCATAGGACTGCTTGCTCTGGTAG-3′, qPinA2 (reverse) 5′-GATTGACCCCTGGATGATGTTG-3′; for Pinb and Sinb, qPinB1 (forward) 5′-AATGAAGTTGGCGGAGGAGGTG-3′, qPinB2 (reverse) 5′-ATACCTCACCTCGCCAAATGCC-3′; and for 18S rRNA, 18S F (forward) 5′-GTGACGGGTGACGGAGAATT-3′, 18S R (reverse) 5′-GACACTAATGCGCCCGGTAT-3′. The reaction was carried out in a 15 μl mixture containing 1× Hot FIREPol EvaGreen qPCR Mix (Solis BioDyne), 0.4 μM of each primer, and 1 μl of the template cDNA. The following temperature profile was used: an initial denaturation step of 95°C for 15 min, 50 cycles of amplification at 95°C for 20 s, 58°C for 15 s, and 72°C for 20 s, and a melting curve profile of 72–95°C with the temperature rising 1°C at each step and continuous fluorescence measurement. Three replicates of each sample were used in the reaction. The relative expression level of Pin and Sin genes was calculated by the standard curve method using 18S rRNA as a normalizer. The non-transgenic control plant was used as a calibrator sample with the expression level set to 1. The normalized values of the tested samples are expressed as x-fold of 1.

Low molecular weight RNA containing the microRNA fraction was separated from total RNA using polyethylene glycol (PEG) as described by Goto et al. (2003). The artificial Pinb-derived miR was detected by stem-loop RT-PCR miRNA assay as described by Varkonyi-Gasic et al. (2007). Two hundred and fifty nanograms of RNA from each sample and 0.05 μM of RT primer were used for the reverse transcription reaction with Maxima reverse transcriptase (Thermo Scientific). The loop region of RT primers contained a sequence complementary to UPL probe #21 (highlighted in bold). Primers for endogenous Tae-miR164 were used in the positive control sample. The sequences for RT and qPCR primers were: for ami-Pb: RT primer 5′-GTTGGCTCTGGTGCAGGGTCCGAGGTATTCGCACCAGAGCCAACGCTAGG-3′ and forward primer 5′- GGTGTTTGAATACCTCACCTAGC -3′ and reverse primer 5′-GTGCAGGGTCCGAGGT-3′; for Tae-miR164: RT primer 5′-GTTGGCTCTGGTGCAGGGTCCGAGGTATTCGCACCAGAGCCAACTGCACG-3′, forward primer 5′-TGGAGAAGCAGGGCACGTGCA-3′ and reverse primer 5′-GTGCAGGGTCCGAGGT-3′; and for U6 RNA: forward primer 5′-CTTCGGGGACATCCGATAAA-3′ and reverse primer 5′-GACCATTTCTCGATTTGTGC-3′.

Samples were first incubated at 65°C for 5 min, and the reverse transcription reaction was performed using the following program: 94°C for 2 min followed by 40 cycles of 94°C for 15 s and 60°C for 1 min. The qPCR reaction was carried out in a 20 μl reaction mixture containing 10 μl of TaqMan Fast Advanced Master Mix (Applied Biosystems), 2 μl of cDNA sample, 0.4 μM of each primer and 0.25 μl of UPL probe #21 (Roche). The reaction was run in a RotorGene Q thermal cycler (Qiagen) using the following program: initial denaturation step at 95°C for 2 min followed by 55 cycles of amplification at 95°C for 5 s, 60°C for 15 s, and 72°C for 5 s. U6 RNA was used as a reference gene. A series of dilutions of amplified amiRNA and U6 RNA was used in the first qPCR reaction to estimate the reaction efficiency. The relative level of amiRNA was calculated using the Pfaffl method (Pfaffl, 2001). The sample with the highest Ct value was used as a calibrator. The level of amiR was determined as x-fold higher compared to the calibrator sample, which was the sample with the lowest detectable amiR level set to 1.0.

Friabilin (complex of puroindoline a and puroindoline b) was extracted from the surfaces of the starch granules using the method of Bettge et al. (1995) with some modifications. To obtain 50 mg of wheat flour, 100–150 mg of kernels were ground in a mortar. After being transferred to a 2 ml microfuge tube containing 0.6 ml of 0.1 M NaCl, the samples were incubated at room temperature for 30 min in a thermoshaker. The aqueous starch suspension was transferred to a new 2 ml microfuge tube and centrifuged at 15 000 × g for 1 min. The pellet was washed with 1 ml of water. Using a pipette tip, the gluten and bran residues were formed into a ball and removed. The starch pellet was resuspended in 1 ml of 80% CsCl and centrifuged at 15 000 × g for 1 min. The CsCl solution was then removed from the tube, and the sample was washed three times with 1 ml of water and once with 0.6 ml of acetone. The pellet was left to dry and the tubes were weighed to determine the amount of starch. Next, the starch pellets were suspended in 150 μl of 1 M NaCl and 150 μl of isopropanol. After incubation for 45 min at 45°C, the samples were centrifuged at 15 000 × g for 1 min. The supernatant was transferred to fresh microfuge tubes and 150 μl of cold acetone was added, and samples were stored overnight at -20°C. Next, the samples were centrifuged at 15 000 × g for 1 min and transferred to fresh microfuge tubes. This fraction was precipitated for 2 h with 1 ml of cold acetone at –20°C and centrifuged at 15 000 × g for 10 min. The resulting protein pellet was resuspended in 30 μl of water, 3 μl of BME and 30 μl of SDS-PAGE sample buffer (Laemmli, 1970), mixed, and incubated at 70°C for 10 min.

The SDS-PAGE buffers and gels were prepared according to Laemmli (1970). The stacking gels were 5% T, 2.6% C, and the resolving gels were 15% T, 2.6% C 1.5 mm thick. 18 cm×16 cm gels were run in Hoeffer SE 660 apparatus at 20 mA until the dye reached the bottom edge of the gel. After electrophoresis, the gels were silver stained using the procedure of Gromova and Celis (2006).

The grain hardness, weight, moisture, and diameter of 50 kernels were measured using the Single Kernel Characterization System (SKCS) 4100 (Perten Instruments) for each individual kernel.

The total grain protein content was determined by the Dumas method (Dumas, 1831). The samples of 250 mg of ground seeds in three replicates were combusted at 960°C in the presence of oxygen, followed by afterburning at 800°C, and dried with P₂O₅. The resultant nitrogen oxide was reduced to nitrogen. The content of nitrogen in the probes was recounted for crude protein using Rapid N Cube (Elementar, Germany). The formula N × 6.25 was applied.

Statistical analysis was performed using MS Excel and R studio³. Obtained data were analyzed by Pearson correlation or one way analysis of variance. Samples that did not meet the ANOVA assumptions were tested by Kruskal–Wallis test.

Artificial microRNA for silencing of Pina or Sina genes (Pina-derived amiR) and Pinb or Sinb genes (Pinb-derived amiR) was constructed on the base on precursor miRNA of wheat Tae-miR164 (Mirbase.org: MI0006173) (Figure 1A). The final sequence of Pina-derived amiR and Pinb-derived amiR cassettes with the highlighted 21 nt precursors is presented in Figure 1B. These specific precursors matched the Pina and Pinb or Sina and Sinb sequences as shown in Figure 1C.

Potentially transgenic plants were selected on hygromycin. The transgenic character of these plants was proven by PCR amplification of T-DNA with two pairs of specific primers.

Transgenic plants were obtained after transformation of two cultivars of wheat (cv. Kontesa and cv. Torka) with the Pinb-derived amiR vector. Transformation efficiency was 2.97 and 0.09%, respectively (Table 1). In contrast, failure to obtain Pina-derived amiR transformed plants after inoculation of 2753 immature embryos of three cultivars of wheat in six experiments was observed.

Similar results were obtained for triticale cv. Wanad. Transformation efficiency was high (3.78%; ±0.32) with the Pinb-derived amiR cassette and very low (0.07%; ±0.04%) with the Pina-derived amiR cassette after inoculation of 4287 immature embryos in five independent experiments.

Segregation ratio of PCR positive (PCR+) to PCR negative (PCR-) plants in 17 T₁ lines of wheat and 12 T₁ lines of triticale indicated that 2/3 of these lines were one-locus transgenics and the remaining 1/3 were two-loci transgenics. Relative expression of both Pina and Pinb or Sina and Sinb in 21 DAP kernels silenced with Pinb-derived amiR was estimated in 42 transgenic T₁ plants from 13 T₀ lines of wheat cv. Kontesa and 18 transgenic T₁ plants from 12 T₀ lines of triticale cv. Wanad (Figures 2A,B). Ten of 13 tested wheat lines were one-locus and three lines were two-loci transgenics. In case of triticale 10 lines were one-locus and two lines were two-loci transgenics. Relative expression of Pinb in T₁ wheat plants, ranging from 0.06 up to 1.59, was below 0.2 in 11 plants (26%) from seven lines and between 0.2 to 0.5 in 20 plants (48%) from 10 lines. All plants with relative expression equal or lower then 0.8 compared to control set to 1.0 (±0.19) are considered as “silenced,” while relative expression over 1.2 indicated overexpression.

Similar results of downregulation in the Pinb-derived amiR transgenic plants were obtained for the second, Pina gene. The same 31 (74%) of T₁ plants showing relative Pinb silencing below 0.5 also showed Pina silencing at a corresponding or even lower level (Figure 2A). The correlation coefficient between silencing of Pinb and Pina in T₁ wheat Pinb-derived amiR transgenics was high and statistically significant (ρ = 0.71; α = 0.01).

These data of Pinb (and Pina) gene(s) silencing correlated with expression of amiR specific to the Pinb gene. The level of Pinb-derived amiR in seven plants with strong silencing of both Pina and Pinb was from 0.3 × 10⁶ to 68 × 10⁶ fold higher compared to the calibrator sample (i.e., the sample with the lowest detectable amiR level set to 1.0). In another set of seven plants with weaker silencing the values, ranging between 3 × 10² and 7 × 10², were from three to five orders of magnitude lower. The highest level of expression of amiR was found in one-locus transgenic plants and lower level was in two-loci plants.

Relative expression of Sinb in T₁ triticale plants ranged from 0.02 up to 2.04 (Figure 2B). Five of them showed very low expression of Sinb, below 0.2, with a corresponding low expression of Sina. Relative expression of Sinb for half of the T₁ triticale plants ranged from 0.46 to 0.92, but the corresponding Sina expression for most of them was above 1.0. Correlation between the expression of Sinb and Sina in T₁ was statistically significant (r = 0.76; α = 0.01). The expression of Pinb and Pina, and Sinb and Sina, was not decreased in very few T₁ plants of both Triticeae species.

The transgenic character of all tested T₂ plants was proved by PCR analysis with specific primers. Relative expression of Pinb in T₂ progeny of wheat cv. Kontesa ranged from 0.08 to 0.98 and for Pina from 0.15 to 2.16 (Figure 2C). About half of the plants out of 23 tested showed relative Pinb expression below 0.5. The Pina expression in the same plants was considerably higher. Relative expression for Pina in plants with 0.5–1.0 values of Pinb silencing was very unstable and in most of the progeny ranged from 1.0 to 2.2. The correlation between Pinb and Pina expression in T₂ (Figure 2C) was strong, positive, and statistically significant (r = 0.81; α = 0.01).

The relative expression of both genes silent in eight tested T₁ plants of wheat was much higher in their T₂ progeny comparing to T₁ (Figure 3A). The significance of difference was confirmed by Kruskall–Wallis test (p = 2.41e-05, α = 0.05). The mean expression value for Pinb in T₁ plants was 0.26 (±0.07) and for Pina was 0.21 (±0.04). The mean values for both Pinb and Pina genes in T₂ progeny were respectively 0.48 (±0.07) and 0.87 (±0.12). The values in the T₂ progeny were higher and more variable for Pina compared to the silenced Pinb. The mean relative expression for Pinb in T₂ was higher in individual lines as well as in all lines compared with T₁. The values for Pinb were less variable than for Pina in the T₂ generation. The data of very weak or lack of Pinb and Pina silencing in T₂ were correlated with the level of significantly weak or lack of Pinb-derived amiR, which ranged from 0 to 3.55 × 10³ (compared to 0.3 × 10⁶ to 68 × 10⁶ in T₁).

Relative expression of Sinb in T₂ progeny of triticale cv. Wanad ranged from 0.12 to 1.41 and for Sina from 0.11 to 1.38 (Figure 2D). About ¼ ; of the progeny out of 36 tested showed relative Sinb silencing up to 0.5. For half of the progeny Sinb relative expression was from 0.5 to 1.0 and for the rest above 1. The correlation between Sinb and Sina expression in T₂ (Figure 2D) was positive and statistically significant (r = -0.58; α = 0.01). The relative expression of Sinb and Sina in the T₂ progeny of seven silenced in T₁ transgenic plants of triticale increased significantly (Kruskall–Wallis: p = 4.844e-05, α = 0.05) (Figure 3B). The mean value for Sinb in T₁ plants was 0.11 (±0.02) and for Sina was 0.15 (±0.06). The mean values for both Sinb and Sina genes in the T₂ progeny were respectively 0.63 (±0.08) and 0.72 (±0.06). Otherwise these values were at a similar level (around 0.6–1.0) in T₂ progeny of non-silenced or weakly silenced T₁ plants. As in the case of Pinb and Pina in wheat, the values of very weak or lack of Sinb and Sina silencing in T₂ correlated with very weak or lack of relative expression of Pinb-derived amiR (from 0 to 13-fold compared to calibrator sample).

The data of very low in T₁ or similar to control in T₂ transcript level of the Pina/Pinb genes in wheat and the Sina/Sinb genes in triticale were correlated with very low or similar to control level of puroindoline a/puroindoline b or secaloindoline a/secaloindoline b (Figures 4A,B).

There were two main groups of T₁ plants with silent Pin and Sin genes in both tested Triticeae species. In the first one, the significantly lower expression of Pin genes in wheat cv. Kontesa and Sinb gene in triticale cv. Wanad were associated with increased grain hardness index, measured by the SKCS (Kruskall–Wallis for Pina/SKCS: p = 2.287e-08, α = 0.05; for Pinb/SKCS: p = 5.422e-09, α = 0.05; for Sina/SKCS: p = 0.1136, α = 0.05; and for Sinb/SKCS: p = 0.0009204, α = 0.05). In the second group, the relative values of SKCS in the T₁ silent transgenics were not changed in wheat cv. Kontesa or were slightly lower in triticale cv. Wanad (Figure 5). Correlation coefficients between these two groups of wheat were 0.289 and 0.294 and for the two groups of triticale were 0.522 and 0.517 and were not significant.

The mean SKCS of the grains collected from all tested 46 lines of the T₁ generation of wheat was 8% higher (not significant as confirmed by ANOVA: p = 0.109684, α = 0.05) than in control plants, and the highest value was 124% of the control (Table 2). The mean SKCS in the grains from 28 tested T₁ lines of triticale was 30% higher than in the control (not significant as confirmed by Kruskall–Wallis: p = 0.2159, α = 0.05). The highest SKCS was 260% of the control.

The mean SKCS in the grains collected from all tested 46 transgenic lines of the T₂ generation of wheat was 5% higher than in control lines, and from 28 tested T₂ lines of triticale was 9% higher than in control.

Mean total protein content in the same grains used for SKCS from silent 46 T₁ lines of wheat cv. Kontesa was 14.65% (±0.23) and was significantly lower than in the control (16.40% ± 0.35), which was confirmed by ANOVA (p = 0.037; α = 0.05) (Table 3). Opposite data were obtained for triticale. Mean total protein content in the grains from 28 silent T₁ was 13.74% (±0.48) and was higher than in control (12.81% ± 0.30), but statistically not significant.

There was no correlation between grain hardness index and total protein in the T₂ grains from T₁ plants with silenced Pina and Pinb (r = -0.05848) (Figure 6). However, a strong positive, statistically significant correlation between hardness index and total protein in triticale transgenics was observed (r = 0.68925; α = 0.05). Total protein content in triticale cv. Wanad was increasing together with the increasing values of grain hardness.