Plant material and retting
Flax seeds (Linum usitatissimum L., fibrous cultivar Nike) were obtained from the flax and hemp collection of the Institute of Natural Fibres of Poland. Fourth-generation plants of the transgenic line and control flax were field-cultivated in Wroclaw on a semi-technical scale during the 2010 growing season. The flax was harvested on the 107th day of its growth. The field-grown plants were retted using the dew method, in which the plants are spread out in a field and left for at least 40 days. During this process, bacteria and fungi degrade the cell wall polysaccharides and middle lamella releasing the fibres from the stems . The quantity of obtained fibres from whole straw is presented as % fibres in the straw (Figure 1).
The plants were transformed using the plasmid pGAglubsens, containing cDNA encoding β-1,3-glucanase from potato (GenBank: AJ586575.1) under a 35S CaMV promoter and Nos terminator. Transgenic plants were pre-selected using PCR and selected using Western blot analysis. Three lines of transgenic plants were selected (B10, B11, B14) for in vitro testing. The B14 line was used for the field trials because it had the best productivity. None of the transgenic lines showed changes in plant height. The transgenic line B14 produced slightly more seeds than the non-transformed plants, while the transgenic lines B10 and B11 showed significant decreases in the yield of seeds .
Infra-red spectrophotometry analysis
Infra-red spectrometry was used to determine the chemical composition and molecular structure of the fibres from the transgenic and control flax plants. The spectra were measured at room temperature using a Biorad 575C FT-IR spectrometer. Data were collected over a spectral range from 50 to 4000 cm-1 with a resolution of 2 cm-1. In the mid infra-red part of this range, samples were prepared in a KBr pellet. In the far infra-red part of this range, samples were suspended in Nujol.
Mechanical analysis of fibre
Tensile tests of the retted fibres were conducted by means of a computer-driven Instron system (model 4452, High Wycombe, UK). The finest axially uniform 2.5- to 4.0-cm long filaments were carefully extracted by hand from the fibre bundles. The ends of the samples were glued and sandwiched between small plastic sheets using a cyanoacrylate adhesive and were pinched within serrated grips connected both to a load cell of 10 N capacity and to the immovable part of the testing machine. The adhesive was spread out in a very thin layer and allowed to partially dry and solidify before the fibre ends were attached to the plastic pieces to avoid undesirable impregnation of the free filament part with cyanoacrylate. Samples of about 10-mm gauge length were extended at a crosshead speed of 1 mm/min. The exact gauge length was determined for each fibre to an accuracy of 0.01 mm at a tensile load of 5 cN. The load displacement curve was recorded and used to evaluate the fibre tensile parameters with Bluehill 2 Software (Instron Co.).
Stiffness, measured as Young’ modulus, was calculated from the slope of a linear part of the load displacement curve (ΔF/Δx), using the formula:
where L is the gauge length (corrected for each sample using Instron readings) and Acw is the effective cell wall cross-sectional area.
The maximum recorded load (Fmax
) per initial cross-sectional area of cell wall material (Acw) was used as a tensile strength measure:
The cross-sectional area of the cell wall material was evaluated using a gravimetric method [49
] and the formula:
where m is the weight of the fibre gauge length and dcw is the cell wall material density, assumed to equal 1540 kg m-3.
The energy to break (determined as the strain energy to maximum load) per unit mass of fibre sample and the relative extension at break were determined using Bluehill 2 software.
The sample weight within the gauge length, needed for the estimation of the cross-sectional area of the cell wall material, was determined soon after tensile measurements were completed, to an accuracy of 1 μg, using a highly precise microbalance (XP6, Mettler, Toledo).
The cellulose content was determined using the colourimetric method with anthrone reagent, as described by Ververis . 15 mg dry, ground flax fibres were incubated with a mixture of nitric and acetic acid (1:8 v/v) for 1 h at 100°C and then centrifuged (5 min, 14000 rpm). The pellet was washed twice with water and then resuspended in 1 ml 67% H2SO4 (v/v). After mixing samples, cold anthrone reagent was added and the cellulose level in these samples was measured spectrophotometrically at 620 nm. Commercially available cellulose after hydrolysis was used for the calibration curve.
The determination of the total lignin content was performed using the acetyl bromide method, as described Iiyama and Wallis . 15 mg dry, ground flax fibres were heated for 2 h at 100°C, then 10 ml water was added to each sample, and the samples were heated for 1 h at 65°C with mixing every 10 min. Then the samples were filtered through a GF/A glass fibre filter and rinsed three times with each of the following solutions: water, ethanol, acetone and diethyl ether. The filters were placed in glass vials and heated overnight at 70°C. After that, 25% acetyl bromide (2.5 ml) in acetic acid was added and the vials were placed at 50°C for 2 h. The cooled samples were mixed with 10 ml of 2 N sodium hydroxide and 12 ml of acetic acid. After incubating in RT overnight, the lignin content was measured at 280 nm. Coniferyl alcohol was used to prepare a calibration curve.
Isolation and fractionation of the cell wall polysaccharides
The isolation and fractionation of the cell wall components was performed using a modified version of the method described by Manganaris  and Vincente .
Fibres from transgenic and non-transgenic flax (1 g dry, ground plant tissue) were boiled in 96% ethanol for 30 min to inactivate the enzymes, extract the low molecular weight components and prevent autolysis. The material was filtered with a Whatman GF/C filter and then sequentially washed with 80% ethanol, chloroform:methanol (1:1 v/v) and acetone, and allowed to dry at 37°C to yield an alcohol-insoluble residue (AIR).
All the AIR obtained from each sample was suspended in 20 ml of water and then stirred at RT for 12 h. After the centrifugation (6000 × g, 4°C, 10 min) the pellet was washed with water and both supernatants were collected for water-soluble fraction (WSF) analysis. The remaining material was resuspended in 50 mM CDTA (trans-1,2-diaminocyclohexane-N,N,N,N-tetraacetic acid) at pH 6.5 and stirred (RT, 12 h). After the centrifugation and wash (as above), the extracted solutions were collected and designated the CDTA-soluble fraction (CSF). The pellet was resuspended in 50 mM Na2CO3 with 20 mM NaBH4, stirred at 4°C for 12 h and washed, and then supernatants were neutralised with glacial acetic acid. These samples were designed the Na2CO3-soluble fraction (NSF). The remaining pelleted material was resuspended in 1 M KOH with 20 mM NaBH4, stirred at RT for 12 h and washed, and then supernatants were neutralised with HCl to yield the 1 M KOH-soluble fraction (K1SF). The same activity was performed with 4 M KOH to obtain the 4 M KOH-soluble fraction (K4SF). Supernatants from the CSF, NSF, K1SF and K4SF were extensively dialysed against water (with a 3.5-kDa cut off) and all of the fractions were additionally lyophilised before use.
Uronic acid measurement
The content of pectin was determined using the biphenyl method  after hydrolysis of the polysaccharides in sulphuric acid . The samples were suspended in 0.1 ml sulphuric acid and stirred in an ice bath for 5 min. Sequentially, 0.1 ml sulphuric acid, 0.05 ml water, 0.05 ml water and 0.7 ml water were added, with stirring between additions. The diluted material was centrifuged for 10 min at 2000 × g at RT, and 0.1 ml of the supernatant was taken and added to a 10-μl 4 M sulphamic acid/potassium sulphamate solution at pH 1.6. Then 0.6 ml of 75 mM Na2B4O7 in sulphuric acid was added for the reaction. The samples were shaken and incubated at 100°C for 20 min. After cooling, 20 μl of m-hydroxy-biphenyl (0.15%) in 0.5% NaOH was added to each sample, and they were incubated at RT for 10 min. The pectin content was measured with a spectrophotometer at 525 nm. Galacturonic acid was used for the calibration curve.
Monosaccharide identification by UPLC
The hydrolysis of polysaccharides, derivatization procedure and UPLC analysis were performed using a modified version of the method describe by Lv  and Yang . The lyophilized tissue samples (10 mg) were hydrolysed with 4 M TFA (500 ml) for 8 h at 110°C, and then cooled and centrifuged (5 min, 1000 rpm, RT). The pellet was discarded and the supernatant was dried under nitrogen and then dissolved in distilled water (1 ml).
In order to determine their monosaccharide composition, the samples (50 μl) were derivatised by incubation for 60 min at 70°C with 0.3 M NaOH (50 μl) and 0.5 M PMP in methanol (50 μl). After cooling, 0.3 M HCl (50 μl) was added to each sample and they were washed three times with chloroform. Each sample was filtered before UPLC analysis.
The samples were analysed on a Waters Acquity UPLC system with a 2996 PDA detector, using Acquity UPLC column BEH C18, 1.0 × 100 mm, 1.7 μm. The mobile phase was A = 50 mM CH3COONa, pH 6.3 with 0.04% TEA and B = acetonitrile with 0.04% TEA, in a gradient flow: 1 min at 96% A/4% B; 5–11 min gradient to 89% A/11% B; 12–13 min gradient to 0% A/100% B; and 14 min gradient to 96% A/4% B with a 0.05-ml/min flow rate. The content was measured at 250 nm.
Total pectin and total hemicellulose contents
The content of total pectin was estimated as the sum of the uronic acids and other monosaccharides from the three pectin fractions (WSF, CSF and NSF).
The total hemicellulose content was calculated in the same way for the other two fractions (K1SF and K4SF).
The determination of the callose content in the flax fibres was performed using a modification of a method described by Hirano . 20 mg dry, ground flax fibres were washed once with 96% ethanol and three times with 20% ethanol. Then, 1 ml of 1 M NaOH was added to the washed tissue, and to solubilise the callose, the tubes were heated at 80°C for 15 min. After the centrifugation (15 min, 10000×g), the supernatant was ready for the callose determination. 0.2 ml of supernatant, 0.4 ml of 0.1% (w/v) aniline blue, 0.21 ml of 1 M HCl and 0.59 ml of 1 M glycine-NaOH buffer (pH 9.5) were mixed and incubated for 20 min at 50°C, and then 30 min at room temperature. The callose content was quantified spectrofluorometrically at excitation and emission wavelengths of 393 and 484 nm, respectively. Curdlan (β-1,3-glucan) was used to prepare a calibration curve.
Phenolic compound extraction and measurement by UPLC
The lyophilized samples (20–100 mg) from each fraction of the cell wall were extracted three times with methanol using an ultrasonic bath (15 min). After centrifugation (5 min, 5000 rpm, RT), the supernatant was collected and the pellet was hydrolysed using 2 M NaOH (2 ml) in the dark. Then, the pH was adjusted to 3.0, and the samples were extracted three times with ethyl acetate and centrifuged (1 min, 5000 rpm, RT). The supernatant was dried, the pellet was resuspended in methanol (0.2 ml) and the samples were analysed on a Waters Acquity UPLC system with a 2996 PDA detector, using an Acquity UPLC column BEH C18, 2.1 × 100 mm, 1.7 μm. The mobile phase was A = 0.1% formic acid and B = acetonitrile, in a gradient flow: 1 min at 95% A/5% B; 12 min gradient to 70% A/30% B; 15 min gradient to 0% A/100% B; and 17 min 95% A/5% B with a 0.1 ml/min flow rate. The detection of coumaric and ferulic acid, syringic aldehyde, vanillin and vitexin was done at 320 nm and that of vanillic acid at 280 nm.
The antioxidant activity was assessed as described by Brand-Williams with some modifications . 10 mg of lyophilised samples from each cell wall fraction were resuspended in 1 ml of water. In a similar way, standards of pectin with different degrees of methylation were prepared. A 1 ml solution of 0.1 mM DPPH (2,2-diphenyl-1-picrylhydrazyl) in water:methanol (1:1) was mixed with 50 μl of sample. After 6 h, the absorbance was measured at 515 nm. Ferulic acid was used as a positive control.
The inhibition of DPPH• radicals of the cell wall fraction was calculated according to the equation:
All of the experiments were independently repeated at least three times. The results are presented as the averages of independent replicates ± standard deviations. Statistical analyses were performed using Statistica 7 software (Statsoft, USA). The significance of the differences between the means was determined using Student’s t test.