Full Text - Elevating the expression level of biologically active recombinant human alpha 1-antitrypsin in Pichia pastoris

Discrimination of population of recombinant inbred lines of rye (Secale cereale L.) for different responses to nitrogen-potassium stress assessed at the seedling stage under in vitro conditions

¹West Pomeranian University of Technology, Department of Plant Genetics, Breeding and Biotechnology, Słowackiego, Szczecin, Poland

Financial support: This work was supported by the Polish Ministry of Science and Higher Education under grant, no. N N302 281936.

Background: Plants differ in the methods used to acquire nutrients from environments with low nutrient availability, and may change the morphology of their ‘root architecture’ to be able to take up nutrients.

Results: In the present study rye response to stress caused by high and low nitrogen-potassium treatments in mature embryos cultures was described within a population consisting of one hundred and thirty eight recombinant inbred lines of rye. Characterization of the response of recombinant inbred lines (RILs) to nutrient stress was presented as the results of analyses of morphological traits, and physiological and biochemical parameters of the seedlings grown in both treatments. A wide range of variability of individual RILs to induced stress was observed in the population of recombinant inbred lines, and was presented as the difference between the means of each of the analysed traits described at high- and low-nitrogen-potassium levels. Lines were grouped using Ward's agglomerative method on the basis of differences in coleoptyle length, with the longest root length and root number used as variables.

Conclusions: Recombinant inbred lines at low nitrogen-potassium treatment developed: longer, shorter, or roots of similar length in comparison with the high nitrogen-potassium treatment. Discriminant function analysis showed that the discriminant variable able to clearly differentiate recombinant inbred lines in terms of their response to nutrient stress was the trait of the longest root length.

Rye (Secale cereale L.) is an interesting source of genes determining tolerance to edaphic stresses. It has the highest tolerance to drought or nitrogen deficiency, and is more productive than other cereals growing in infertile, sandy, or acid soils, as well as in the poorly-prepared land where it is more often grown (Cakmak et al. 1997; Geiger and Miedaner, 2009; Masojć and Kosmala, 2012).

About 60% of cultivated soils worldwide have plant-growth-limiting problems caused by mineral nutrient deficiencies and toxicity (Lynch, 2007; Liu et al. 2008; Zhang et al. 2010). This situation occurs in Poland, where, for example ca. 60% of the crop production area consists of light-textured or acid soils (Górny and Szołkowska, 1996).

Nitrogen (N) limitation is associated with decreased activity of enzymes that are required for energy metabolism, such as photosynthesis and respiration (Marschner, 1995). Potassium (K) deficiency leads to growth arrest and to impaired sugar and nitrogen balance through inhibition of protein synthesis and long-distance transport (Marschner, 1995; Epstein and Bloom, 2005).

Nitrogen-Potassium interaction (N x K) exists at the morphological, physiological, biochemical and genetically stages during the whole plant growth and generally exists in agricultural ecosystems. K improves the use of N, results in higher yield and quality (Rengel, 2005). K ensures utilization of N and storage of carbohydrates in roots, thus improving NUE (Nutrient-Use Efficiency) (Zhang et al. 2010). Currently, this interaction is a subject of interest in many studies and reviews (Shin et al. 2005; Zhang et al. 2010; Postma and Lynch, 2011; Tsay et al. 2011).

Plants have evolved multiple mechanisms to maintain nutritional homeostasis in diverse edaphic environments (Bloom et al. 1985; Van der Ploeg et al. 1999). Some of these responses can be genetically simple, but most traits (e.g. root architecture) are genetically complex (Liu et al. 2008).

A morphological change in root elongation and an increase in root number in response to different nutrient status is a ‘common’ property of many plant species (Zhang and Forde, 1998; Walch-Liu et al. 2006, Zhang et al. 2007; Ruffel et al. 2011). It results from, among other things, the genetic mechanism in plant roots that senses the presence of NO₃^- ions participating as a signalling factor in the signalling pathway of root elongation (Zhang and Forde, 1998). A significant role is played here by the products of the ANR1 gene (Transcription factors, TFs) and the AXR4 gene (Zhang et al. 1999). It is still unknown how NO₃^- affects ANR1, how AXR4 acts downstream of ANR1, and how the signal is transferred to the genes involved in the process of cell proliferation (Zhang et al. 1999). Remans et al. (2006) suggested that the AtNTR1.1 gene may act upstream of ANR1 and could be one of the elements of the transcriptional regulation of ANR1, functioning as the nitrate sensor. Gan et al. (2005) identified several genes, out of which seven of the MADS-box genes presented similar expression patterns to ANR1, suggesting the existence of undiscovered genes involved in the response to localized nitrate (Zhang et al. 2007).

The application of biotechnology methods in the identification of nutrient-efficient genotypes has an exciting potential. On the one hand, they constitute an alternative method, compared to root evaluation of field-grown plants implemented under controlled conditions (e.g. hydroponics, pots or in vitro) at an early stage of plant growth (Rzepka-Plevneš et al. 1997a; Sanguineti et al. 2006; Tuberosa et al. 2011; Lynch, 2007). On the other hand, successful regeneration of plants selected under in vitro conditions form cell/callus lines proves the usefulness of such methods (Rai et al. 2011).

For species of cultivated plants, inbred lines have a wide range of genotypic variability regarding growth at lower mineral concentrations, and selection of tolerant genotypes is carried out depending on the root system (Lynch, 2007; Tuberosa and Salvi, 2007; Löschenberger et al. 2008; Messmer et al. 2011). Bearing in mind that the root is the initial site of nutrient perception, whereas plant root plasticity is a primary effect of the activity of the mechanisms sensing and optimizing consumption and utilization of water and nutrients, the possibility of describing the different responses within segregants in a mapping population enables the investigation of plant responses to low N, K or N-K deprivation; the inhibitory effect of nitrate on root development (Zhang and Forde, 1998; Van der Ploeg, 1999; Tsay et al. 2011; Ruffel et al. 2011); and the possible roles of auxin (Walch-Liu et al. 2006) or ABA (De Smet et al. 2006).

Thus, the aims of this study were to perform phenotype screening and discrimination of population of recombinant inbred lines of rye according to their morphological, physiological and biochemical responses to high- and low- N and K level in medium, to separate the population of RILs into groups differing significantly in terms of their response in order to prepare the tested material for research into the identification of molecular markers linked to QTLs of tolerance to induced stress with the use of the bulked segregant analysis (BSA) method.

One hundred and thirty-eight recombinant inbred lines (RILs - F₇) of rye (Secale cereale L.), deriving from a cross between inbred lines 153/79-1 and Ot1-3, were used in the study. In a preliminary study Rzepka-Plevneš et al. (1997a) have characterized line 153/79-1 as tolerant, whereas Ot1-3 as susceptible to nutrient stress in medium assessed at the seedling stage.

Plant embryos culture under in vitro conditions and nutrient treatments were establish according to the method described by Rzepka-Plevneš et al. (1997a), Rzepka-Plevneš et al. (1997b) and Rzepka-Plevneš (1999). Rye seeds were surface-sterilized by soaking them for 15 min respectively in 1% and then 7% solution of H₂SO₄ and NaOCl. Then they were washed three times for 10 min in distilled water and left in sterile water for 24 hrs. Mature embryos were prepared with the use of a preparation needle, sterilized in 10% NaOCl solution for 10 min, and then washed three times in sterile water and transferred onto appropriate media. Four embryos were placed in each glass tube (9 x 3.5 cm), containing 30 ml of proper media. The tubes were closed with aluminium foil and wrapped with parafilm, than were transferred for 11 days into a phytotron with temperature of 25ºC under cool white light with 16 hrs photoperiod (36 µmol·m^-2·s^-1) and 75% RH. The nutrient treatments were composed of MS medium (Murashige and Skoog, 1962). High nitrogen-potassium content (HNK) medium consisted of 6.00 mM N and 2.00 K and used as control, whereas low nitrogen-potassium (LNK) consisted of 0.334 mM of N and 0.333 mM of K. pH of the nutrient media was adjusted to 5.7 (Rzepka-Plevneš et al. 1997a; Rzepka-Plevneš et al. 1997b). A completely randomized design was used with replicates. One glass tube with four seedlings was adopted as one replicate. The glass tubes were evenly laid out in a phytotron with equal light availability. The parental lines and each of the one hundred thirty eight RILs were represented by ca. 150 embryos (ca. 40 glass tubes) and 50 and 100 embryos were placed in each of the media - HNK and LNK, respectively. The plants were harvested after 11 days when they had one (two) visible leaves. At this stage plants grown in LNK became yellow.

The seedlings were taken from the agar medium and then washed with distilled water. Biometric measurements, such as: CL (coleoptyle length (cm)), the longest root length (LRL (cm)), and root number (RN), were carried out. Seedlings of each of the tested RILs were taken from the HNK and LNK media respectively, then were pooled and stored at room temperature and -20ºC, respectively, and subjected to physiological and biochemical measurements, respectively. These parameters were assessed each in three replicates for each of the RILs.

After 11 days of treatment, seedling samples were taken for the assessment of chlorophyll concentration (mg g^-1 FW), according to Lichtenthaler and Wellburm (1983), and of carotenoid (Car) content, according to Hager and Meyer-Berthenrath (1966). Total chlorophyll [Chl (a + b)], chlorophyll a (Chl a), chlorophyll b (Chl b) and Car contents were determined. CO₂assimilation rate (A) and transpiration rate (E) were measured using a TPS-2 gas analyser by PP Systems (UK) according to the standard method described by Von Caemmerer and Farquhar (1981).

Total catalase (CAT, EC 1.11.1.6) activity was measured spectrophotometrically (UV/VIS Lambda Bio, Perkin-Elmer) according to the method by Lück (1965) by monitoring a decline in absorbance at λ = 240 nm for 60 sec as H₂O₂ was consumed. Total peroxidase (POX, EC 1.11.1.7) assay contained 20 mM pyrogallol, 1 mM H₂O₂, and 50 ml of plant extract in 50 mM potassium phosphate buffer (pH 7.0) in a total volume of 1 ml (Chance and Maehly, 1955). Purpurogallin formation was measured spectrophotometrically (Nova 400, Merck) at λ = 430 nm at 25ºC for 4 min. Total superoxide dismutase (SOD, EC 1.11.1.5) activity was analysed according to Abassi et al. (1998). Increasing volumes (25, 50, 100 and 200 µl) of leaf extracts were added to the reaction mixture to the final volume of 3 ml. The reaction mixture contained 50 mM potassium phosphate buffer (pH 7.8), 13 mM L-methionine, 0.1 mM EDTA, 2 μM riboflavin and 75 μM nitroblue tetrazolium (NBT).

The experiment was carried out in stages by characterization of 8-10 subsequent RILs in every stage in terms of all the mentioned parameters. In the case of contamination of rye embryo cultures, single contaminated glass tubes were removed. When more embryos were contaminated, the experiment was repeated for a given line.

Two-way analysis of variance was carried out for the traits of the parent lines, 138 RILs and treatments (HNK and LNK). Changes of root elongation under nutrient stress are a typical plant adaptation mechanism and depend on genotype and initiated defense strategy. Response of each of the one hundred thirty eight RILs and parent lines to nutrient stress caused by N and K deficiency has been presented as value of difference (d_) between means for the seedlings growing at HNK and LNK (Figure 1). Their significance was tested using the Student's t-test. For each RIL the value of difference (d_) between means was positive or negative, depending on examined trait and RIL response. The differences (d_) between the means for three seedling traits: d_CL, d_LRL and d_RN were used to carry out cluster analysis with the use of Ward's agglomerative method (Euclidean distances). Grouping variables (‘a’ and ‘b’) were assigned to the RILs grouped in the groups of Ward's dendrogram together with parental lines, while variable ‘c’ to other RILs. The groups were characterized according to investigated parameters. Discriminant function analysis (DFA) was performed in order to answer the question, whether other physiological or biochemical parameter, apart from seedling traits - i.e. CL, LRL or RN, are also important in discrimination of RILs' population. The parental lines were added to such proposed a posteriori division of the 138 RILs into three groups. The verification of the traits by forward selection method in the DFA was based on the significance of Wilk's λ test, approximate value of F test. The significance level corresponding to the F-value for entering or staying on a specific trait at each step was set at P = 0.20. All traits that remained in the model, when the stepwise selection process stopped, were considered to discriminate significantly between groups. Also, coefficients of canonical variables after their standardization were calculated, as well as accumulated percentage of explained variability in a given model. The assessment of model significance was verified with the use of χ² test. The ability of the created models to divide RILs into groups was illustrated on a chart of scatter of canonical values. The above-presented calculations were made using Statistica 9 software package (StatSoft PL).

A wide range of variability in the response of the examined rye genotypes to nutrient stress caused by the LNK treatment was found both for the parental lines and within the population of the 138 RILs (F₇). The mean values of seedling CL, LRL and RN, physiological (Chl a, b, a+b, a/b, Car, A, E) and biochemical (total CAT, SOD and POX) parameters assessed at the HNK and LNK treatments for both parental lines and 138 RILs are presented in Table 1 and Table 2.

It was concluded that inbred lines responded to induced stress (LNK) in different ways. Under HNK conditions they generally generated longer coleoptyles in comparison to LNK. Line 153/79-1 developed significantly longer LRL at LNK than at HNK. Line Ot1-3 responded in a different way, and the differences were described as significant (Table 1). For RN of line 153/79-1, no significant differences were noted between HNK and LNK. Differences were found for line Ot1-3 (Table 1). When other physiological parameters were considered, it was demonstrated that under HNK conditions, the parental lines generally had greater values of the assessed parameters in comparison with LNK. The differences were not always significant. They are presented in detail in Table 1. Line Ot1-3 had significantly lower biochemical parameters at LNK in comparison to HNK. A similar response was found for line 153/79-1, and a significant difference was described only for POX activity (Table 1).

Different seedling parameter values resulted from not only the experimental conditions but also from genotype. Line 153/79-1, as opposed to Ot1-3, was characterized by longer coleoptyle, longer LRL and more numerous roots (Table 1). In terms of both physiological and biochemical parameters, genotypic differences between the lines had different values. They are presented in detail, together with the significance of the differences in Table 1.

Among the RILs significant differences were found in all the examined seedling traits and in all physiological and biochemical parameters in both NK treatments (Table 2).

Variance analysis showed that the effects of NK treatments, genotype and NK x genotype interaction were highly significant for all the seedling traits, as well as for physiological and biochemical parameters (Table 2). In HNK and LNK treatments some RILs generally developed shorter coleoptyles, longer roots and a greater number of roots. Also, on average lower concentrations of Chl a, b and Car and lower A and E as well as decreased of CAT and POX activity and increased SOD activity were found (Table 2).

In both NK treatments highly significant correlations between CL, LRL and RN were found (Table 3). Similarly, significant and highly significant correlations were assessed between SOD and CL, and LRL and RN, respectively. Highly significant correlations between photosynthetic pigments, A and E, were also found in both NK treatments (Table 3). In HNK, a significant correlation between the activity of CAT and POX was observed, and this was not found for the LNK treatment (Table 3).

It has been demonstrated that RILs developed longer CL at HNK when compared to LNK, and some RILs developed significantly longer LRL and more numerous roots (RN) at HNK in comparison to LNK (Figure 1a). The other RILs developed significantly shorter LRL and less numerous roots (RN) at HNK in comparison to LNK (Figure 1b).

According to the calculations performed with the use the of Student t-test, some of the differences (d_) were significant, highly significant or non-significant, for which differences between means (d_) of individual traits were at a similar level (data not shown) (Figure 1c). The range of variability for individual parameters of RILs is presented in Table 4 as difference (d_) and standard deviation values (SD), and also as minimum and maximum values. It was demonstrated that at HNK the examined RILs on average developed longer coleoptyles (d_CL = 0.74), shorter LRL (d_LRL = -0.24) and a similar number of roots (d_RN = 0.01). For the other parameters mean differences (d_) for the whole population had positive values, which proved that a decrease in their values was observed for individual RILs. The SOD parameter was an exception - a greater activity of the enzyme was found at LNK in comparison to HNK (-3.03) (Table 4 and Figure 1).

Cluster analysis of the set of RILs and two parent lines based on the d_CL, d_LRL and d_RN indicated that all the genotypes were grouped in seven discrete groups (Figure 2).

Groups ‘a’ and ‘c’ consisted of eighteen RILs each. Groups ‘b’ and ‘d’ included twenty-three and twenty-five RILs, while groups ‘e’, ‘f’ and ‘g’ - twelve, twenty-one and twenty (Figure 3). In general, the RILs in groups ‘a’ and ‘b’ exhibited a significantly shorter CL, LRL and less RN in the LNK in comparison to the HNK, whereas the RILs in groups ‘f’ and ‘g’ were comprised of genotypes expressing a significantly longer CL, LRL and more RN in the LNK in comparison to the HNK (Figure 3). The characteristics of d_CL, d_RN, and especially of d_LRL between groups ‘a’ and ‘g’ showed the possibility of discrimination of the whole population according to their different responses to nutrient stress and their classification into extreme groups: susceptible (‘a’) and tolerant (‘g’), while lines with similar responses were in the groups - ‘b’ and ‘f’ (Figure 3a). Figure 3b and Figure 3c present the physiological and biochemical responses of the RILs, which were classified as shown in Figure 3a. Positive values of (d_) for individual physiological parameters, proving their decrease at the LNK in comparison with the HNK, were found in each of the groups. Assessment of CAT, POX and SOD activity in the individual groups demonstrated that differences between the means were positive and/or negative (Figure 4c). A decrease in the activity of CAT and SOD and a slight increase in POX were observed under the LNK treatment in group ‘a’, while in group ‘g’ a decrease in CAT and POX and a distinct increase in SOD activity were noted (Figure. 3c).

After application of forward stepwise DFA to the data (138 RILs x 13 variables), the variables presented in Table 5 were retained in the model. A highly significant statistical value of the Wilk’s λ test = 0.214; F = 16.530; P < 0.0000 F was demonstrated; hence the discrimination of RIL response to nutrient stress was highly significant. Nine variables (d_LRL, d_CAT, d_POX, d_Car, d_Chl b, d_SOD, d_A, d_Chl a and d_E) were used in the model of discriminant function analysis (Table 5).

The greatest observable contribution came from d_LRL (λ = 0.333, F = 128.020), followed by d_CAT (λ = 0.932, F = 4.650) and d_Car (λ = 0.937; F = 4.322). The other variables, despite their inclusion in the model, turned out to be non-significant in predicting affiliations to rye groups tolerant or susceptible to nutrient stress. Eigenvalues and the corresponding standardized canonical discriminant function coefficients are presented in Table 5. The first function (Root 1) generated the eigenvalue of 3.13, while the second one gave 0.131 (Root 2). The variance calculated for the first function, and the percentage of the total explained variance expressed in the first function was 95.98 (Table 5). The highest standardized discriminant coefficients in Root 1 originated from: d_LRL (1.004), d_Car (-0.624) and d_Chl b (0.316), whereas in Root 2 they were from: d_Car (-0.870), d_CAT (0.782), and d_Chl a (0.541). Both functions were statistically significant (Root 1: R-canon = 0.871; Wilk’s λ = 0.214; χ² = 203.59; P < 0.0000; Root 2: R-canon = 0.340; Wilk’s λ = 0.884; χ² = 16.26; P < 0.038); and the first function was responsible for ca. 96% of the explained variance. A two-dimensional scatter plot using the discriminant scores of the RILs along Roots 1 and 2 indicates a satisfactory separation of the recombinant inbred lines according to their response to nutrient stress assessed under in vitro conditions at the seedling stage (Figure 4).

Figure 4 presents the classification matrix of the RILs, from which it appears that 93.5% of the RILs were correctly classified into groups ‘a’, ‘b’ and ‘c’ with the use of the calculated classification functions.

They have also been used in many other experiments, including for describing the response of rice to drought (Joshi et al. 2011) or tolerance to saline and heat stresses in wheat (Benderradji et al. 2012). Rakoczy-Trojanowska and Malepszy (1995) found significant differences between the in vitro response of several inbred rye lines, both in the case of immature inflorescences and immature embryos.

Elicitation of the response potential per se of each of the 138 investigated recombinant inbred lines of rye to stress caused by nitrogen and potassium deficiencies was obtained by analysis of morphological changes in CL, LRL and RN in the seedlings growing in the HNK and LNK treatments. The state, in which the examined rye seedlings generally developed shorter CL in the LNK treatment in comparison with the HNK, may be the result of changes in root-to-shoot ratio, and can be explained on the one hand by the functioning of the mechanism of root elongation described by Zhang et al. (2007), or on the other by allocation of assimilates from the shoot to the root, taking place in plants under nutrient stress (Bloom et al. 1985; Cai et al. 2012). According to Mi et al. (2007) N-efficient maize inbred lines allocate more N to root development, and can develop more root compared to inefficient genotypes.

It was demonstrated that CL such as LRL and RN - elements of ‘root architecture’, had different values in individual RILs. The traits depend on genotype (Messmer et al. 2011). Despite the fact that CL in the parental lines was not affected by the NK treatments, a significant influence of genotypic differences was found. In the inbred lines LRL was significantly affected by both genotype and the treatments. The seedlings of line 153/79-1 developed longer LRL at LNK than at HNK, while the converse occurred with seedlings of line Ot1-3 - they developed on average shorter LRL, and the differences were highly significant. Neither genotypes in the both parental lines nor the treatments for line 153/79-1 had any influence on the differences in RN. Line Ot1-3 developed a lower number of roots in the LNK (Table 1). On average longer CLs (6.47 cm) were noted for the whole population of the RILs tested in the HNK treatment in comparison with the LNK (5.42 cm). Longer LRLs were described for the RILs tested in the LNK (on average 3.43 cm) in comparison to the HNK (3.21 cm); similarly, a greater RN was noted in the RILs tested in the LNK. A similar response from maize seedlings affected by both low N and high N was presented by Wang et al. (2005) and Liu et al. (2008). Different responses from the seedlings of individual RILs compared to parental lines may be explained by genotypic differences and possible transgression, which was described by Górny (1995) for barley, by Górny and Szołkowska (1996) for oat and by Liu et al. (2008) for maize.

A morphological change of in root elongation and an increase in root number is a ‘common’ response of many plant species to different nutrient status (Zhang and Forde, 1998; Walch-Liu et al. 2006). It may be assumed that the RILs which developed shorter roots in the LNK treatment in comparison with the HNK, may have genes changing the mechanism of ‘root elongation’ by specific TFs ‘modulating’ or silencing the expression of e.g. ANR1 (Zhang and Forde, 1998; Walch-Liu et al. 2006). Inhibition of root elongation is in accordance with the ‘dormant strategy’ (Chun et al. 2005; Guo et al. 2005). A possible explanation for the observed reaction in the selected RILs may be the inhibition of their root elongation resulting from initiation of the mechanism sensing N nutritional status of a plant as extremely low. Such a response type was described as differentiating inbred maize lines by Chun et al. (2005) and Guo et al. (2005).

Walch-Liu et al. (2006) suggest that up- and down-regulation of the ANR1 gene under conditions of low and high N status (not necessarily of NO₃^-) can be viewed as a mechanism modulating the intensity of lateral root response to localized N supply. Authors state that the molecular mechanism of plant response to N stress may involve (apart from ANR1) dozens of genes of the MADS-box gene family (TFs).

The low concentration of nitrogen and potassium in the medium could have resulted in increased pressure under laboratory conditions for the genotypes responding to nutrient stress (-N-K) in different ways. Research on the transcriptome of model plants indicates that in general plant response to different abiotic factors, including N and K deficiencies, often has a common genetic background (Ruffel et al. 2011).

Cakmak and Engels (1999) and Zhang et al. (2010) state that very low rates of potash fertilizer application result in ineffective utilization of the applied nitrogen fertilizers. Shin et al. (2005) showed that changes in ROS concentrations in wild-type Arabidopsis and in a set of root hair mutants suggest that the hair root cells are important for the response to nitrogen and potassium starvation. K starvation induces the expression of AtNRT1.5 and increases the activity of the GS (glutamine synthetase)/GOGAT (glutamate synthase)/GDH (glutamate dehydrogenase) cycle in K-starved roots. The latest research has shown that kinase CIPK23 (CBL-interacting protein kinase 23) is involved in regulating both K⁺ and NO₃^- uptake (Tsay et al. 2011).

The response of the examined rye RILs to nutrient stress caused by different concentrations of N and K in HNK and LNK treatments is in accordance with the results of the study presented by Rzepka-Plevneš (1999) for open-pollinated, selected populations or breeding strains of rye. Rzepka-Plevneš (1999) conducted triple mass selection among seedlings of eleven rye populations and two strains, regenerated from mature embryos in vitro culture by assessment of CL, LRL and RN. The author distinguished genotypes tolerant to nutrient deficiencies in medium, which, as later demonstrated in the author's field experiment, did not differ significantly from the control in terms of the selected parameters of yield (Rzepka-Plevneš and Kulpa, 1999). This results support the opinion presented by Rai et al. (2011) on the usefulness of the method for preliminary evaluation/selection of response to abiotic stress of mature embryos culture conducted for different genotypes of rye and other cereals.

Significant correlations between RN and CL, RN and LRL, and CL and LRL were reported. The results are in accordance with the results presented for inbred lines of rye and their hybrids by Górny and Geiger (1982) and Rzepka-Plevneš (1999). Górny and Geiger (1982) presented (highly) significant correlations for inbred lines of rye and their hybrids between: root number (RN) and total root system length (TRSL); RN and total root system weight (TRSW); the longest root length (LRL) and TRSL; LRL and TSRW; and TRSL and TRSW. The authors reported high heritability coefficients for root traits of rye LRL (h² = 0.50) and shoot fresh weight (h² = 0.75). Similar results were obtained in the present study. Highly significant correlations were found for RN and CL, RN and LRL, and CL and LRL. In another paper Górny et al. (1982) demonstrated significant correlations between rye traits, such as: LRL, TRSL, TRSW and grain number (GN), 1000-grain weight (TGW), grain yield per spike (GYS), assessed for seedlings (under laboratory conditions) and adult plants (micro-plot data). Moreover, the authors showed highly significant correlations between LRL and plant height (0.41) and grain yield per ha (0.51), assessed under laboratory conditions, and adult plants assessed from yield trials, respectively.

A decrease in the concentration of Chl a and b, Car, assimilation (A) and tanspiration (E) rate in parental lines and among the population of 138 RILs examined in the LNK in comparison with the HNK is a typical consequence of physiological and biochemical transformations. It results from rearrangements of plant metabolism in response to occurring nutrient stress. The results are in accordance with the results presented by Kumar-Terawi et al. (2004) and Huang et al. (2004), who demonstrated that deficiency of macroelements caused an increase in the activity of antioxidant enzymes in maize and rice plants. Deficiencies of P, K and Ca in a medium resulted in increased activity of CAT, while N deficiency caused a not significant decrease in the enzyme activity. No significant differences were observed in the activity of POX between control plants and plants growing in LNK, although the means were higher for plants under stress. A deficiency in all the examined macro-elements resulted in an increase in SOD. Kumar-Tewari et al. (2007) demonstrated that N and K deficiencies modify antioxidative responses of the examined plants in different ways. The activity of CAT, POX and SOD in the population of RILs was different in comparison with parental lines. On the one hand, despite the different responses to stress caused by N and K deficiencies noted in parental lines, generally in the population of RILs nutrient stress caused a reduction in the activity of CAT and POX and an increase in the activity of SOD. Significantly different responses were often observed in selected RILs as a higher activity of CAT, POX and SOD. It should be emphasized that high SOD activity, lower CAT and the lowest for POX were found in the distinguished extreme tolerant group, and low activities of CAT and SOD and higher activity of POX were found in the susceptible group. The increase in SOD activity in the group of tolerant RILs in comparison to susceptible RILs may demonstrate different genetic background determining the response of RILs to abiotic stress (Kumar-Tewari et al. 2004, Shin et al. 2005).

Ward's agglomerative method was used to group RILs according to the d_CL, d_RN and d_LRL values, in order to reveal the effective response of each RIL to N-K stress. RILs were clustered in seven discrete groups (‘a’-‘g’), among which those differing extremely were clustered in groups ‘a’ and ‘g’. Characterization of individual clusters is presented in Figure 3. It was demonstrated that RILs responding to stress in a similar way as those clustered in groups ‘a’ and ‘g’ were identified in the neighboring clads (Figure 2). If we adopt biometric seedling parameters as a criterion for the response of RILs to nutrient stress, it may be concluded that Ward's agglomerative method can be used to distinguish groups of RILs with extremely different responses from the examined population. The division has been verified with the use of the k-means method (data not shown). The examined population was divided into seven groups prior to its de novo agglomeration. It was demonstrated that the division was feasible. Variances between clusters were definitely higher than within groups. The d_LRL trait was the highest, highly significant statistic, which confirmed its usefulness as a predictor of the division of the population of RILs.

Manschadi et al. (2008) used Ward's agglomerative method to group wheat cultivars with different tolerances to drought in terms of two traits (variables): growth angle and seminal root number. The authors demonstrated that seminal root architecture is closely linked to the angle of seminal root axes at the seedling stage and can be used in breeding for selection for water-limited environments.

Discriminant function analysis was employed by Ebdon et al. (1998) in the identification of low- and high-water use Kentucky bluegrass cultivars, while Kanbar et al. (2010) used it for selection of deep- and shallow-rooted plants in an early segregating generation of rice to identify drought-tolerant lines. In this study the greatest observable contribution comes from LRL (λ = 0.333, F = 128.020), and also from CAT, SOD; in addition Car turned out to be significant in the model. The result of canonical analysis was in agreement with the result of discriminant function analysis. RILs with high d_LRL ‘in minus’ were linked with deeper roots and, in the opinion of many authors, with better NUE (Rzepka-Plevneš et al. 1997a; Rzepka-Plevneš et al. 1997b; Lynch, 2007; Liu et al. 2008; Zhang et al. 2010).

In conclusion, - the range of morphological, physiological and biochemical changes described for the seedlings of recombinant inbred lines obtained from mature embryos was typical of such a study type. In spite of the many limitations and imperfections presented by Hazarika (2006), this fact according to the authors may support the commonly held opinion on the usefulness of the employed method as one of the possible laboratory methods for describing potential of plant response per se to nutrient stress at the seedling stage. Genotypes responding in different way to nutrient stress enable identification of molecular markers useful for selection in low-input breeding, such as could constitute interesting material for the monitoring of gene expression.

The author wishes to thank J. Kurek for help and efficient assistance during the experiments.

ABASSI, N.A.; KUSHAD, M.M. and ENDRESS, A.G. (1998). Active oxygen-scavenging enzymes activities in developing apple flowers and fruits. Scientia Horticulturae, vol. 74, no. 3, p. 183-194. [CrossRef]

BENDERRADJI, L.; BRINI, F.; KELLOU, K.; YKHLEF, N.; DJEKOUN, A.; MASMOUDI, K. and BOUZERZOUR, H. (2012). Callus induction, proliferation, and plantlets regeneration of two bread wheat (Triticum aestivum L.) genotypes under saline and heat stress conditions. ISRN Agronomy, vol. 2012, no. 367851. [CrossRef]

BLOOM, A.J.; CHAPIN, F.S. and MOONEY, H.A. (1985). Resource limitation in plants an economic analogy. Annual Reviews of Ecology, Evolution and Systematics, vol. 16, p. 363-392. [CrossRef]

BOLAÑOS, J.; EDMEADES, G.O. and MARTINEZ, L. (1993). Eight cycles of selection for drought tolerance in lowland tropical maize. III. Responses in drought-adaptive physiological and morphological traits. Field Crops Research, vol. 31, no. 3-4, p. 269-286. [CrossRef]

CAI, J.; CHEN, L.; QU, H.; LIAN, J.; LIU, W.; HU, Y. and XU, G. (2012). Alteration of nutrient allocation and transporter genes expression in rice under N, P, K, and Mg deficiencies. Acta Physiologiae Plantarum, vol. 34, no. 3, p. 939-946. [CrossRef]

CAKMAK, I.; EKIZ, H.; YILMAZ, A.; TORUN, B.; KOLELI, N.; GULTEKIN, I.; ALKAN, A. and EKER, S. (1997). Differential response of rye, triticale, bread wheat and durum wheats to zinc deficiency in calcareous soils. Plant Soil, vol. 188, no. 1, p. 1-10. [CrossRef]

CAKMAK, I. and ENGELS, C. (1999). Role of mineral nutrients in photosynthesis and yield formation. In: RENGEL; Z. (ed). Mineral Nutrition of Crops: Mechanisms and Implications. The Haworth Press, New York, USA, p. 141-168.

CHANCE, B. and MAEHLY, A.C. (1955). Assay of catalase and peroxidases. In: CALONIC, C.P. and KAPLAN, N.O. (eds). Methods in Enzymology, New York, Academic Press, vol. 2, p. 764-775.

CHUN, L.; MI, G.; LI, J.; CHEN, F. and ZHANG, F. (2005). Genetic analysis of maize root characteristics in response to low nitrogen stress. Plant and Soil, vol. 276, no. 1-2, p. 369-382. [CrossRef]

DE SMET, I.; ZHANG, H.; INZE, D. and BEECKMAN, T. (2006). A novel role for abscisic acid emerges from underground. Trends in Plant Sciences, vol. 11, no. 9, p. 434-439. [CrossRef]

EBDON, J.S.; PETROVIC, A.M. and SCHWAGER, S.J. (1998). Evaluation of discriminant analysis in identification of low- and high-water use Kentucky bluegrass cultivars. Crop Science, vol. 38, no. 1, p. 152-157. [CrossRef]

EPSTEIN, E. and BLOOM, A.J. (2005). Mineral Nutrition of Plants: Principles and Perspectives. Sinauer Associates Inc., 380 p. ISBN-13- 978-0878931729.

GAN, Y.; FILLEUR, S.; RAHMAN, A.; GOTENSPARRE, S. and FORDE, B.G. (2005). Nutritional regulation of ANR1 and other root-expressed MADS-box genes in Arabidopsis thaliana. Planta, vol. 222, no. 4, p. 730-742. [CrossRef]

GEIGER, H.H. and MIEDANER, T. (2009). Rye breeding. In: CARENA, M.J. (ed). Cereals. Germany, Springer-Science+Business Media; vol. 3, p. 157-181.

GÓRNY, A.G.; GEIGER, H.H.; MORGENSTERN, K. and SINGH, R.K. (1982). Correlations between seedling and adult plant characters in hybrids and inbred lines of rye (Secale cereale L.). Tagungsbericht Akademie der Landwirtschaftswissenschaften der DDR, Berlin, vol. 198, p. 389-399.

GÓRNY, A.G. and GEIGER, H.H. (1982). Variation and covariation among juvenile shoot and root characters of inbred lines and hybrids in rye (Secale cereae L.). Tagungsbericht Akademie der Landwirtschaftswissenschaften der DDR, Berlin, vol. 198, p. 445-454.

GÓRNY, A.G. (1995). Direct effects of cyclic selection for longer seminal roots in spring barley (Hordeum vulgare L.). Journal of Applied Genetics, vol. 36, no. 1, p. 17-26.

GÓRNY, A.G. and SZOŁKOWSKA, A. (1996). Effects of selection for more vigorous seminal roots in two cross populations of oat (Avena sativa L.). Journal of Applied Genetics, vol. 37, no. 4, p. 331-344.

GUO, Y.; MI, G.H.; CHEN, F. and ZHANG, F. (2005). Effect of NO₃^- supply on lateral root growth in maize plants. Journal of Plant Physiology and Molecular Biology, vol. 31, no. 1, p. 90-96.

HAGER, A. and MEYER-BERTHENRATH, T. (1966). Die Isolierung und quantitative Bestimmung der Carotinoide und Chlorophylle von Blattern, Algen und isolierten Chloroplasten mit Hilfe dunnschichtchromatographischer Methoden. Planta, vol. 69, no. 3, p. 198-217. [CrossRef]

HAZARIKA, B.N. (2006). Morpho-physiological disorders in in vitro culture of plants. Scientia Horticulturae, vol. 108, no. 2, p. 105-120. [CrossRef]

HUANG, Z.A.; JIANG, D.A.; YANG, Y.; SUN, J.W. and JIN, S.H. (2004). Effects of nitrogen deficiency on gas exchange, chlorophyll fluorescence, and antioxidant enzymes in leaves of rice plants. Photosynthetica, vol. 42, no. 3, p. 357-364. [CrossRef]

JOSHI, R.; SHUKLA, A. and SAIRAM, R.K. (2011). In vitro screening of rice genotypes for drought tolerance using polyethylene glycol. Acta Physiologiae Plantarum, vol. 33, no. 6, p. 2209-2217. [CrossRef]

KANBAR, A.; TOORCHI, M.; MOTOHASHI, T.; KONDO, K. and SHASHIDHAR, H.E. (2010). Evaluation of discriminant analysis in identification of deep and shallow rooted plants in early segregating generation of rice (Oryza sativa L.) using single tiller approach. Australian Journal of Basic and Applied Sciences, vol. 4, no. 8, p. 3909-3916.

KROUK, G.; TILLARD, P. and GOJON, A. (2006). Regulation of the high-affinity NO₃^- uptake system by NRT1.1-mediated NO₃^- demand signaling in Arabidopsis. Plant Physiology, vol. 142, no. 3, p. 1075-1086. [CrossRef]

KUMAR-TEWARI, R.; KUMAR, P.; TEWARIA, N.; SRIVASTAVAA, S. and SHARMA, P.N. (2004). Macronutrient deficiencies and differential antioxidant responses-influence on the activity and expression of superoxide dismutase in maize. Plant Science, vol. 166, no. 3, p. 687-694. [CrossRef]

KUMAR-TEWARI, R.; KUMAR, P. and SHARMA, P.N. (2007). Oxidative stress and antioxidant responses in young leaves of mulberry plants grown under nitrogen, phosphorus or potassium deficiency. Journal of Integrative Plant Biology, vol. 49, no. 3, p. 313-322. [CrossRef]

LICHTENTHALER, H.K. and WELLBURN, A.R. (1983). Determinations of total carotenoids and chlorophylls a and b of leaf extracts in different solvents. Biochemical Society Transactions, vol. 11, no. 5, p. 591-592.

LIU, J.; LI, J.; CHEN, F.; ZHANG, F.; REN, T.; ZHUANG, Z. and MI, G. (2008). Mapping QTLs for root traits under different nitrate levels at the seedling stage in maize (Zea mays L.). Plant and Soil, vol. 305, no. 1-2, p. 253-265. [CrossRef]

LÖSCHENBERGER, F.; FLECK A.; GRAUSGRUBER, H.; HETZENDORFER, H.; HOF, G.; LAFFERTY, J.; MARN, M.; NEUMAYER, A.; PFAFFINGER, G. and BIRSCHITZKY, J. (2008). Breeding for organic agriculture: The example of winter wheat in Austria. Euphytica, vol. 163, no. 3, p. 469-480. [CrossRef]

LÜCK, H. (1965). Catalase. In: BERGMEYER, H.U. (ed). Methods of enzymatic analysis, New York, Verlag Chemie, Academic Press, p. 885-888.

LUDLOW, M.M. and MUCHOW, R.C. (1990). A critical evaluation of traits for improving crop yields in water limited environments. Advances in Agronomy, vol. 43, p. 107-153. [CrossRef]

LYNCH, J.P. (2007). Roots of the second green revolution. Australian Journal of Botany, vol. 55, no. 5, p. 493-512. [CrossRef]

MANSCHADI, A.M.; HAMMER, G.L.; CHRISTOPHER, J.T. and deVOIL, P. (2008). Genotypic variation in seedling root architectural traits and implications for drought adaptation in wheat (Triticum aestivum L.). Plant Soil, vol. 303, no. 1-2, p. 115-129. [CrossRef]

MARSCHNER, H. (1995). Nutritional physiology. In: MARSCHNER, H. (ed). Mineral Nutrition of Higher Plants. 2^nd ed. San Diego, USA, Academic Press, p. 229-299.

MASOJĆ, P. and KOSMALA, A. (2012). Proteomic analysis of preharvest sprouting in rye using two-dimensional electrophoresis and mass spectrometry. Molecular Breeding, vol. 30, no. 3, p. 1355-1361. [CrossRef]

MESSMER, M.; HILDERMANN, I.; THORUP-KRISTENSEN, K. and RENGEL, Z. (2011). Nutrient management in organic farming and consequences for direct and indirect selection strategies. In: LAMMERTS VAN BUEREN, E.T. and MYERS, J.R. (eds). Organic crop breeding. Oxford, UK, Blackwell & Wiley, p. 21-27.

MI, G.; CHEN, F. and ZHANG, F. (2007). Physiological and genetic mechanisms for nitrogen use efficiency in maize. Journal of Crop Science and Biotechnology, vol. 10, no. 2, p. 57-63.

MURASHIGE, T. and SKOOG, F. (1962). A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiologia Plantarum, vol. 15, no. 3, p. 473-497. [CrossRef]

POSTMA, J.A. and LYNCH, J.P. (2011). Root cortical aerenchyma enhances the growth of maize on soils with suboptimal availability of nitrogen, phosphorus and potassium. Plant Physiology, vol. 156, no. 3, p. 1190-1201. [CrossRef]

RAI, M.K.; KALIA, R.K.; SINGH, R.; GANGOLA, M.P. and DHAWAN, A.K. (2011). Developing stress tolerant plants through in vitro selection - An overview of the recent progress. Environmental and Experimental Botany, vol. 71, no. 1, p. 89-98. [CrossRef]

RAKOCZY-TROJANOWSKA, M. and MALEPSZY, S. (1995). Genetic factors influencing the regeneration ability of rye (Secale cereale L.). II. Immature embryos. Euphytica, vol. 83, no. 3, p. 233-239. [CrossRef]

REMANS, T.; NACRY, P.; PERVENT, M.; FILLEUR, S.; DIATLOFF, E.; MOUNIER, E.; TILLARD, P.; FORDE, B.G. and GOJON, A. (2006). The Arabidopsis NRT1.1 transporter participates in the signaling pathway triggering root colonization of nitrate-rich patches. Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 50, p. 19206-19211. [CrossRef]

RENGEL, Z. (2005). Breeding crops for adaptation to environments with low nutrient availability. In: ASHRAF, M. and HARRIS, P.J.C. (eds). Abiotic stresses plant resistance through breeding and molecular approaches. New York, USA, Food Products Press, p. 239-259.

RUFFEL, S.; KROUK, G.; RISTOVA, D.; SHASHA, D.; BIRNBAUM, K.D. and CORUZZI, G.M. (2011). Nitrogen economics of root foraging: Transitive closure of the nitrate-cytokinin relay and distinct systemic signaling for N supply vs. demand. Proceedings of the National Academy of Sciences of the United States of America, vol. 108, no. 45, p. 18524-18529. [CrossRef]

RZEPKA-PLEVNEŠ, D.; KUREK, J. and MARCINIAK, H. (1997a). Agronomic properties of rye forms selected in vitro for tolerance to nutrient deficiency. Plant Breeding and Seed Science, vol. 41, no. 2, p. 91-100.

RZEPKA-PLEVNEŠ, D.; MARCINIAK, H. and ŚMIECH, M. (1997b). The evaluation of rye (S. cereale L.) inbred lines tolerance to nutrition deficiency by in vitro test (in Polish). Biuletyn IHAR, vol. 203, p. 137-146.

RZEPKA-PLEVNEŠ, D. and TOMCZAK, P. (1998). Agronomic properties of rye populations selected for tolerance to nitrogen deficiency in hydroponic cultures. Plant Breeding and Seed Science, vol. 42, no. 2, p. 65-74.

RZEPKA-PLEVNEŠ, D. (1999). Variability of tolerance to nitrogen and potassium deficiencies in original (S₀) and selected (S₁-S₃) rye populations; assessed during in vitro cultures. Plant Breeding and Seed Science, vol. 43, no. 1, p. 48-63.

RZEPKA-PLEVNEŠ, D. and KULPA, D. (1999). Agronomic properties of rye populations selected for tolerance to nutrition deficiency under laboratory conditions (in Polish). Biuletyn IHAR, vol. 211, p. 259-265.

SANGUINETI, M.C.; DUVICK, D.N.; SMITH, S.; LANDI, P. and TUBEROSA, R. (2006). Effects of long-term selection on seedling traits and ABA accumulation in commercial maize hybrids. Maydica, vol. 51, p. 329-338.

SHIN, R.; BERG, R.H. and SCHACHTMAN, D.P. (2005). Reactive oxygen species and root hairs in Arabidopsis root response to nitrogen, phosphorus and potassium deficiency. Plant Cell Physiology, vol. 46, no. 8, p. 1350-1357. [CrossRef]

TSAY, Y.F.; HO, C.H.; CHEN, H.Y. and LIN, S.H. (2011). Integration of nitrogen and potassium signaling. Annual Review of Plant Biology, vol. 62, p. 207-226. [CrossRef]

TUBEROSA, R. and SALVI, S. (2007). From QTLs to genes controlling root traits in maize. In: SPIERTZ, J.H.J.; STRUIK, P.C. and VAN LAAR, H.H. (eds). Scale and complexity in plant systems research. Heidelberg, Germany, Springer, p. 15-25.

TUBEROSA, R.; SALVI, S.; GIULIANI, S.; SANGUINETI, M.C.; FRASCAROLI, E.; CONTI, S. and LANDI, P. (2011). Genomics of root architecture and functions in maize. In: COSTA DE OLIVEIRA, A. and VARSHNEY, R.K. (eds). Root genomics. Heidelberg, Germany, Springer, p. 179-204.

VAN DER PLOEG, R.; BÖHM, W. and KIRKHAM, M.B. (1999). On the origin of the theory of mineral nutrition of plants and the law of minimum. Soil Science Society of America Journal, vol. 63, no. 5, p. 1055-1062. [CrossRef]

VON CAEMMERER, S. and FARQUHAR, G.D. (1981). Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta, vol. 153, no. 4, p. 376-387. [CrossRef]

WALCH-LIU, P.; IVANOV, I.I.; FILLEUR, S.; GAN, Y.; REMANS, T. and FORDE, B.G. (2006). Nitrogen regulation of root branching. Annals of Botany, vol. 97, no. 5, p. 875-881. [CrossRef]

WANG, Y.; MI, G.; CHEN, F.; ZHANG, J. and ZHANG, F. (2005). Response of root morphology to nitrate supply and its contribution to nitrogen accumulation in maize. Journal of Plant Nutrition, vol. 27, no. 12, p. 2189-2202. [CrossRef]

WIESLER, F. and HORST, W.J. (1994). Root growth and nitrate utilization of maize cultivars under field conditions. Plant and Soil, vol. 163, no. 2, p. 267-277. [CrossRef]

ZHANG, H. and FORDE, B.G. (1998). An Arabidopsis MADS box gene that controls nutrient induced changes in root architecture. Science, vol. 279, no. 5349, p. 407-409. [CrossRef]

ZHANG, H.; JENNINGS, A.; BARLOW, P.W. and FORDE, B.G. (1999). Dual pathways for regulation of root branching by nitrate. Proceedings of the National Academy of Sciences of the United States of America, vol. 96, no. 11, p. 6529-6534. [CrossRef]

ZHANG, H.; RONG, H. and PILBEAM, D. (2007). Signalling mechanisms underlying the morphological responses of the root system to nitrogen in Arabidopsis thaliana. Journal of Experimental Botany, vol. 58, no. 9, p. 2329-2338. [CrossRef]

ZHANG, F.; NIU, J.; ZHANG, W.; CHEN, X.; LI, C.; YUAN, L. and XIE, J. (2010). Potassium nutrition of crops under varied regimes of nitrogen supply. Plant and Soil, vol. 335, no. 1-2, p. 21-34. [CrossRef]

Note: Electronic Journal of Biotechnology is not responsible if on-line references cited on manuscripts are not available any more after the date of publication.