•Zinc (Zn) deficiency has been recognized as a potential risk for human health in many developing regions where staple food with low micronutrient density represents a major proportion of the diet. The success of strategies to increase Zn content in the edible part of crops requires better understanding of Zn transport to, and distribution within, the grains.
•The transfer of Zn from the growth medium to wheat (Triticum aestivum) grains in an ear culture system was investigated by using the stable Zn isotope 70Zn, and the spatial distribution of Zn within the grains was studied by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS).
•Zinc was readily transported in the stem up to the rachis. More Zn accumulated in the stem when higher amounts of Zn were supplied to the medium. Once Zn was transported into the grain, Zn accumulated particularly in the crease vascular tissue. The gradient of 70Zn concentration between crease vascular tissue, aleurone layer and endosperm demonstrates that Zn is distributed within grain through the crease phloem.
•These results suggest that two barriers of Zn transport into wheat grains may exist: between the stem tissue rachis and the grain, and the maternal and filial tissues in the grain.
Although substantial efforts have been made to understand Zn transport in plants (Herren & Feller, 1994; Pearson & Rengel, 1994; Pearson et al., 1995, 1996a,b, 1998, 1999; Rengel, 2001; Tauris et al., 2009; Waters et al., 2009), still no clear picture exists of how Zn is transported to ears and stored in the grains. One of the barriers to a better in-depth understanding of metal transport and accumulation in plants is the lack of appropriate methods that allow the localization of individual metals at the tissue and cellular levels (Mills et al., 2005; Grotz & Guerinot, 2006). This applies, in particular, to the quantification of the spatial distribution of mineral elements within the seeds. There are several methods available dealing with the spatial distribution of elements in seeds, but each has its limits. One approach is the mechanical separation of different seed tissue that, at best, only allows a rough estimation of major differences in tissue mineral concentrations (Bityutskii et al., 2002), even though such analysis has contributed to describing the differences among the embryo, seed coat and endosperm. A simple but powerful tool, particularly for breeding purposes, is the use of dyes for staining of specific elements, such as dithizone for Zn (Ozturk et al., 2006; Choi et al., 2007) and potassium ferrocyanide for Fe (Choi et al., 2007; Pintasen et al., 2007). However, these techniques can only provide a rough and semiquantitative estimate of metal distribution and concentration. Much more powerful combinations of X-ray analysis and electron microscopy techniques, such as energy-dispersive X-ray (EDX) analysis (Otegui et al., 2002), microproton-induced X-ray emission (micro-PIXE, Bhatia et al., 2003; Vogel-Mikušet al., 2007) and synchrotron X-ray fluorescence (S-XRF) mapping (Young et al., 2007; Meharg et al., 2008; Lombi et al., 2009), as well as high-resolution secondary ion mass spectrometry (SIMS) (Heard et al., 2001; Feeney et al., 2003; Moore et al., 2010) are currently available methods for better localization of individual components. In particular, Kim et al. (2006) convincingly demonstrated the potential of combining S-XRF microtomography in relating molecular studies on micronutrient transport to mapping element distribution in Arabidopsis seeds. By using a more precise laser capture microdissection technique and gene expression profiling, Tauris et al. (2009) proposed a roadmap for zinc trafficking in the developing barley grain.
Laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) is an emerging technique for measuring metal distribution in plant tissues (Becker et al., 2008; Shi et al., 2009; Wirth et al., 2009; Wu et al., 2009). The advantages over other methods might be the ease of sample preparation and handling, the low cost and, most importantly, the possibility of tracing the movement of the mineral elements using stable isotopes introduced to the plant at specific developmental stages with high sensitivity.
Ear culture is an established method for growing detached ears of small grain cereals in culture medium (Donovan & Lee, 1977; Herren & Feller, 1994; Pearson et al., 1996a,b; Zhou et al., 2006). This method is particularly suited to metals at specific developmental stages during reproductive growth. In this experiment, we grew detached wheat ears in the culture medium with the 70Zn isotope, in order to trace the movement of the applied Zn to the grain. Using LA-ICP-MS, the spatial distribution of Zn within the grains was studied in order to elucidate the Zn transport pathway within the developing grain.
Materials and Methods
The seeds of the spring wheat Triticum aestivum L. cv ‘Segetoria’, kindly supplied by I. Cakmak, Sabanci University, Turkey, were sown in Micherlich pots, each pot containing 4 l peat substrate (Einheitserde type P, 10% clay, bulk density 500 g l−1). The plants were grown in a glasshouse with natural light and a temperature regime (day : night) of 25 : 20°C (aeration : heating set). The plants were watered and fertilized regularly and treated with fungicides or insecticides when necessary. At the beginning of anthesis, individual ears were tagged to be able to select and compare ears at exactly the same developmental stage.
Wheat ear culture
Detached wheat ears were cultured according to Singh & Jenner (1983) and Sharma et al. (1995). At anthesis (Expt 1) or 3 wk after anthesis (Expt 2), the tagged ears with the flag leaf removed were cut at the penultimate stem for in vitro culture. The stems were sterilized in 70% ethanol for 1 min and 1% NaClO for 5 min, and then washed with sterile distilled water. After sterilization, the ears were further cut c. 2 cm from the bottom of the penultimate stem under sterile distilled H2O and then transferred into a cotton-plugged conical flask containing 100 ml liquid culture medium. The basic components of the culture medium were as in the MS culture medium (Murashige & Skoog, 1962) with modifications in nitrogen and glutamine concentrations according to preliminary experiments aimed at matching grain growth rates of intact ears in the pots (Table 1). Zn concentration and form in the culture medium were varied according to the treatments.
Table 1. Composition of the wheat ear culture medium
Class of nutrient
Concentration (mg l−1)
8.6, 17.2, 34.4, 68.8 or 25.8
Expt 1 was designed to investigate how the Zn concentration in the culture medium affects Zn uptake and partition in detached wheat ears at the early growth stage of grain setting. Treatments represented normal to high Zn concentrations of 30, 60, 120 and 240 μM in the form of natural isotopes of ZnSO4. The ear culture started at anthesis and the treatment duration was 7 d.
In Expt 2, we studied how late-supplied Zn was transported to wheat ears and distributed in the grains by using the stable isotope 70Zn. Natural isotopes of Zn and the 99% enriched stable isotope 70Zn (STB Isotope Germany GmbH, Hamburg, Germany) in the form of ZnSO4 were used in the study. Zinc concentrations of 30 and 90 μM were used to represent normal and high Zn supply. Ear culture was started at 21 d after anthesis (DAA) and the treatment duration was 9 d.
All media and containers were autoclaved. Detached ears were cultured in conical flasks and incubated in a growth cabinet under a day : night temperature cycle of 25 : 22°C, a 16 h photoperiod and a photosynthetically active radiation (PAR) of 120 μmol m−2 s−1.
Zn analysis of different organs of the wheat ear
After termination of the ear culture, the wheat ears were separated into grains, rachis, glumes (including the lemma and palea), peduncle and the node right below the peduncle. From each ear, three uniform grains in the middle position were selected for LA-ICP-MS analysis. The other grains and organs from the same ear were oven-dried at 60°C for 3 d and the dry weights were recorded. The dried samples were ground by a vibrating cup mill (Puverisette; Fritsch GmbH, Idar-Oberstein, Germany) and used for total Zn analysis.
The samples were dry-ashed in a muffle oven at 480°C for 16 h. The ash was dissolved in 1 ml of 1 : 3 (HNO3 : H2O) ultrapure nitric acid, diluted with further 9 ml of deionized water. After passing the solution through filter paper (Rotilabo-Rundfilter AP52.1, 180s; Carl Roth GmbH, Karlsruhe, Germany), the solution was analysed by ICP-MS (7500cx; Agilent Technologies, Santa Clara, California, USA) for the isotopes 64Zn, 66Zn, 67Zn, 68Zn and 70Zn, separately.
Localization of Zn in grains using LA-ICP-MS
Sample and calibration standard preparation Fresh grains were freehand-sectioned. The cross-sections were fixed on glass slides using double-face fotostrip (Tesa AG, Hamburg, Germany) and subsequently air-dried for at least 24 h before analysis.
For calibration, wheat flour (10 g) was spiked with different Zn standard solutions (10 ml) and dried at 60°C for 48 h. The Zn-spiked flour was homogenized using a vibrating cup mill (Puverisette; Fritsch GmbH) for 10 min. The flour was then pressed into pellets of 5 mm diameter. The Zn concentration was determined by digesting the flour in nitric acid and measuring by ICP-MS (7500cx; Agilent Technologies).
For calculation of the grain Zn concentration, the Zn : 13C ratio (13C as internal standard) of the calibration standards and the grains were used. Four cross-sections from four grains of each treatment were analysed; the images shown in Fig. 5 and Fig. 6 are representative of four independent measurements of each treatment.
Zn localization A quadrupole ICP-MS coupled to a laser ablation system, UP193SS (New Wave Research, Fremont, CA, USA), was used to determine the Zn concentration and distribution in wheat grains. Samples and matrix-matched calibration standards were arranged in a laser ablation chamber (Supercell; New Wave Research) and ablated under the same experimental conditions, in order to allow the calculation of Zn concentrations. The ablated material was transported by argon as a carrier gas to the plasma. The ICP-MS instrument tuning was optimized with respect to the maximum ion intensity of low masses. The flow rate of the carrier gas and the make-up gas were adjusted to optimum output; the flow rate was 0.25 l min−1 for the carrier gas and 1.25 l min−1 for the make-up gas. Radiofrequency (RF) power was set to 1300 W and the reaction mode was off. The laser parameters were set to 2 J cm−2 of output energy, 10 Hz repetition rate, 50 μm diameter of the crater size, and 20 μm s−1 scan speed. Measured elements were 64Zn, 66Zn, 67Zn, 68Zn, 70Zn and 13C. 13C was used as an internal standard to normalize for different quantities of seed tissue ablated.
For the calculation of 70Zn concentration in the endosperm of seed cross-sections, two locations (two lines in the length of 500 μm) from both sides of the crease of each cross-section were chosen. The 70Zn concentration of the endosperm was the mean of 200 measurements obtained by LA-ICP-MS run at the speed of one measurement per 5 μm. Three cross-sections per treatment were measured.
Zn partitioning in different organs of detached wheat ears
In Expt 1, we examined how the Zn concentration in the culture medium affects Zn uptake and partitioning in detached wheat ears at the early growth stage of grain setting (0–7 DAA). The Zn concentrations and contents in the grain, glumes (including the lemma and palea), rachis and peduncle increased in response to increasing Zn supply in the culture medium, but the magnitude of the increase varied among different tissues of the wheat ear (Fig. 1). With a low Zn supply of 30 μM, all parts of the wheat ear had similar Zn concentrations, in the range 52–56 μg g−1 DW. At higher Zn supply in the culture medium, the increase in Zn concentration in the rachis was much greater than in other tissues (Fig. 1a). The Zn concentration in the rachis reached 628 μg g−1 DW at 240 μM Zn supply, while the whole-grain Zn concentration was only 145 μg g−1 DW. When the Zn supply increased eightfold (from 30 to 240 μM) in the medium, the Zn concentration in the rachis increased 11-fold but only 2.6-fold in the grain. Because of the small contribution of the rachis to the dry matter of the ear, the Zn content of the rachis represented only 20.3% of the grain Zn content at low Zn supply (Fig. 1b). However, with increasing Zn supply, the Zn content of the rachis reached almost the same amount as that of the grains, indicating a transport barrier at the interface between rachis and grain. This barrier appears to become increasingly effective at high Zn supply.
When the ears were transferred to the culture medium later during seed formation (21–30 DAA, Expt 2), increasing the Zn supply from 30 to 90 μM also significantly enhanced the Zn concentrations (Fig. 2a) and Zn contents (Fig. 2b) of all plant organs, showing that at this developmental stage Zn is also transported into the vegetative and reproductive parts of the ear. Again the Zn concentration of the rachis was particularly increased at the higher Zn supply. At this developmental stage, however, the proportion of Zn stored in the rachis (Zn content, Fig. 2b) had decreased because of the lower contribution of its dry matter to that of the total ear compared with the grains.
In this experiment Zn was applied to the culture medium either as 99% enriched 70Zn or as natural isotopes of Zn (control), allowing the quantification of the Zn uptake into the ear during this developmental stage. When natural isotopes of Zn were supplied in the ear culture, the percentage of each Zn isotope in each organ did not change with higher Zn supply, and the percentages represent the natural distribution of Zn isotopes, as follows: 48.6%64Zn, 27.9%66Zn, 4.1%67Zn, 18.8%68Zn and 0.6%70Zn (Fig. 3). The supply of 70Zn as the only Zn source during the ear-culture period changed the proportion of the Zn isotopes in each organ and reflects the uptake of Zn into the organ during the culture period. The proportion of 70Zn in all plant organs increased with the 70Zn supply. Between the organs, the 70Zn accounted for 71.0 and 86.7% (30 and 90 μM 70Zn supply, respectively) of the total Zn in the rachis and decreased through the peduncle and the glumes to the grain (26.5% at 30 μM and 44.4% at 90 μM Zn supply). A transport barrier for Zn between the rachis and the grain is further supported by the comparison of the relative concentration of 70Zn in the organs along the transport pathway when the 70Zn supply in the culture medium is increased from 30 to 90 μM (Table 2). When the 70Zn supply in the culture medium increased threefold, the 70Zn concentration increased by a factor of 4.1, 4.2, 5.2, 5.1 and 3.6 in the peduncle node, peduncle (lower half), peduncle (upper half), rachis and glumes, respectively, and by a factor of only 2.3 in grains.
Table 2. The concentrations of 70Zn in different organs of detached wheat ears (21 d after anthesis (DAA)) cultured in medium containing the stable isotope 70Zn at 30 μM (30) or 90 μM (90) for 9 d
Zn supply (μM)
70Zn concentration (μg g−1 DW)a
70Zn concentration in culture medium (μM)
Peduncle (upper half)
Peduncle (lower half)
aValues are means ± SD (n =5).
11.2 ± 2.0
9.9 ± 2.0
24.8 ± 4.1
16.2 ± 1.1
7.1 ± 2.3
15.9 ± 5.8
25.8 ± 2.5
35.2 ± 4.7
127.3 ± 19.5
83.5 ± 11.9
30.0 ± 7.6
64.7 ± 8.4
Ratio 90 : 30
Zn distribution in wheat grains
The wheat grains did not reach maturity at the end of the ear culture Expt 2. The average grain DW was 38 mg, which was 16.7% less than that of mature grains when intact plants were allowed to grow in the pots until maturity. Zinc treatment during the 9 d of in vitro culture did not affect grain DW compared with ears of plants grown in pots to the same growth stage (data not shown). During this period, the Zn concentration in the grain was 42.0 μg g−1 DW at 30 μM Zn supply and 58.9 μg g−1 DW at 90 μM Zn supply (Fig. 4). The uptake of Zn as reflected by the increase in 70Zn concentration amounted to 26.6 or 44.8% of the total grain Zn at 30 or 90 μM Zn supply, respectively, demonstrating the importance of Zn transport to the grains at later grain-filling stages.
The localization of Zn within the grains using LA-ICP-MS (Fig. 5) clearly shows that the Zn concentration in the vascular tissue of the crease was much higher than in the aleurone layer or the seed coat. The endosperm had the lowest Zn concentration. The mean Zn concentrations of the crease vascular tissue, the aleurone layer and the endosperm were 190, 76 and 8 μg g−1 DW, respectively, at a 30 μM Zn supply (Fig. 5a,c). Higher Zn supply in the culture medium increased the overall Zn concentration of the grain, particularly in the crease vascular tissue and the endosperm around the crease vascular tissue. The mean Zn concentrations of the crease vascular tissue, the aleurone layer and the endosperm were 579, 138 and 10 μg g−1 DW, respectively, at a 90 μM Zn supply (Fig. 5b,d). The average increases in mean Zn concentration were 200, 82 and 28% for the crease vascular tissue, the aleurone layer and the endosperm, respectively. As expected, very low 70Zn concentrations were observed in the natural isotopes of Zn treatment. When 70Zn was applied to the ears for 9 d, the 70Zn concentration in the grain increased in the endosperm as well as in the aleurone layer. However, the greatest increase was again in the vascular tissue of the crease (Fig. 5c,d).
The accumulation of late-applied 70Zn in the crease vascular tissue and the distribution gradient between crease vascular tissue and endosperm is more clearly shown in Fig. 6. This time the laser beam was running parallel to the crease vascular tissue with increasing distance from the crease vascular tissue aiming at repeated measurements in the same grain tissue (Fig. 6a,b). The steep decreasing gradient of the 70Zn concentration of the ablation line L1 shows that the ablation started in the crease vascular tissue but then passed into the seed coat layer (Fig. 6b,c). The laser beam running through the aleurone layer (L2) and the endosperm at increasing distance from the crease vascular tissue (L3, L4, L5) shows 6.4 μg g−1 DW of 70Zn in the aleurone layer (L2) and 3.2, 0.9 and 0.8 μg g−1 DW in the endosperm (L3, L4, L5). The ratios 70Zn : total Zn were 0.42, 0.14, 0.14, 0.12 and 0.11 for the crease vascular tissue, aleurone layer and endosperm L3, L4, L5, respectively. The steep concentration gradient from the crease vascular tissue to the endosperm concentrations clearly demonstrates the role of the crease vascular tissue in loading Zn into the grain.
Zinc distribution in detached wheat ears
The ear culture technique of small cereals was developed to study the grain-filling process, including amino acids and sucrose uptake (Graham & Morton, 1963; Jenner, 1968; Donovan & Lee, 1977), mineral transport into the grain (Herren & Feller, 1994; Pearson et al., 1995, 1996a,b, 1999; Riesen & Feller, 2005), and the regulation of starch and protein synthesis (Sasaki et al., 2005; Zhou et al., 2006). In contrast to the intact plants, the growth of cereal ears in liquid culture initially involves the movement of the culture medium into the plant tissue via the xylem. Normally the developing seed depends mainly on import through the phloem from the vegetative parts of the plant for its supply of carbohydrates, amino acids, minerals and other minor constituents (Donovan et al., 1990). Solutes can be selectively removed from the xylem sap during its acropetal translocation and loaded into the phloem. Mineral nutrients and assimilates can be transferred from the xylem to the phloem in the peduncle and in other stem parts of maturing wheat. In the ear culture system, the driving force of solution movement in the stem is probably the ear transpiration occurring in the lemma and glumes (Pearson et al., 1996b), since no leaves are attached in most cases. The nutrient transfer from xylem to phloem may take place through loading into the sieve tubes by a xylem-to-phloem transfer in the stem (Herren & Feller, 1994), but most abundant transfer probably takes place in the attachment region where close vascular connections between the grain, lemma, palea, glume and rachilla are involved (Pearson et al., 1995). In this region, the vascular connections are complex and many xylem and phloem transfer cells are present (O’Brien et al., 1985).
There is a xylem discontinuity at the point of the attachment of the grain to the rachis (Zee & O’Brien, 1970). This xylem discontinuity obviously functioned as the first barrier in Zn transport to grain in this study. There was a great concentration gradient of 70Zn between grain and rachis, particular at high Zn supply in the culture medium (Table 2). The concentration difference between the two adjacent plant tissues (Table 3) could be an indicator of the Zn transport efficiency through this barrier. A higher Zn concentration in the culture solution enhanced the Zn concentration difference. At 240 vs 30 μM Zn supply, the Zn concentrations were higher by factors of 11 and 2.6 in the rachis and the grain, respectively, supporting the previous statement on the transport efficiency through barrier I from the rachis to the grain. With different experimental approaches, Pearson et al. (1999) proposed that a protective barrier preventing excessive accumulation of Zn in the grain existed at high concentrations of solution Zn. Herren & Feller (1994) observed a similar phenomenon when investigating Zn unloading from the xylem and transfer to the phloem in the peduncle of wheat at very high Zn concentrations in the feeding solution (100–1000 μM). They suggested that Zn accumulation in the stem may contribute to the maintenance of relatively constant Zn concentrations in the grains. Alternatively, the transfer barrier may result from the saturation of the Zn transport system (Pearson et al., 1995).
Table 3. Zinc (Zn) concentration gradients across the proposed barriers in wheat as affected by the Zn supply in the culture medium
Zn supply in culture solution (μM)
Ratio of Zn concentration grain : rachisa
70Zn supply in culture solution (μM)
Ratio of 70Zn concentration endosperm : crease vascular tissuea
aValues are means ± SD; n =5 for barrier I (early grain-filling stage, 0–7 d after anthesis (DAA)) and n =3 for barrier II (late grain-filling stage, 21–30 DAA).
(10.8 ± 4.6) × 10−1
(14 ± 1) × 10−3
(4.5 ± 0.5) × 10−1
(3.8 ± 0.9) × 10−1
(8 ± 2) × 10−3
(2.3 ± 0.1) × 10−1
The regulation of Zn transport within stems of maturing cereals has not yet been addressed experimentally. But studies of Zn transport in roots or shoots of young seedlings of model plants indicate that Zn concentrations are tightly controlled at both the cellular and plant organ levels; yet the key mechanisms in such processes are unclear (Berg & Shi, 1996; Gitan et al., 1998; Waters et al., 2009).
Zinc distribution in wheat grains
The wheat grain consists of four major tissues: the embryo, the aleurone, the starchy endosperm and the outer layers (testa and pericarp). Zinc was preferentially localized in the aleurone and the embryo (Mazzolini et al., 1985; Ozturk et al., 2006). Our results support this view, but, in addition, the LA-ICP-MS technique revealed a Zn concentration gradient in the endosperm. The distribution of the stable isotope 70Zn clearly demonstrated that Zn is delivered to the endosperm through the crease vascular tissue during grain filling (Figs 5, 6). The large difference between crease vascular tissue and mean endosperm Zn concentrations (Table 3) suggests that the translocation from the maternal to filial tissues may be an additional bottleneck for Zn accumulation in the endosperm. Again, a higher Zn supply in the culture solution enhanced the concentration difference between the two grain tissues. At 90 vs 30 μM Zn supply, the Zn concentrations were higher by factors of 3.0 and 1.3 in the crease vascular tissue and the endosperm, respectively, supporting the existence of a second barrier limiting the transport of Zn from the crease vascular tissue to the endosperm. According to Thorne (1985), in wheat, assimilates are unloaded from the phloem in the crease vascular bundle symplastically through the chalaza and then into the apoplast via transfer cells in the nucellus. The transfer cells of the nucellar projection constitute the interface between the maternal tissues and an apoplastic space, the endosperm cavity, which separates the maternal and filial tissues (Wang et al., 1995). Laser microdissection and gene expression profiling of barley grains revealed that many genes coding for Zn transporters are expressed at high levels in these transfer cells (Tauris et al., 2009). These authors proposed that the transfer cells accumulate Zn in vacuoles for temporary storage from where it is mobilized, transferred to the cytosol and pumped into the endosperm cavity. In agreement with the X-ray energy spectra (Mazzolini et al., 1985), our results using LA-ICP-MS show a high concentration of Zn in this region (Fig. 5). Therefore, the transfer cells and endosperm cavity separating the maternal and filial tissues might constitute a second barrier for Zn transport into the grain endosperm. One important limiting factor for Zn accumulation in the endosperm could be the number and efficiency of the transporters in the transfer cells responsible for Zn efflux and/or the Zn-importing transporters in the modified aleurone cells facing the endosperm cavity.
The Zn transporters involved in the first and second transport barriers discussed earlier are probably concentration-regulated, controlling the efficiency of Zn translocation into the grain. In all organisms, zinc uptake is tightly controlled to ensure that adequate amounts of the metal are accumulated while potentially toxic over-accumulation is prevented (Guerinot & Eide, 1999). There is convincing evidence showing that specific Zn transporters in the plant roots or shoots are up- or down-regulated at suboptimal or supraoptimal Zn supply, respectively (see reviews by Guerinot & Eide, 1999; Eide, 2005, 2006), but so far there is little information on Zn transporters in seeds (Tauris et al., 2009). Deciphering the regulation mechanism of Zn transport in developing grains may speed up the breeding of genotypes with higher Zn density in seeds.
In conclusion, the data presented suggest the existence of two barriers along the Zn transport pathway to the grains. It would be important to know whether these barriers are specific for Zn or whether they also apply to other nutritionally favourable or undesirable heavy metals. The significance of these barriers is tremendous, because the increase of grain Zn content needs to break these barriers. This, however, may impose the risk of over-accumulation in grains of Zn and other heavy metals toxic to humans and animals. Therefore, the precise control of these barriers seems important in future attempts to increase grain Zn concentration.
The financial support by the German Research Foundation (DFG) to W.J.H. is gratefully acknowledged (grant HO 931/23-1). We wish to thank the reviewers for their helpful suggestions and I. Cakmak, Sabanci University, Turkey, for providing the seeds.