Index | Search | Home | Table of Contents
Jasso de Rodríguez, D., J. Romero-García, R. Rodríguez-García, and J.L. Angulo Sánchez. 2002. Characterization of proteins from sunflower leaves and seeds: Relationship of biomass and seed yield. p. 143149. In: J. Janick and A. Whipkey (eds.), Trends in new crops and new uses. ASHS Press, Alexandria, VA.
Diana Jasso de Rodríguez, Jorge Romero-García, Raúl Rodríguez-García, and José Luis Angulo Sánchez
Sunflower (Helianthus annuus L., Asteraceae), currently cultivated for its seeds, is the worlds fourth largest oil-seed crop. World seed production was 25.2 million tonnes (t) during 1995/1996 from approximately 50 million ha of cultivated land. In the US, during 1999, sunflower seed production was 1,971,205 t from 1,438,000 ha, according to the National Agricultural Statistics Service, U.S. Department of Agriculture. Sunflower meal is a potential source of protein for human consumption (Sosulski 1979) due to its high nutritional value and lack of antinutritional factors (Smith 1968). About 3 to 7 t of plant biomass per ha, including 10% of the flowerhead, is left in the field and represents biomass that may have other uses (Marechal and Rigal 1999).
Information on the biochemical and biophysical composition of sunflower seed proteins is limited and there are fewer reports dealing with leaf proteins. The majority of seed proteins are globulins (Gheyasuddin et al. 1970) with sedimentation constants of about 11S (Joubert 1955). The reported molecular masses are 300,000350,000 Da for the sunflower globulins (Sabir et al. 1973). The major seed storage globulin in several sunflower species (heliantinin) has been studies by Anisimova (1992) who found three main groups of polypeptides: one with basic polypeptides of 20,000 Da molecular mass and two with acidic polypeptides (30,000 and 40,000 Da). However, the composition was highly variable.
In Mexico, approximately 80% of oil-seeds for human food are imported so increasing sunflower production is an option to reduce this deficit. The Universidad Autónoma Agraria, Antonio Narro (UAAAN) has developed sunflower cultivars for the semi-arid regions. Understanding the relationship between biomass and seed yield and their correlation with leaf and seed proteins is important for plant selection. The objectives of this paper were to characterize the leaf and seed proteins from six sunflower cultivars and relate these to biomass and seed yield.
This study was carried out at the experimental field of the UAAAN at Buenavista, Saltillo, Coahuila, Mexico. Six sunflower cultivars were studied, four developed by the UAAAN Plant Breeding Department (SAN-3C, GORDIS, SANE 1278, and SANE 23578), and two Argentinean commercial cultivars (RIB 77 and KLM 123). Temperature, rainfall, and evapotranspiration data were recorded daily during the experiment (Table 1). The cultivars phenology was monitored during the crop development cycle.
Table 1. Temperatures and rainfall recorded in the growing area during the development cycle of six sunflower varieties.
Seeding was performed on June 10, 1996 in a randomized block design with six treatments (cultivars) and four replications. The plot size was 4 × 7 m for each of the 24 experimental units. The distance was 0.8 m between rows and 0.25 m between plants yielding 50,000 plants/ha. An 80N60P00K fertilizer of ammonium sulfate and simple superphosphate was applied prior to seeding. Each plot received heavy irrigation six days prior to seeding. Water was also applied at three and ten days after seeding to ensure plant emergence. During the remainder of the season, the available soil water (ASW) was 140 mm/100 cm depth, and rainfall (Rf) was 240 mm from seeding to physiologic maturity.
The level of water deficit (LWD) was defined (Blanchet and Merrien 1990) as:
where RET=Real evapotranspiration and PETcrop=Potential evapotranspiration of the crop.
According to equation 1, when LWD equals unity, no water deficit is observed and the total water requirements of the crop are satisfied. When RET is lower than PETcrop, a water deficit is observed; the lower the ratio value, the higher the LWD. In this paper, RET was considered equal to the sum of the total available soil water to 1 m depth (ASWtotal) plus Rf, hence equation 1 becomes:
|LWD =||ASWtotal + Rf
we also assumed that ASW is equally available at all moisture levels to satisfy water demand or potential conditions. This holds true until the water availability is exhausted. When soil moisture is depleted, rainfall was considered as RET during that period and:
The potential evapotranspiration of the crop was calculated as:
with EV = Pan evapotranspiration and Kc = Crop coefficient. The levels of water deficit for the varieties at the different growing stages were estimated by means of equations 24.
Variables monitored included biomass (total and by plant organ), grain yield, soluble protein concentrations, and molecular weight.
Sampling and measurements were performed at the different crop development stages identified according to the CETIOM scale (CETIOM-AFNOR 1986): Star (E1); bud-flowering (E3); flowering-start (F1); flowering-end (F4); physiologic maturity (M2); and harvest (M4).
Six plants per plot were sampled and separated by plant parts as leaves, petioles, stems, buds, or flowerhead. These parts were placed in aluminum pans and oven dried at 80°C for 72 hr.
Grain yield determination at harvest was performed by sampling all the plants in a 2 m2 area in two central furrows of the 4×7 m plot. Grain moisture was determined by oven-drying at 103°C for 16 hr.
Soluble proteins were determined using 712 leaves from four plants (Merrien et al. 1988) and stored at 20°C until analysis.
Leaf or grain samples were freeze-dried. The extraction and measurement of soluble proteins was performed according to the method reported by Bradford (1976). The material was frozen in liquid nitrogen and pulverized in a porcelain capsule. From this material, approximately a 300 mg sample was suspended in 25 ml of acetone at 4°C for 60 min and centrifuged at 15,000 rpm for 15 min. A second acetone extraction and centrifugation was performed to extract all pigments. Afterwards the material was suspended in 3 ml of phosphate buffer (pH 7) with 2% poly vinyl pyruvate (PVP), allowed to set for 30 min at 4°C and centrifuged at 15,000 rpm for 20 min. The supernatant (crude extract) was stored in refrigeration until analysis. A sample of 0.1 ml from the crude extract was diluted with 10 ml of phosphate buffer (pH=6); from this dilution 2.4 ml were extracted and 0.6 ml of staining reactive (Bio-Rad) added with the protein concentration measured at 595 nm by UV spectroscopy. The calibration curve was constructed using bovine g-globulin as a standard.
Denaturing electrophoresis was performed according to Laemmli (1970), in 12% acrylamide gel with SDS purchased from Bio-Rad and used without further purification. The gel mold was prepared by mixing water (33 ml), 40 ml acrylamide (30% solution), 25 ml Tris 1.5 molar (pH=8.8), 1 ml SDS (10% solution), 1 ml ammonium persulfate (10% solution) and 0.04 ml TEMED. The stacking gel with 5% acrylamide was prepared by mixing 20.4 ml water, 5.1 ml acrylamide (30% solution), 3.75 Tris 1.5 molar (pH=6.8), 0.3 ml SDS (10% solution), 0.3 ml ammonium persulfate (10% solution) and 0.03 ml TEMED.
Electrophoresis was performed at an initial voltage of 130V, raised to 180V when the tracking dye reached the gel mold; analysis time was 9 h. The protein molecular weight markers (Sigma Chemical Company) were: a2-macroglobulin (205,000 Da), b-galactoside (116,000 Da), phosphorilase (97,000 Da), albumin serum (66,000 Da), fumarase (48,500 Da), carbonic anhydrase (29,000 Da), b-lactoglobulin (18,400 Da), and a-lactalbumin (14,200 Da). Proteins were simultaneously fixed and stained using a solution containing 0.24 g Coomassie brilliant Blue R250 in 90 ml of a 1:1 (V/V) methanol:water and 10 ml of glacial acetic acid.
A calibration curve was constructed using the data from the markers, plotting the logarithm of molecular weight versus distance. The gel was stained with silver nitrate.
Simple correlations between the variables were estimated using the equation proposed by Miller et al. (1958) and a statistical software package (Mstat-C 1990).
Saumell (1976) reported that temperature affects oil content and composition. Water requirement for sunflower is 400500 mm according to Robles (1985), rainfall during the crop development was 240 mm. The number of days required by each variety to reach a particular growing stage is presented in Table 2.
Table 2. Days after sowing for different growing stages of six sunflower cultivars.
|Growing stages (days after sowing)|
There was no water deficit from seeding to the bud-flowering stage, since LWD (Eq. 24) was equal for all varieties. During the flowering-end to physiologic maturity period, SAN 3-C and KLM 123 suffered LWD of 0.63 (highest water stress) during three days, which represent 13% of the growth stage time. For the rest of the growth stage, rainfall supplied the plants so the LWD reached equilibrium. SANE 23578, SANE 1278, and RIB 77 had LWDs of 0.68 during 40% of the growth stage, but during the remaining 60% there was adequate water supplied by rainfall. Finally, GORDIS variety had an LWD of 0.77 (lowest water stress) during 38% of the growth stage, but no water deficit for the remainder of the growth stage.
The results for biomass accumulation (g/plant) for the above ground organs and total plant of GORDIS are presented in Fig. 1. Other cultivars behaved similarly. The leaf biomass was highest at 60 days after sowing (DAS) and 76 DAS for petioles. The stem and head plus grain biomass was highest at the harvest stage. The head plus grain biomass increased noticeable after 60 DAS continuing until harvest. The total biomass production for SAN 3-C was 153.9 g/plant, 143.36 g/plant for GORDIS, 138 g/plant for RIB 77, 118.9 g/plant for SANE 1278, 115.2 g/plant for KLM 123, and 97.1 g/plant for SANE 23578. SAN 3-C yielded the highest biomass, possibly due to its relative longer development cycle and short water stress period during the flowering-end to physiologic maturity stages.
These results showed that biomass production depended on cultivar and level of water deficit. The biomass accumulated in a classical growing kinetics, increasing in the stem up to harvest. There were translocation and redistribution of photosynthates from leaves and stem to the head and grain, mainly at the flowering stage, this agreeing with Merrien (1992).
Total biomass (g/plant) in the different growing stages for the six cultivars is presented in Fig. 2. The cultivars were plotted in increasing order of total biomass at the harvest stage. At the star and bud-flowering stages, biomass accumulation was low and very similar for the 6 varieties, but at the flowering-start and end, differences among varieties were observed. The differences were more evident at the harvest stage reaching a value of 58% for the difference between the highest (SAN 3-C) and the lowest (SANE 23578) production.
|Fig. 1. Biomass accumulation in GORDIS
sunflower. E1 = star, E3 = bud flowering, F1
= flowering start, F4 = flowering end, M2 = physiological
maturity, M4 = Harvest.
||Fig. 2. Total biomass accumulation for
six sunflower cultivars at different growth stages.
GORDIS had the highest grain yield with 78 g/plant followed by SAN 3-C with 56 g/plant, KLM 123 with 42.6 g/plant, RIB 77 with 39.8 g/plant, and SANE 1278 with 38.4 g/plant while SANE 23578 showed the lowest value with 29.6 g/plant. The high grain yield in GORDIS was possibly due to the fact that it had the lowest water stress during its development.
The soluble protein concentrations (mg/g, dry weight basis) in the leaves and grain at different growth stages is shown in Fig. 3. Two groups were differentiated at the flowering-end stage, one with relatively high protein concentrations (100217.1 mg/g) which included SAN 3-C, and SANE 1278, and the other group with concentrations below 100 mg/g included GORDIS, RIB 77, KLM 123, and SANE 23578. Protein concentration in the grain was similar for GORDIS (86.9 mg/g), SAN 3-C (75.9 mg/g), and RIB 77 (71.8 mg/g) and lower for KLM 123 (54.9 mg/g), SANE 1278 (39.7 mg/g), and SANE 23578 (31.3 mg/g).
Fig 3. Soluble protein concentration in the leaves and grain of six
sunflower cultivars at different growth stages.
GORDIS showed an almost constant leaf protein concentration during the development cycle and the highest concentration in grain (86.9 mg/g). SAN 3-C behaved in a different manner with the protein concentration at the star stage being low at 46.35 mg/g, but increasing to 217 mg/g at flowering-end. The protein concentration in the grain was slightly lower (75.9 mg/g) than in GORDIS. The rest of the cultivars behaved either like GORDIS or SAN 3-C. A decrease in leaf protein concentration for French cultivars from flowering to seed maturity has been reported by Martínez (1987). The protein content was lower (11.5 mg/g fresh weight basis at flowering) than the present results.
It should be noted that GORDIS suffered less water stress (highest LWD) at the flowering-end-physiologic maturity period, than the rest of the varieties; this might influence protein translocation yielding a higher protein concentration in grain. Water stress promoted a reduction in protein concentration (Quartacci and Navari-Izzo 1992), which supports our results.
Simple correlation analysis between soluble proteins in the grain and several other variables indicated a positive and highly significant correlations between total biomass at flowering-start (0.83**), total biomass at maturity (0.85**), biomass at harvest (0.91**) and grain yield (0.83**). This information may be useful for selecting higher yielding varieties for commercial sunflower production.
The SDS-PAGE profiles of the leaf and grain proteins in the crude extracts for all varieties at different growing stages showed a similar number of protein bands (twenty three) in the electrophoretograms with only the flowering-end stage shown. The difference among polypeptides is the relative intensity of the stained bands rather than its number. The highest molecular weight protein observed was 154,880 Da and the lowest 28,800 Da. There are reports on sunflower proteins showing bands at 21,00057,000 Da (Jiang et al. 1994). Three main groups of polypeptides were identified (Anisimova 1992): one basic (approximately 20,000 Da), and two acidic groups (30,000 and 40,000 Da), with variable composition. Polypeptides with molecular masses of 40,00080,000 Da, 37,000 Da, 29,000 Da, 20,000 Da, and 17,000 Da has been reported (Jung et al. 1993). The most representative bands of soluble proteins are 43,300 and 14,400 Da (Navari et al. 1992) and the bands at 60,000 and 56,000 Da increased (21% and 37% respectively) following water stress. These results concur with our study in the banding pattern observed since our plants were under water stress from flowering-end to physiologic maturity.
The electrophoretogram for leaf soluble proteins at the flowering-end stage for the six sunflower cultivars is shown in Fig. 4. There were 11 high intensity bands at: 148,000, 116,000, 106,000, 104,000, 95,500, 83,170, 79,000, 69,500, 57,500, 45,000, and 28,800 Da. Similar high intensity patterns were observed for KLM 123 and SANE 23578, whereas SAN 3-C and GORDIS showed low intensity bands. Considering that the band intensity is directly related to concentration, it is possible to conclude that KLM 123 and SANE 23578 have higher protein concentrations in the leaves than GORDIS and SAN 3-C. All cultivars showed very similar patterns despite the relative differences observed in the different growth stages for leaves protein electrophoretograms. We conclude that soluble proteins in the leaves have different relative concentrations depending on the growing stage and sunflower variety.
The proteins in the grain at harvest showed five high intensity bands at: 79,000, 77,000, 57,500, 45,000, and 28,800 Da (Fig. 5). The bands at 57,500, 45,000, and 28,800 Da were particularly noticeable and apparently not dependent on the variety. In this figure, there are two evident features: only the low molecular weight proteins (lower than 66,000 Da) produced high intensity bands, while the high molecular weight proteins were present in lower relative concentration. Proteins with 60,000, 54,000, 48,000, and 40,000 Da in sunflower grain have also been observed (Dalgalarrondo et al. 1984).
|Fig. 4. SDS polyacrylamide gel electrophoretogram of leaf soluble crude protein extracts of six sunflower cultivars at the flowering-end stage.||Fig. 5. SDS polyacrylamide gel electrophoretogram of the grain soluble crude protein extracts of six sunflower cultivars at the flowering-end stage.|
The results from Fig. 4 suggest that the low protein concentrations in GORDIS and SAN 3-C may be due to translocation of leaf soluble proteins to the grain. The relative concentration of leaf proteins was a function of the growing stage and cultivar, whereas in the grain apparently it is independent of cultivar.
Finally, we conclude that biomass, grain yield, and soluble protein concentration in the leaves and grain, at the different growing stages, were affected by the level of water deficit and cultivar. GORDIS and SAN 3-C were outstanding because they showed the highest values for these characteristics as well as high adaptation.