II.20. High lysine barley - a summary of the present research development in Sweden.
Lars Munck. The Swedish Seed Association, S-268 00 Svalöf, Sweden.
Hiproly barley (CI 3947) was found in the world barley collection (Hagberg and Karlsson 1969, Munck et al. 1969a) with the dyebinding (DBC) screening techniques (Mossberg 1969). Hiproly barley although high in protein content (about 16%) showed a protein and amino acid composition similar to low protein (about 10%) commercial lines (Munck et al. 1969b). This means increased content in % of Kjeldahl protein (Nx6.25) of albumines, lysine, asparatic acid and methionine and decreased content of prolamines, glutamic acid, cysteine and proline. It was found that these effects were due to a major change restricted to the endosperm part of the seed (Munck 1970) depending on a major gene segregating 1:3 (Munck et al. 1970). Significant deviations for 1:3 were found in some crosses pointing out the possibility of genetic modifiers. The gene was by linkage or as a pleiotropic effect associated with a morphological trait resulting in a strong starch protein adherence of meal preparations as seen in the light microscope (Munck et al. 1970). Analysis of segregating F2 populations in a phytotron experiment with two daylengths (Hagberg et al. 1970) revealed no difference between high lysine and normal segregants regarding daylength response, plant height, number of spikes per plant, numbers of seeds per plant. The 1000-kernel weight was, however, about 5 absolute grams lower as a mean in the high lysine part of the material. A significant amount of positive deviates was found among the high lysine lines. The shriveled Hiproly seed could be considerably improved in seed quality by crossing without affecting amino acid composition. Endosperms of different developmental stages from Hiproly and CI 4362 ("isogenic normal sisterline") were homoenized with an Ultraturrax at 2°C and extracted twice with water followed by 1.2% NaCl and 70% EtOH. Centrifugations were made in a high speed centrifuge and extracts were analysed with micro-Kjeldahl. The water extract was diluted with EtOH to a strength of 70% to determine the EtOH precipitable and EtOH soluble Kjeldahl protein. Hiproly showed increased ability to produce EtOH precipitable albumines especially at rather late stages in the endosperm development (30-40 days after anthesis) compared to CI 4362. At yellow ripening stage (about 40% water content) Hiproly total protein contained 11.3% of this fraction compared with 5.9% in CI 4362. This difference was 30% less after drying similar endosperms indicating denaturation effects on the proteins. The prolamines were strongly reduced in Hiproly protein - 26.5% compared with 38.3% in the sisterline. Electrophoresis studies of the EtOH soluble H20 fraction and the EtOh soluble proteins (prolamines) in polyacrylamid (7.5%) gels using beta-alanine buffer at pH 5.0 indicates differences in extractability of the catodic component between the two barley lines. A characteristic pattern of four bands soluble in EtOH and in 0.1 M formic acid appeared in the alcohol soluble subfraction of the water extract in Hiproly and in the EtOH (prolamine) fraction in CI 4362. Further studies regarding the anodic component of the albumines with electrophoresis using 15% gels in TRIS buffer at pH 8.6 confirm the great quantitative increase of this fraction introduced by the Hily gene (see also Munck 1972). These studies were made on single endosperm level from segregating populations, always with yellow ripe non dried frozen seeds (water content about 40%) to avoid denaturation effects. No certain qualitative differences could be ascertained. An extra band in the globulines of the Hiproly and Hily seeds could be found in the albumines of CI 4362 and normal segregants. Extracts from single endosperms (table 1) analysed with electrophoresis were confirmed with amino acid analysis, It was possible by analysing several seeds per plant to detect normal and high lysine homozygotes as well as heterozygotic plants (extracts on seed base). A segregation not significant: deviating from 1:3 was found in the F2 generation from the cross Hiproly x Kristina. The relative strength between different bands varied considerably between homozygotic high lysine plants indicating the presence of modifiers. It was also possible to isolate such lines with the DBC method (see table 2). Other amino acid levels changed in the Hiproly barley were also modified either towards the direction of the normal or Hiproly. These differences have been confirmed two years with several analyses on single plant and population bases.
The DBC screening analysis was tested on 235 F4 lines emerging from F2 plants homozygotic for the gene according to amino acid analysis. Twelve low DBC lines were found which were all found to be high in lysine according to amino acid controls in F4. Starch protein relationship showed the adherance type. These lines gave more coarse meal in milling with a hammer mill. Rechecking the lines in F5 another year did not confirm the low DBC values of F4.
Morphological studies on sections reveal that both Hiproly and CI 4362 have a dense protein subaleuron layer, comparable to that of hard wheats. This is not the case with the variety Kristina which more resembles the structure of soft wheats. These facts seem to affect milling characteristics, Hiproly and CI 4362 giving more coarse meal than Kristina.
Another morphological character separated from the above described is the starch protein adherence character, Kristina and CI 4362 giving good separation between starch granules and protein in meal preparations in contrast to Hiproly. Out of 235 high lysine lines in F4 seventeen lines gave low starch protein adherence as Kristina and CI 4362, Amino acid composition was not different from 15 control plants and the standard deviation of protein and amino acids g/16 g N were of the same magnitudes. Nevertheless, the usual negative correlation between e.g, lysine g/16 g N and crude protein as well as the positive correlation between e.g. glutamic acid g/16 g N and crude protein were only significant for the seventeen lines with deviating starch protein adherence, while there were no signs of such correlation for the control material. Tentatively the adherence trait could affect the timing of the protein synthesis so that low lysine prolaimines are starting and finishing earlier at higher water contents in the endosperm, independently of the amount of nitrogen translocated to the seed. Such a changed pattern as well as the adherence trait itself could also be related to the apparent smaller cell size of the Hiproly endosperm. The starch protein adherence character is difficult to use as an indicator for the high lysine gene in backcrosses which could indicate that it is inherited separately as a gene, linked with the high lysine gene.
Similar effects could be obtained with morphological modifiers which do not affect the overall amino acid composition, but which neutralize a tentative pleiotropic morphological side effect of the high lysine gene. A marked starch protein adherence effect was found in otherwise normal barley varieties (e.g. Monte Christo). Further integrated genetical-ultrastructural-biochemical work is needed to grasp this complex.
Nutritional experiments with mice and rats confirm the improved nutritional value of the new barley lines obtained from the crosses. An increase (at 9.4% protein level in diet) of net protein utilization (NPU) from 59.8 to 68.7% and biological value from 71.2 to 80.6% could be obtained for restrictively fed rats when lysine g/16 g N increased from 3.25 to 4.13. Protein content of the seeds were 12.6 and 12.4%. Neither the particle size of the barley meals nor the starch protein adherence character interacted significantly with the feeding results.
Mapping studies show that the high lysine gene is located in the 7th chromosome (K.-E. Karlsson, Barley Genetics Newsletter No. 2, 1972). The symbol lys is suggested for the gene referring to the practically important increase of the amino acid lysine which is limiting for the use of barley protein in e.g. pig feeds. It should be remembered that several other amino acids are affected by the gene. The symbol lys should be revised when the basic action of the gene is better understood
Contributions from the following persons and institutions are involved in this note: The Swedish Seed Association, Svalöf: A. Hagberg and G. Persson (plant breeding), K.-E. Karlsson (DBC screening, crossing, gene mapping), L. Munck (coordination, biochemistry, nutrition). Institute of Genetics, Lund: P. Malnoe, A. L. Tallberg and P. Knutsson (electrophoresis), K. Hazell (morphology). Institute of Biochemistry, Uppsala: D. Eaker and R. Thorzelius (amino acid analyses). National Institute of Animal Sciences, Copenhagen, Denmark: B. 0. Eggum (rat tests).
Table 1. Amino acid analyses of extracts from single endosperms with and without the Hily gene.
Literature:
Hagberg, A. and Karlsson, K.-E. 1969. IAEA (Vienna) Symp. Proceed. STI/PUB/212:17-21.
Hagberg, A., Karlsson, K.-E. and Munck, L. 1970. IAEA (Vienna) Symp. Proceed. STI/PUB/258:121-132.
Mossberg, R. 1969. IAEA (Vienna) Symp. Proceed. STI/PUB/212:151-160.
Munck, L. 1970. IAEA (Vienna) Symp. Proceed. STI/PUB/258:319-329.
Munck, L. 1972. American Chemical Society Symp. on seed proteins. Los Angeles, March 30, 1971 (In press)
Munck, L., Karlsson, K.-E. and Hagberg, Aw 1969a. J. Swed. Seed Ass. 79:196-205.
Munck, L., Karlsson, K.-E. and Hagberg, A. 1969b. Bartey Genetics II Pullman, Wash.:544-558.
Munck, L., Karlsson, K.-E., Hagberg, A. and Eggum, B. 0. 1970. Science 168:985-987.
