SUMMARY
Starch quality is one of the most important agronomic traits in rice (Oryza sativa L). In this study, we identified single nucleotide polymorphisms (SNPs) in the Waxy and Alk genes of eight rice varieties and their associations with starch physicochemical properties.vi.e.vamylose content (AC) and gelatinization temperature (GT). Seven Sri Lankan rice varieties, Pachchaperumal, Herathbanda, At 354, Bg 352, Balasuriya, H 6 and Bw 295-5 were detected as high amylose varieties while Nipponbare exhibited low amylose content. In silico analysis of the Waxy gene revealed that all tested Sri Lankan varieties possessed ‘G’ (Wxa allele) instead of ‘T’ in the first intron which could explain varieties with high and intermediate amylose content. All Sri Lankan varieties had ‘A’ instead of ‘C’ in exon 6 of the Waxy gene and this fact was tally with the varieties showing high amylose content. Therefore, possessing the Wxa allele in the first intron and ‘A’ in exon 6 could be used as a molecular marker for the selection of high amylose varieties as validated using several Sri Lankan varieties. All Sri Lankan varieties except, Bw 295-5 exhibited the intermediate type of GT which could not be explained using the so far reported allelic differences in the Alk gene. However, Bw 295-5 which is a low GT variety had two nucleotide polymorphisms in the last exon of the Alk gene, i.e. ‘G’ and ‘TT’ that represent low GT class. Therefore, it can be concluded that sequence variations of Waxy and Alk genes reported in this study are useful in breeding local rice varieties with preferential amylose content and GT class.Key word Alk gene, amylose content, single nucleotide polymorphism, Waxy gene.INTRODUCTIONRice (Oryza sativa L.) is one of the leading food crops of the world. More than half of the world’s population relies on rice as the major daily source of calories and protein (Sartaj and Suraweera, 2005). After grain yield, quality is the most important aspect of rice breeding. Grain size and shape largely determine the market acceptability of rice, while cooking quality is influenced by the properties of starch. In rice grains starch is the major component that primarily controls rice quality. Starch consists of two forms of glucose polymers, relatively unbranched amylose and a highly branched amylopectin. Starch-synthesizing genes may contribute to variation in starch physicochemical properties because they affect the amount and structure of amylose and amylopectin in rice grain (Kharabian-Masouleh et al., 2012). Amylose content (AC), gelatinization temperature (GT) and gel consistency (GC) is the three most important determinants of eating and cooking quality. Amylose content is the ratio of amylose amount present in endosperm to total starch content. Rice varieties are grouped based on their amylose content into waxy (0-2%), very low (3-9%), low (10-19%), intermediate (20-25%), and high (> 25%) (Kongseree and Juliano, 1972). The most widely used method for amylose determination is a colorimetric assay where iodine binds with amylose to produce a blue-purple color, which is measured spectrophotometrically at a single wavelength (620nm). Low amylose content is usually associated with tender, cohesive and glossy cooked rice; while, high amylose content is associated with firm, fluffy and separate grains of cooked rice. The Waxy (Wx) gene, which encodes granule-bound starch synthase I (GBSSI), is the major gene controlling AC in rice (Nakamura, 2002). The Waxy gene is located on chromosome six and various single nucleotide polymorphisms (SNPs) of Wx were found, including a ‘G’ to ‘T’ SNP of the first intron, ‘A’ to ‘C’ SNP of the sixth exon and ‘C’ to ‘T’ SNP of the tenth exon (Larkin and Park, 2003). The ‘AGGTATA’ sequence at the 5’splice-junction coincides with the presence of the Wxa allele, while the ‘AGTTATA’ sequence coincides with the presence of the Wxb allele. Therefore, all intermediate and high amylose cultivars had ‘G’ nucleotide while low amylose cultivars had ‘T’ nucleotide at the putative leader intron 5' splice site. The cytosine and thymidine (CT) dinucleotide repeats in the 5’- untranslated region (UTR) of the Waxy gene were reported to be a factor associated with AC. However, the relationship between these polymorphisms and amylose contents is not clear. Amylopectin chain length distribution plays a very important role to determine GT in cooked rice. The time required for cooking is determined by the gelatinization temperature of starch. It is important because it affects the texture of cooked rice and it is related to the cooking time of rice. The gelatinization temperature is estimated by the alkali digestibility test. It is measured by the alkali spreading value (ASV). The degree of spreading value of individual milled rice kernels in a weak alkali solution (1.7% KOH) is very closely correlated with gelatinized temperature. According to the ASV, rice varieties may be classified as low (55 to 69°C), intermediate (70 to 74°C) and high (> 74°C) GT classes. In a breeding program ASV is extensively used to estimate the gelatinization temperature. The synthesis of amylopectin is more complex than that of amylose. Polymorphisms in the starch synthase IIa (SSIIa) gene which is recognized as the Alk gene are responsible for the differences in GT in rice (Umemoto and Aoki, 2005; Waters et al., 2006). Two single nucleotide polymorphisms (SNPs) in the last exon of the Alk gene are responsible for the differences in GT in rice. The biochemical analysis clearly showed that the function of the amino acids caused by these two SNPs is essential for SSIIa enzyme activity (Nakamura et al., 2005) and those are ‘G’/‘A’ SNP at 4424 bp position and ‘GC’/‘TT’ SNPs at 4533/4534 bp position with reference to Nipponbare rice genomic sequence. Based on the SNPs, Low SSIIa enzyme activity results in S-type amylopectin, which is enriched in short chains whereas high SSIIa enzyme activity produces L-type amylopectin (Umemoto et al., 2004). Therefore, the combination of ‘G’ at SNP3 and ‘GC’ at SNP4 is required to produce L-type rice starch and this has a higher GT relative to S-type starch. GC is a standard assay that is used in rice improvement programs to determine the texture of softness and firmness in high amylose rice cultivars. Intermediate and low amylose rice usually has soft gel consistency. Sequence variation in exon 10 of the Waxy gene associates with GC (Tran et al., 2011).OBJECTIVES The objectives of this study were to detect polymorphisms in major starch synthesizing genes among several rice cultivars as models and to determine the relationship between their SNP variations and starch physicochemical properties. Also, we analyzed major starch synthesizing gene sequences of several Sri Lankan rice varieties in silico aiming at utilizing this information in rice breeding programs.MATERIALS AND METHODSPlant materials: Seeds of eight Oryza sativa L. accessions were obtained from the Rice Research and Development Institute (RRDI), Bathalagoda, Sri Lanka and Gene Bank of Plant Genetic Resource Center (PGRC), Gannoruwa.Characterization of grain physical parameters: Grain length and width were determined using a vernier caliper. Ten grains from each sample were collected randomly and measured to obtain the average length and width of the milled rice. The average length and width were recorded as their length and width. Based on the length and width of the grains, the milled rice grains were classified into four classes (table 1) according to the method accepted by RRDI Bathalagoda, Sri Lanka.According to the scale L/S – Long Slender, L/M – Long Medium, I/B – Intermediate Bold and S/R –Short RoundAnalysis of amylose content: Initially, rice samples were dehusked and polished prior to milling. Ten whole – milled rice kernels of eight rice samples were ground separately by using mortar and pestle. Amylose content per 100 mg was determined by measuring the blue value of rice varieties as described by Juliano (1971). About 100mg rice sample was shifted into a 100 mL volumetric flask and 1mL of 95% ethanol was added. Then 9mL of 1N NaOH was added and the content was boiled for 20min. at boiling temperature to gelatinize the starch. After cooling the content, the volume was made up to 100mL and 5mL of starch solution was pipetted out into a 100mL volumetric flask. The blue color was developed by adding 1mL of 1N acetic acid and 2 mL of iodine solution (0.2g iodine and 2.0g potassium iodine in 10 mL aqueous solution). Then volume was made up to 100mL with distilled water and the solution was kept for 20min. after shaking. Finally, the absorbance of the solution was measured at 620nm using Spectrophotometer T80 (PG Instruments Limited) as described by Juliano (1971). The standard curve was prepared using 40mg of potato-amylose to calculate the amylose content of rice varieties through absorbance values. Forty mg of potato amylose was put into a 100 mL of volumetric flask and 1ml of 95% ethanol and 9mL of NaOH were added and content was heated for 20min at boiling temperature. After cooling the content volume of the solution was made up to 100mL using distilled water. Then 1mL, 2mL, 3mL, 4mL and 5mL of amylose solution were pipetted out into 100mL flasks. Then 0.2mL, 0.4mL, 0.6mL, 0.8mL and 1mL of 1N acetic acid were added to the flasks respectively. Finally, 2mL of iodine solution was added to each flask and volume was made up to 100mL with distilled water. Solutions were stood up for 20min. after shaking and absorbance values were measured at 620nm. Measured absorbance values were plotted at 620nm against the concentration of anhydrous amylose (mg).Analysis of gelatinization temperature: GT was indirectly measured on rice by the alkali spreading value. Husked and polished seeds per accession were used for the analysis. Selected duplicate sets of six milled grains without cracks of each sample were put into Petri dishes. About 10mL of 1.7% KOH was added and grains were spread in the petri dish to provide enough space. The constant temperature at 30°C was maintained to ensure better reproducibility. After 23hrs, the degree of disintegration was quantified by a standard protocol with a numerical scale of 1–7 (table 2) as reported by Cruz and Khush (2000). As reported by Juliano (2003), GT of rice was determined using the alkaline spreading scale, where 1.0-2.5: High (74-80 °C), 2.6-3.4: High-intermediate (70-74 °C), 3.5-5.4: Intermediate (70-74 °C) and 5.5-7.0 Low: (55-70 °C).Bioinformatics and statistical analysis: The available literature was used to identify the most likely candidate genes associated with rice starch quality and their SNPs of each gene (Hirose et al., 2006; Waters and Henry, 2007; Tran et al., 2011). In all the tested varieties except Bg 352 and At 354, the DNA sequence of each gene was retrieved from the Rice SNP Seek database (http://snp-seek.irri.org/). The gene sequences of At 354 and Bg 352 were obtained from the National Research Council 16-016 project, Wayamba University of Sri Lanka. Multiple sequence alignment was conducted for the DNA sequence using Clustal Omegavsoftware (https://www.ebi.ac.uk/Tools/msa/clustalo/). Starch physiochemical data obtained were subjected to a one-way analysis of variance (ANOVA) followed by Duncan’s New Multiple Range Test (DNMRT) to determine the statistical differences among varieties at the significance level of p = 0.05. Statistical analysis was done using SAS version 9.1 (SAS, 2004).ESULTS AND DISCUSSION: Physical properties of rice grains: Physical properties such as length, width, size, shape and pericarp color of rice grains obtained from eight different rice varieties are given in table 3. Classification of rice grains was carried out, according to their sizes and shapes based on Juliano (1985). The size of the rice grains was determined as per grain length while grain shape was determined by means of length and width ratio of the rice kernel. In the local market, rice is classified as Samba (short grain), Nadu (intermediate grain) and Kora (long/medium) based on the size of the grain (Pathiraje et al., 2010). Lengths of rice kernels were varied from 5.58 to 6.725 mm for all varieties. The highest grain length and width were given by At 354 and Pachchaperumal respectively. The varieties, Bw 295-5 and H 6 showed a length: width ratio over 3 which is considered as slender in grain shape. Bw 295-5, H 6, At 354, Bg 352 and Nipponbare possessed white pericarp and others possessed red pericarp.Relationship between amylose content and SNPs variation of waxy loci in selected varieties: Amylose content was measured in seven Sri Lankan rice varieties and one exotic rice variety. Amylose content of the evaluated varieties varied significantly with p = 0.05 with the lowest of 15.11% and highest of 28.63% which were found in Nipponbare and Bw 295-5, respectively (table 4). The majority of the evaluated varieties fell into the high AC category (between 25-28%). Only Nipponbare could be clearly categorized under the low amylose group (table 4). The amylose content of Bg 352, Pachchaperumal and Herathbanda have already been determined by early studies of Rebeira et al. (2014) and Fernando et al. (2015). Most of the data obtained in the present experiment has agreed with the results of previous studies. Major genes such as Waxy and their functional SNPs have a major influence on amylose in rice (Nakamura et al., 2005). Accordingly, single nucleotide polymorphism, ‘G’/‘T’, at the 5’ leader intron splice site of the GBSSI has explained the variation in amylose content of varieties. Accordingly, high and intermediate amylose varieties have ‘AGGTATA’ while low amylose varieties have the sequence ‘AGTTATA’, which might lead to a decrease in the splicing efficiency. Therefore, the GBSSI activity of Nipponbare might be considerably weak and resulted in starch with low amylose content. Hence, producing ‘G’/‘T’ polymorphism clearly differentiates low amylose rice varieties, as reported by Nakamura et al. (2005). In GBSSI, Larkin and Park (2003) identified an ‘A’/‘C’ polymorphism in exon 6 and a ‘C’/‘T’ polymorphism in exon 10 which resulted in non- synonymous amino acid change. Chen et al. (2008) reported that the non-synonymous ‘A’/‘C’ SNP at exon 6 had the highest possible impact on GBSSI. Accordingly, the ‘A’/‘C’ polymorphism in exon 6 causes a tyrosine/serine amino acid substitution while the ‘C’/‘T’ polymorphism in exon 10 causes a serine/proline amino acid substitution. In view of this information, there is a relationship between the polymorphism detected by in silico analysis and amylose content obtained from our experiment. Out of the eight tested rice varieties, only one variety, Nipponbare was categorized as low amylose variety (10-19%) and it exhibited ‘T’ nucleotide at the intron splice site (table 4; figure 1). Varieties such as Pachchaperumal, Balasuriya, Bw 295-5, H 6, Herathbanda, At 354 and Bg 352 which contained high amylose (> 25%), had ‘G’ and ‘A’ nucleotides at intron splice site and exon 6 respectively (table 4; figure 1). The predominant allelic pattern of intron splice site and exon 6 are different in varieties containing intermediate amylose content (20-25%) which showed ‘G’ and ‘C’ nucleotides respectively. Of these selected rice varieties, none of the intermediate type amylose variety was found.Relationship between gel consistency and SNPs variation in Waxy loci: In this study, GC data of Herathbanda, Hondarawalu, Kuruluthuda, Pachchaperumal and Bg 352 were obtained from Fernando et al. (2015). The results of Tran et al. (2011) showed that the exon 10 ‘C’/‘T’ SNP of Wx has mainly affected GC. Accordingly, rice with a ‘C’ at exon 10 had soft and viscous gels once cooked. However, a sample with a ‘T’ had short and firm gels. In this study, Herathbanda, Hondarawalu, Kuruluthuda and Pachchaperumal had ‘C’ nucleotide and Bg 352 had ‘T’ nucleotide in exon 10 (table 5; figure 2). However, ‘C’/‘T’ substitution analysis could not be used to explain the GC of tested varieties.Relationship between gelatinized temperature and SNPs variation of Alk loci in selected rice varieties: Although there were differences in the scores, the degree of disintegration of all samples was saturated at 23 hrs. Most of the selected rice varieties showed the intermediate disintegration score. Varieties, Pachchaperumal, Balasuriya, H 6, Herathbanda, At 354 and Bg 352 were categorized into intermediate GT class (70–74°C) as indicated by an alkali spreading (AS) value of 5 (table 6; figure 3). Nipponbare and Bw 295-5 showed the highest disintegration score indicating the dispersion of all grains. Hence these varieties were categorized into low GT class (55-69°C) as indicated by an AS value of 6 (table 6; figure 3). However, high GT class rice varieties (> 74°C) were not found in the tested samples. Chromosomal mutation within the Alk gene has led to a number of single nucleotide polymorphisms (SNPs). Umemoto et al. (2004) identified four SNPs in Alk gene. Thus, SNP3 and SNP4 may be important genetic polymorphisms that are associated with GT class. According to the SNP3 and SNP4, eight rice varieties could be classified into either high GT or low GT types. If there is ‘A’ instead of ‘G’ at 4424 bp position of Alk gene with reference to Nipponbare rice genomic sequence, it codes methionine instead of valine amino acid residue in SSIIa, whilst two adjacent SNPs at bases 4533 and 4534 code for either leucine (‘GC’) or phenylalanine (‘TT’). Rice varieties with high GT starch had a combination of valine and leucine at these residues. Rice varieties with low GT starch had a combination of either methionine and leucine or valine and phenylalanine at these same residues. Nipponbare carried the ‘A’ and ‘GC’ nucleotides, while Bw 295-5 carried the ‘G’ and ‘TT’ nucleotides. Hence these varieties were classified into low GT class. Varieties such as Pachchaperumal, Balasuriya, H 6, Herathbanda, At 354 and Bg 352 carried ‘G’ and ‘GC’ nucleotides and these varieties were classified into high GT rice varieties. However, intermediate GT status could not be determined by SNP3 and SNP4 mutation of Alk gene (table 6; figure 4).In silico analysis of the polymorphisms in GBSSI gene and Alk genes of rice varieties retrieved from Rice-SNP-database: In this study, GBSSI gene and Alk gene were compared with the sequences retrieved from the Rice-SNP-Seek database to validate the SNPs further. As previously reported by Ayres et al. (1997), all low amylose varieties had the sequence ‘AGTTATA’ in exon 1. In agreement with preliminary work done by Larkin and Park (2003), all of the intermediate amylose varieties have the allelic pattern of GCC. All of the high amylose varieties have either the GAC or GAT allele of GBSSI. Among 42 rice accessions with the Sri Lankan pedigree, four allelic patterns were found; TAC, GCC, GAC and GAT (table 7). In this allelic pattern, the first letter corresponds to the ‘G’/‘T’ polymorphism in 5’ leader intron splice-junction, the second letter corresponds to the ‘A’/‘C’ polymorphism in exon 6 and the third letter corresponds to the ‘C’/‘T’ polymorphism in exon10 of Waxy gene. Analysis of the ‘G’/‘T’ polymorphism in the Wx locus showed that 41 rice cultivars shared the same ‘AGGTATA’ sequence at the 5’ leader intron splice-junction. But only 1 rice cultivar, Puttu nellu was found with ‘T’ nucleotide in intron1/exon1 junction site, which could be categorized as a low amylose variety (table 7). As discussed above, varieties with an intermediate level of apparent amylose could be reliably distinguished from those with higher apparent amylose based on a SNP in exon 6. Hence, only three rice varieties Nalumoolai Karuppan, Pannithi and Godawel with ‘C’ nucleotide in exon 6 exhibited the possibility of containing intermediate amylose content (table 7). High activity of GBSSI produces high amylose content leading to a non-waxy, non-sticky or non-glutinous phenotype. Therefore, according to the in silico genotypic results, rest of the 38 rice varieties may produce high amylose content in the endosperm (table 7). Proving this phenomenon. Abeysekera et al. (2017) has reported that usually, most of Sri Lankan rice varieties contain high amylose content. Targeted sequence analysis of exon 8 of the Alk gene in 42 different rice cultivars were found with three SNP polymorphisms that resulted in a changed amino acid sequence and, of these three SNPs, two SNPs were reported to be correlated with possible GT differences. Accordingly, Puttu nellu and 3210 rice varieties carried the ‘G’ and ‘TT’ nucleotides in SNP3 and SNP4 respectively (table 7). Hence these varieties can be classified into low GT class and except these two; other rice varieties carried the ‘G’ and ‘GC’ nucleotides in SNP3 and SNP4 respectively. Therefore, those varieties can possibly be classified into high GT rice varieties (table 7). However, further experiments are necessary to check the phenotypic variations for grain amylose content and GT class of in silico analyzed rice varieties. CONCLUSION Present results revealed the relationship between SNPs variation at Waxy loci and the amylose content of selected rice varieties. Accordingly, Pachchaperumal, At 354, Bg 352, Herathbanda, H 6, Balasuriya and Bw 295-5 with high amylose content had ‘G’ instead of ‘T’ in the first intron exhibiting the presence of Wxa allele with reference to Nipponbare which had low amylose content. Also all tested varieties had ‘A’ in exon 6 of the Waxy gene. Thus present findings i.e. presence of Wxa allele and SNP ‘A’ in exon 6 could be used as a potential molecular marker for the selection of high amylose varieties. In addition, Bw 295-5 which is a low GT variety, had two SNPs variations in the last exon of the Alk gene i.e. ‘G’ and ‘TT’ which is likely to be used to represent low GT class. Accordingly, sequence variations identified in Waxy and Alk genes could be utilized in the future rice breeding programs for the development of varieties with preferential amylose content and GT class.ACKNOWLEDGMENTSDirector and staff of the Gene Bank, Plant Genetic Resources Center, Gannoruwa are acknowledged for giving rice accessions.CONFLICT OF INTERESTAuthors have no conflict of interest.REFERENCESAbeysekera, W., G. Premakumara, A. Bentota and D. S. Abeysiriwardena, 2017. Grain amylose content and its stability over seasons in a selected set of rice varieties grown in Sri Lanka. Journal of agricultural sciences Sri Lanka, 12(1): 43-50.Ayres, N., A. McClung, P. Larkin, H. Bligh, C. Jones and W. Park, 1997. Microsatellites and a single-nucleotide polymorphism differentiate apparentamylose classes in an extended pedigree of us rice germ plasm. Theoretical applied genetics, 94(6-7): 773-781.Chen, M.-H., C. Bergman, S. Pinson and R. Fjellstrom, 2008. Waxy gene haplotypes: Associations with apparent amylose content and the effect by the environment in an international rice germplasm collection. Journal of cereal science, 47(3): 536-545.Cruz, N. D. and G. Khush, 2000. Rice grain quality evaluation procedures. Aromatic rices, 3: 15-28.Fernando, H., T. Kajenthini, S. Rebeira, T. Bamunuarachchige and H. Wickramasinghe, 2015. Validation of molecular markers for the analysis of genetic diversity of amylase content and gel consistency among representative rice varieties in sri lanka. Tropical agricultural research, 26(2): 317-328.Hirose, T., T. Ohdan, Y. Nakamura and T. Terao, 2006. Expression profiling of genes related to starch synthesis in rice leaf sheaths during the heading period. Physiologia plantarum, 128(3): 425-435.Juliano, B., 1971. A simplified assay for milled rice amylose. Journal of cereal science today, 16: 334-360.Juliano, B. O., 1985. Rice: Chemistry and technology. The american association of cereal chemists. Inc. St. Paul, Minnesota, USA, 774.Juliano, B. O., 2003. Rice chemistry and quality. Island publishing house. Island publishing house, Manila: 1-7.Kharabian-Masouleh, A., D. L. Waters, R. F. Reinke, R. Ward and R. J. Henry, 2012. Snp in starch biosynthesis genes associated with nutritional and functional properties of rice. Scientific reports, 2(1): 1-9.Kongseree, N. and B. O. Juliano, 1972. Physicochemical properties of rice grain and starch from lines differing in amylose content and gelatinization temperature. Journal of agricultural food chemistry, 20(3): 714-718.Larkin, P. D. and W. D. Park, 2003. Association of waxy gene single nucleotide polymorphisms with starch characteristics in rice (Oryza sativa L.). Molecular Breeding, 12(4): 335-339.Nakamura, Y., 2002. Towards a better understanding of the metabolic system for amylopectin biosynthesis in plants: Rice endosperm as a model tissue. Plant cell physiology, 43(7): 718-725.Nakamura, Y., P. B. Francisco, Y. Hosaka, A. Sato, T. Sawada, A. Kubo and N. Fujita, 2005. Essential amino acids of starch synthase iia differentiate amylopectin structure and starch quality between Japonica and Indica rice varieties. Plant molecular biology, 58(2): 213-227.Pathiraje, P., W. Madhujith, A. Chandrasekara and S. Nissanka, 2010. The effect of rice variety and parboiling on in vivo glycemic response. Journal of tropical agricultural research, 22(1): 26-33.Rebeira, S., H. Wickramasinghe, W. Samarasinghe and B. Prashantha, 2014. Diversity of grain quality characteristics of traditional rice (Oryza sativa L.) varieties in sri lanka. Tropical agricultural research, 25(4): 470-478.Sartaj, I. Z. and S. A. E. R. Suraweera, 2005. Comparison of different parboiling methods on the quality characteristics of rice. Annals of the Sri Lankan Department of Agriculture, 7: 245-252.Tran, N., V. Daygon, A. Resurreccion, R. Cuevas, H. Corpuz and M. Fitzgerald, 2011. A single nucleotide polymorphism in the waxy gene explains a significant component of gel consistency. Theoretical applied genetics, 123(4): 519-525.Umemoto, T. and N. Aoki, 2005. Single-nucleotide polymorphisms in rice starch synthase iia that alter starch gelatinisation and starch association of the enzyme. Functional plant biology, 32(9): 763-768.Umemoto, T., N. Aoki, H. Lin, Y. Nakamura, N. Inouchi, Y. Sato, M. Yano, H. Hirabayashi and S. Maruyama, 2004. Natural variation in rice starch synthase iia affects enzyme and starch properties. Functional plant biology, 31(7): 671-684.Waters, D. L. and R. J. Henry, 2007. Genetic manipulation of starch properties in plants: Patents 2001-2006. Recent patents on biotechnology, 1(3): 252-259.Waters, D. L., R. J. Henry, R. F. Reinke and M. A. Fitzgerald, 2006. Gelatinization temperature of rice explained by polymorphisms in starch synthase. Plant biotechnology journal, 4(1): 115-122.