IDENTIFICATION OF POLYMORPHIC MARKERS FOR BREEDING OF RICE TOLERANT TO PHOSPHORUS DEFICIENCY 1)

Transcription

1 Identification Indonesian Journal of polymorphic of Agriculture markers 3(1), for 2010: breeding 1-8of rice... 1 IDENTIFICATION OF POLYMORPHIC MARKERS FOR BREEDING OF RICE TOLERANT TO PHOSPHORUS DEFICIENCY 1) Joko Prasetiyono a), Hajrial Aswidinnoor b), Sugiono Moeljopawiro a), Didy Sopandie b), and Masdiar Bustamam a) a) Indonesian Center for Agricultural Biotechnology and Genetic Resources Research and Development Jalan Tentara Pelajar No. 3A, Bogor 16111, West Java, Phone: (0251) , , Facs.: (0251) borif@indo,net.id b) Bogor Agricultural University, Kampus IPB, Dramaga, Bogor 16680, West Java ABSTRACT Information on polymorphisms among rice parents is very important in rice breeding for tolerance to phosphorus deficiency. A study was conducted at Molecular Biology Laboratory of the Indonesian Center for Agricultural Biotechnology and Genetic Resources Research and Development, Bogor, West Java, from October 2006 to July 2007 to identify polymorphic markers of six rice genotypes. The rice genotypes, i.e. Dodokan, Situ Bagendit, Batur, Kasalath, NIL-C443, and K were analyzed for their polymorphisms using 496 SSR markers that covered the rice genomes. Seven of the 496 markers were used as foreground and recombinant selection markers, whereas the rests were used as background selection markers. PCR amplifications were separated on a 5% polyacrylamide gel and colored by the silver staining method. Three different markers among the seven foreground and recombinant selection markers, which are tightly linked with Pup1 gene and have a distance of less than 5 cm, were selected from each crossing. These markers were Dodokan vs Kasalath (RM277, SSR3, RM519), Dodokan vs NIL-C443 (RM277, SSR3, RM519), Dodokan vs K (RM277, SSR3, RM519), Situ Bagendit vs Kasalath (RM28102, SSR3, RM519), Situ Bagendit vs NIL-C443 (RM28102, SSR3, RM519), Situ Bagendit vs K (RM511, SSR3, RM519), Batur vs Kasalath (RM277, RM1261, RM519), Batur vs NIL-C443 (RM277, RM1261, RM519), and Batur vs K (RM28102, SSR3). Variations in the background selection primers were found in each chromosome and in each parent combination. Primers on chromosome 4, 5, and 12 showed the lowest polymorphisms. More primers are needed for these chromosomes. [Keywords: Oryza sativa, plant breeding, genetic markers, phosphorus defficiency] INTRODUCTION Soil mineral deficiency and toxicity are major problems in the world. In Asia, about 50% of the rice fields are experiencing phosphorus (P) deficiency. In Indonesia, the land area of acid soil with P deficiency is estimated to 1) Article in bahasa Indonesia has been published in Jurnal AgroBiogen Vol. 4 No. 2, 2008, p reach million ha or approximately 24.3% of the total land available (BPS 2004). P deficiency constitutes a major problem in agriculture, where farmers often cannot afford to buy fertilizers and phosphate, while the acid soils have high P binding capacity. In acid soils, the P element will be bound by iron (Fe) under submerged conditions, and will be bound by aluminum (Al) under dry conditions. P deficiency will also reduce the plant tillers, the rate of leaf expansion, and the rate of assimilate production per leaf area (Radin and Eidenbock 1984). Use of rice varieties tolerant to low P in marginal lands is a more strategic choice than to change the land condition with liming and fertilization which is often not practical and economical and can even threat sustainable agriculture. Marker assisted backcrossing (MABc) is the use of molecular markers for selection (marker assisted selection, MAS or marker assisted breeding, MAB), which in its development, combined with a backcross breeding program. MABc shortens the breeding time to get varieties or lines with the desired properties. According Ribaut and Hoisington (1998), using the MABc method restored 98% of plant genomes such as the parent restorers of elders needed only twice backcross, whereas the conventional method (without the aid of molecular markers) required 4-5 time backcrosses. If only one gene segment without any interference from other follower genes (no linkage drag), when not aided by molecular marker, 100 times of backcrossing are required and it took 50 years, whereas when using an improved MABc method (using foreground, recombinant, and background markers) only BC2F3 population is needed. It is very beneficial in plant breeding because it can shorten the time to develop new varieties. Further development of the MABc method is the use of markers across the whole chromosomes in addition to markers which are linked to the desired gene. This method is known as marker assisted advanced backcrossing (AdvMABc). The method involves two stages. The first stage is selection of results from crosses using markers tightly linked to the desired gene, called the foreground selection and recombinant selection. Foreground selection is to use a marker (or several) which is tightly linked to the

2 2 Joko Prasetiyono et al. desired properties (if a gene in question is already known, it can be used a marker for the gene). Recombinant selection is selection by using two markers between foreground markers. These recombinant markers are located adjacent to the recombinant gene segment in question. The second stage is to use as many markers scattered throughout the chromosome. This stage is called background selection. The use of background selection will accelerate the recovery of the parental restorer genomes. Individuals that generate dominant homozygote markers will be selected for the next step of crossing. The background selection has two objectives, namely (1) reducing the proportion of donor genome on chromosome carrying the donor, and (2) reducing the donor genome in other chromosomes (Friscth et al. 1999). Research that explores the existence of genes that regulate tolerance to P deficiency has been conducted since 10 years ago. The first publication on exploration of markers that are tightly linked to the nature of P deficiency in rice was reported by Wissuwa et al. (1998). In their research, near isogenic lines (NIL) population from crosses of Kasalath and Nipponbare and RFLP markers of C498 (chromosome 6) and C443 (chromosome 12) were used to make the selection. At the end of the study, NIL-C498 population genetically contains 96% Nipponbare, whereas NIL-C443 population contains 91% Nipponbare (Wissuwa and Ae 2001). Wissuwa et al. (2002) continued their research by transferring QTLs from the previous studies, namely by designing Pup1, and found it in chromosome 12 with a distance of 3 cm (centimorgan) between markers S14025 and S In the end, Pup1 mapped a better way within 0.6 cm distance, and physical mapping is now placing Pup1 in a 240 kb sequence in three bacterial artificial chromosome (BAC) clones. The Pup1 place was identified and acquired approximately 30 genes (putative) that allegedly was involved in setting Pup1. NIL-C443 that carries the donor allele from Kasalath, which increases the binding of P, can be used as parental donors. The study aimed to obtain polymorphic markers between the Indonesian rice parents sensitive to P deficiency (Dodokan, Situ Bagendit, Batur) and the IRRI rice parents tolerant to P deficiency (Kasalath, NIL-C443, K ) for use in the selection of rice hybrids. MATERIALS AND METHODS The study was conducted at the Laboratory of Molecular Biology and Greenhouse of the Indonesian Center for Agricultural Biotechnology and Genetic Resources Research and Development (ICABIOGRAD), Bogor, West Java from October 2006 to July Rice plant materials used in the study came from three Indonesian upland rice varieties (Dodokan, Situ Bagendit, and Batur) and three rice genotypes from IRRI (Kasalath, NIL-C443 [NIL-Pup1], and K ). All the three upland rice varieties used are susceptible to P deficiency based on preliminary research results. Total number of primers used for this research was 496 primers, consisting of 7 primers for foreground and recombinant selections and 489 primers for background selection (sequences of the 489 primers can be found in The primers used in the foreground and recombinant selections are shown in Table 1. Rice leaves were taken from 2 week-old plants of the six parent varieties. DNA of the leaves were isolated by miniprep using a modified method of Dellaporta et al. (1983). A section of fresh leaf was placed in a 2-ml microtube, poured with liquid nitrogen, and then pulverized into powder by using a chopstick. Extract buffer solution which contained Na-disulphite (0.8 g/100 ml extract buffer) was added into the microtube containing the leaf powder, followed by heating in a water bath at a temperature of 65 C for 15 minutes. The microtube was shuffled for 5 minutes and 500 µl chloroform isoamilalcohol (chisam) was added. The tube was shaken by hand or machine slowly for minutes, and then centrifuged at 12,000 rpm for 15 minutes. Supernatant DNA (solution above) was Table 1. Primers used in the foreground and recombinant selections of rice, ICABIOGRAD, Bogor, West Java, Primer Chromosome Forward Reverse Remark RM CGGTCAAATCATCACCTGAC CAAGGCTTGCAAGGGAAG Flanking marker RM CACTAATTCTTCGGCTCCACTTTAGG GTGGAAGCTCCGAGAAAGTGC RM GTCCATGCCCAAGACACAAC GTTACATCATGGGTGACCCC SSR3 1 2 xxxxx xxxxx RM ACTCCACTATGACCCAGAG GAACAATCCCTTCTACGATCG RM CTTCGATCCGGTGACGAC AACGAAAGCGAAGCTGTCTC RM AGAGAGCCCCTAAATTTCCG AGGTACGCTCACCTGTGGAC Flanking marker xxxxx = unpublished, SSR3 was developed from new markers located in between RM277 and RM519 markers.

3 Identification of polymorphic markers for breeding of rice... 3 transferred into a new 2-ml plastic tube and added with 96% ethanol or isopropanol to reach a 2-ml volume. If appropriate (DNA liquid colored brown), before adding absolute ethanol or isopropanol, Na or K acetate as many as 10% of the volume of the solution may be added to the mixture. The microtube was then incubated for 1 hour in a freezer (-20 C) and centrifuged at 12,000 rpm for 10 minutes. After centrifugation, the supernatant was discarded and the DNA was dried using a vacuum machine. The DNA pellet was dissolved in µl of 1 x TE buffer according to the amount of pellet. Quantity and quality of DNA was not measured, but the DNA solution was directly dissolved into ddh 2 O at a ratio of 1:50. PCR reactions were performed in a 20 µl volume containing 1 x PCR buffer [10 mm Tris-HCl (ph 8.3), 50 mm KCl, 1.5 mm MgCl 2, 0.01% gelatine], 100 µm dntps (datp, dctp, dgtp, dttp), 0.5 µm primers (F and R), 1:10 DNA, and 1 unit of Taq DNA polymerase (IRRI Taq). The PCR program used in this study was 5 minutes at a temperature of 94 C for early denaturation, then performed 35 cycles consisting of 60 seconds at a 94 C for denaturation, 60 seconds at 55 C for primer attachment, and 2 minutes at 72 C for primer extension. The final primer extension was done at 72 C for 7 minutes. The PCR results were separated using a 5% polyacrylamide gel, and staining was done using the silver staining method. Primers that gave polymorphism between the two parents (one Indonesian rice parent and one rice parent from IRRI), were selected and grouped based on the chromosome. Each of the polymorphic primers was viewed its position in the microsatellite map within the rice chromosomes. Polymorphic primers that had cm distances were selected for the background selection. The main primers (Table 1) which are polymorphic were used in the foreground and recombinant selection. RESULTS AND DISCUSSION Research on polymorphisms among rice parents is very important to determine primers that can be used for selection of the progeny of each crossing. The selection of microsatellite markers as markers for this type of selection is due to co-dominant character of the markers, so that it can detect allelic diversity at a high level (able to detect the origin of alleles from each locus), easy and economical to apply because it uses the PCR process, and repeatability is high enough (Temnykh et al. 2000). Microsatellite primers which were used for the foreground and recombinant selections were used in the initial selection to select the progeny from crosses that can be used for the background selection. Only the polymorphic markers between the two parents that could be used for further selection events. Monomorphic markers between two parents could not be used for selection activities because they could not distinguish the origin of allele donations in one locus. Dodokan, Situ Bagendit, and Batur are Indonesian improved upland rice varieties which are commonly grown by farmers in the dry land. Dodokan and Situ Bagendit can be grown in both dry land and paddy fields, while Batur is better suited for the dry land. All the three varieties were used as recipient because based on results of a preliminary study, they were sensitive to P deficiency. These varieties have the advantages as mentioned in the description of Indonesian improved rice varieties (Pusat Penelitian dan Pengembangan Tanaman Pangan 1993; Badan Penelitian dan Pengembangan Pertanian 2002). Kasalath is a landrace (local rice) originated from India. This rice is an upland rice which is highly tolerant to P deficiency. The use of Kasalath as a parent for crossing with Nipponbare (sensitive to P deficiency) has increased the P uptake by 28-55% higher than with Nipponbare. The Pup1 gene possessed by Kasalath is predicted function better under an upland condition (high altitude and dry), so the use of upland rice varieties as parental recipient will give good results. Apparently, Kasalath also released organic acids (citric acid) when grown in the Yoshida solution containing Fe-P as a source of P. This has caused the P becomes available for the plants. Besides increasing the binding of P before entering into the plant, Kasalath was also efficient in the use of P. This is evidenced by the introduction of OsPTF1 (Oryza sativa L. phosphate transcription factor 1) gene. This gene was cloned from Kasalath. Introduction of this gene into a rice variety sensitive to P deficiency (Nipponbare) using Agrobacterium tumefaciens increased the number of tillers, root and canopy dry weight by 30% in the nutrient solution and 20% in soil media (Yi et al. 2005). This means that Kasalath has two mechanisms simultaneously in facing P deficient conditions, namely the external and internal mechanisms. NIL-C443 or NIL-Pup1 was derived by crosses of Nipponbare with Kasalath (BC5). K is a derivative of crosses between IR36 and Kasalath. These three rice varieties contain the same gene from Kasalath, namely Pup1. The survey results of parents that were done can be seen in Figure 1 and Table 2. SSR3 is a microsatellite primer developed by IRRI researchers; its position tightly linked with the Pup1 gene and located between RM277 and RM519. Some markers that can give information about the Pup1 gene itself are currently being made by IRRI researchers. In the future, the expected primers can be used as markers for foreground selection (Dr. Joong Hyoun Chin 2007, pers. comm.). RM277 and RM519 represent two microsatellite markers known to be linked with the Pup1 gene with a distance between the two markers is approximately 5.1 cm. These markers are

4 4 Joko Prasetiyono et al. 100 bp DNA ladder a b c d e f RM28102 RM1261 SSR3 Amplification not found RM277 RM260 RM511 RM519 Figure 1. Amplification results of DNA from six rice parentages using main primers for foreground and recombinant selections that were separated by 5% polyacrylamide gel; a = Dodokan; b = Situ Bagendit; c = Batur; d = Kasalath; e = NIL-C443; f = K (primer RM260 did not yield bands). Arrows indicated the presence of two alleles in the K genotipe. Table 2. Scoring of amplification results from DNA of six rice parentages using the major primers. Primer Crosses 1 vs 4 1 vs 5 1 vs 6 2 vs 4 2 vs 5 2 vs 6 3 vs 4 3 vs 5 3 vs 6 RM277 P** P** P** M M M P** P** M RM511 P P P M M P P P M RM28102 P P P P** P** M M M P** SSR3 P * P * P P * P * P * M M P * RM1261 P P P M M P** P* P* M RM519 P** P** P** P** P** M P** P** M 1 = Dodokan; 2 = Situ Bagendit; 3 = Batur; 4 = Kasalath; 5 = NIL-C433; 6 = K P = polymorphic; M = monomorphic; *Selected primer for foreground selection; **Selected primer for recombinant selection. potential markers that can be used for recombinant selection. However, not all parental combinations gave polymorphic results, such as RM277 that produced monomorphic results in parents of Situ Bagendit vs Kasalath, NIL-C443 or K , and Batur vs K Similarly, RM519 produced monomorphic results in parents of Situ Bagendit vs K and Batur vs K To overcome this problem, therefore, it needs to be used a marker which is located in between two pincer markers to provide greater polymorphic opportunities. This marker can be used for selection, because a segment of 5 cm long still can be monitored using some markers in between the two pincer markers (Collard et al. 2005). Based on Figure 1, K with RM277, RM1261 and RM519 markers produced two bands; one of the bands was monomorphic with Batur (see arrows in Figure 1). Although this primer is polymorphic, it cannot be used for selection, because one of the bands was the same as that of the Batur parent. This condition will be confusing in the selection of progeny from crossing these two parents. Two alleles to be generated can be derived from both parental alleles or from seeds mixed with K seeds. The markers used for the foreground and recombinant selections are extremely necessary to help maintain segment of the donor DNA carrying the Pup 1 gene in the chromosomes of progenies from the crosses. In the

5 Identification of polymorphic markers for breeding of rice... 5 selection of F1 generation (F1, BC1F1, BC2F1, BC3F1, etc.), only individuals that give heterozygote bands (two bands) for the entire foreground and recombinant primers that must be selected, whereas when F2 generation (F2, BC1F2, BC2F2, BC3F2, etc.) have been used, the individuals that produce homozygote bands like the donor parents must be selected. Segments of this donor DNA must be maintained so as not to be lost during the recombination process when the crossing occurred. Therefore, the use of molecular markers is very helpful in plant breeding. Selection of microsatellite markers that will be used in the background selection was done by the PCR amplification using microsatellite primers spread throughout the rice chromosomes. The total numbers of primers that were used in this analysis were 489. Amplification of some of the markers can be seen in Figure 2 and results of amplification of all primers are given in Table 3. Comparison of the number of polymorphic primers in each cross and in every chromosome is presented in Figures 3 and 4. Data in Table 3 and Figure 3 indicate that the number of polymorphic primers in each cross was over 200 primers. This means that each chromosome can use about 20 primers. The high number of polymorphic primers was always found in crosses with NIL-C443 (Dodokan, Situ Bagendit or Batur vs NIL-C443). This is due to the striking difference in genome arrangement between NIL-C443 to the three improved varieties. NIL-C443 is a Nipponbare (japonica rice) which crossed with Kasalath (indica rice) by the backcross method (BC5), so most of the NIL-C443 genomes are japonica rice. The existence of genetic differences between indica and japonica rice has also been proved by Chakravarthi1 and Naravaneni (2006), even there are also differences in genome structure of their chloroplasts (Garris et al. 2005), although there are several loci that have similar genetic composition. Average number of the polymorphic microsatellite primers in each chromosome varied. The highest average number was found in chromosome 1, so that the number of options that microsatellite primers used was more than those in other chromosomes. Chromosome 12 had the least number of polymorphic primer (<10), thus requiring the addition of new primers that even more. This is particularly because the chance to get polymorphism among the parental species of indica rice is also low. Primers to be used should be closed, at least 10 cm distance between two primers to reduce the effect of linkage drag that usually arises during selection of the result from crossing. Selection of polymorphic primers as the background of selection is very important, since the condition of the chromosome progeny from crosses must be returned maximally as close as possible to the genome of the female parent, namely the Indonesian improved rice varieties. The d b f a c e Figure 2. Results of DNA amplification of six rice parentages using microsatellite primers for backgrond selection as separated by 5% polyacrylamide gel (PCR results could only be loaded two times with three well differences at second loading); a = Dodokan; b = Situ Bagendit; c = Batur; d = Kasalath; e = NIL-C443; f = K , 1 = RM241, 2 = RM17, 3 = RM22, 4 = RM25, 5 = RM49, 6 = RM60, 7 = RM70, 8 = RM124, 9 = RM162, 10 = RM169, 11 = RM172, 12 = RM223, 13 = RM178, 14 = RM210, 15 = RM395, 16 = RM204, 17 = RM30, 18 = RM194, 19 = RM55, 20 = RM256, 21 = RM175, 22 = RM334, 23 = RM245, 24 = RM169, 25 = RM13, 26 = RM44, 27 = RM291, 28 = RM52, 29 = RM146, 30 = RM32, 31 = RM314, 32 = RM170.

6 6 Joko Prasetiyono et al. Table 3. Results of polymorphisms on nine cross combinations of rice using 489 microsatellite primers for background selections. Number of polymorphic and monomorphic bands from each cross Number of primers used Chromosome 1 vs 4 1 vs 5 1 vs 6 2 vs 4 2 vs 5 2 vs 6 3 vs 4 3 vs 5 3 vs 6 for each chromosome Poly Mono Poly Mono Poly Mono Poly Mono Poly Mono Poly Mono Poly Mono Poly Mono Poly Mono Total Average = Dodokan; 2 = Situ Bagendit; 3 = Batur; 4 = Kasalath; 5 = NIL-C443; 6 = K Poly = polymorphic band; Mono = monomorphic band.

7 Identification of polymorphic markers for breeding of rice... 7 more polymorphic primers generated, the greater the chances of return to the parental genomes of the recipient. However, an increasing number of primers analyzed did not warrant an increasing number of polymorphism primers produced. Based on the calculation of the percentage of polymorphism (Figure 5), the lowest was achieved by chromosome 12 (35%) and the highest was for chromosome 10 (70%). So the opportunity to get polymorphic primers is larger in chromosome 10 than that in chromosome 12. Based on the tabulated results, more additional microsatellite primers are required in some chromosomes, such as chromosomes 4, 5, and 12. Number of polymorphic primers Number of polymorphic primers Percentage of polymorfphic primers Figure 3. Figure 4. Figure 5. 1 vs 4 1 vs 5 1 vs 6 2 vs 4 2 vs 5 2 vs 6 3 vs 4 3 vs 5 3 vs 6 Proportion of the number of polymorphic primers in each combination from nine crosses of rice; 1 = Dodokan, 2 = Situ Bagendit, 3 = Batur, 4 = Kasalath, 5 = NIL-C443, 6 = K Chromosome Average numbers of polymorphic primers in each chromosome from nine crosses of rice Chromosome Average percentages of polymorphic primers in each chromosome from nine crosses of rice. CONCLUSION Microsatellite markers that can be used for foreground and recombinant selections were six primers, i.e. RM277, RM511, RM28102, SSR3, RM1261, and RM519, with three different primers for each crossing, namely Dodokan vs Kasalath (RM277, SSR3, RM519 ), Dodokan vs NIL-C443 (RM277, SSR3, RM519), Dodokan vs K (RM277, SSR3, RM519), Situ Bagendit vs Kasalath (RM28102, SSR3, RM519), Situ Bagendit vs NIL-C443 (RM28102, SSR3, RM519), Situ Bagendit vs K (RM511, SSR3, RM519), Batur vs Kasalath (RM277, RM1261, RM519), Batur vs NIL-C443 (RM277, RM1261, RM519), and Batur vs K (RM28102, SSR3). Most of the polymorphic microsatellite markers that can be used for background selection were found in chromosomes 1 and 2, whereas the least number was found in chromosome 12. Increasing the number of primers is needed, especially for identification of the level of polymorphisms in chromosomes 4, 5, and 12. ACKNOWLEDGEMENTS This activity is a part of a research project entitled Revitalizing Marginal Lands: Discovery of Genes for Tolerance of Saline and Phosphorus Deficient Soils to Enhance and Sustain Productivity. A big thank you delivered to Dr. Abdelbagi M. Ismail (IRRI senior scientist and the PI project), who provided Kasalath, NIL-C443, and K as sources of gene resistance to P deficiency. The research project was funded by the Generation Challenge Program, Subprogram 2. REFERENCES Badan Penelitian dan Pengembangan Pertanian Deskripsi Varietas Unggul Baru Padi. Badan Penelitian dan Pengembangan Pertanian, Jakarta. 56 hlm. BPS (Badan Pusat Statistik) Statistik Indonesia. Badan Pusat Statistik, Jakarta. Chakravarthil, B.K. and R. Naravaneni SSR marker based DNA fingerprinting and diversity study in rice (Oryza sativa. L). African J. Biotechnol. 5(9):

8 8 Joko Prasetiyono et al. Collard, B.C.Y., M.Z. Jahufer, J.B. Brouwer, and E.C.K. Pang An introduction to markers, quantitative trait loci (QTL) mapping and marker-assisted selection for crop improvement: The basic concepts. Euphytica 142: Dellaporta, S.L., J. Wood, and J.B. Hicks A plant DNA minipreparation: Version II. Plant. Mol. Biol. Rep. 1(4): Friscth, M., M. Bohn, and A.E. Melchinger Minimum sample size and optimal positioning of flanking markers in markerassisted backcrossing for transfer of a target gene. Crop Sci. 39: Garris, A.J., T.H. Tai, J. Coburn, S. Kresovich, and S. McCouch Genetic structure and diversity in Oryza sativa L. Genetics 169: Pusat Penelitian dan Pengembangan Tanaman Pangan Deskripsi Varietas Unggul Padi Pusat Penelitian dan Pengembangan Tanaman Pangan, Bogor. 123 hlm. Radin, J.W. and M.P. Eidenbock Hydraulic conductance as a factor limiting leaf expansion of phosphorus-deficient cotton plants. Plant Physiol. 75: Ribaut, J.M. and D. Hoisington Marker-assisted selection: New tools and strategies. Mol. Breed. 3: Temnykh, S., W.D. Park, N. Ayres, S. Cartinhour, N. Hanck, L. Lipovich, Y.G. Cho, T. Ishii, and S.R. McCouch Mapping and genome organization of microsatellite sequences in rice (Oryza sativa L.). Theor. Appl. Genet. 100: Wissuwa, M., M. Yano, and N. Ae Mapping of QTLs for phosphorus-deficiency tolerance in rice (Oryza sativa L.). Theor. Appl. Genet. 97: Wissuwa, M. and N. Ae Further characterization of two QTLs that increase phosphorus uptake of rice (Oryza sativa L.) under phosphorus deficiency. Plant Soil 237: Wissuwa, M., J. Wegner, N. Ae, and M. Yano Substitution mapping of Pup1: A major QTL increasing phosphorus uptake of rice from a phosphorus-deficient soil. Theor. Appl. Genet. 105: Yi, K., Z. Wu, J. Zhou, L. Du, L. Guo, Y. Wu, and P. Wu OsPTF1, a novel transcription factor involved in tolerance to phosphate-starvation in rice (Oryza sativa L.). Plant Physiol. 138: