当前位置:首页 >> 农林牧渔 >>

Development of pre-isogenic lines for rice blast-resistance by marker-aided selection from a recomb


Theor Appl Genet (1996) 93:560-567

9 Springer-Verlag 1996

T. I n u k a i 9 R. S. Z e i g l e r 9 S. S a r k a r u n g 9 M . B r o n s o n L. V. D u n g 9 T. K i n o s h

i t a 9 R. J. N e l s o n

Development of pre-isogenic lines for rice blast-resistance by marker-aided selection from a recombinant inbred population

Received: 15 November 1995 / Accepted: 22 March 1996

To increase the available set of near-isogenic lines (NILs) for blast-resistance in rice, we have developed a general method for establishing NILs from populations of fixed recombinants that have been used for gene mapping. We demonstrated the application of this method by the selection of lines carrying genes from the rice cultivar Moroberekan. Moroberekan is a West African japonica cultivar that is considered to have durable resistance to rice blast. Multiple genes from Moroberekan conferring complete and partial resistance to blast have previously been mapped using a recombinant inbred (RI) population derived from a cross between Moroberekan and the highly and broadly susceptible indica cultivar CO39. To analyze individual blast-resistance genes, it is desirable to transfer them individually into a susceptible genetic background. This RI population, and the associated data sets on blast reaction and restriction fragment length polymorphism (RFLP) genotypes, were used for selection of lines likely to carry individual blast-resistance genes and a minimum number of chromosomal segments from Moroberekan. Because skewed segregation in the RI population favored CO39 (indica) alleles, resistant lines carrying 8.7-17.5% of Moroberekan alleles (the proportion expected after two or three backcrosses) could be selected. We chose three RI lines carrying different complete resistance genes to blast and two RI lines carrying partial resistance genes to blast as potential parents for the development of NILs. These lines were subjected to genetic analysis, which allowed clarification of some issues that could not be resolved during the initial gene-mapping study.
Abstraet

Introduction
Near-isogenic lines (NILs) carrying single genes for blastresistance are useful for genetic analysis (Kiyosawa 1967; Yokoo and Kiyosawa 1970; Kiyosawa 1972; Mackill and Bonman 1992; Inukai et al. 1994a), for gene tagging (Yu et al. 1991; Miyamoto et al. 1993; Satoh et al. 1993), and for characterizing pathogen isolates (Inukai et al. 1994b; Zeigler et al. 1995). However, two problems have been encountered using the conventional approach to the development of NILs. First, while backcrossing and selection is effective for the transfer of single genes, it may not be efficient when multiple genes are present in the donor parent. If multiple genes for blast-resistance are not differentiated by the isolates used for selection, some of the loci may be "lost". In developing a set of NILs, for instance, Mackill and Bonman (1992) used four blast-resistance donors known, or suspected, to carry multiple genes for blastresistance. To "capture" as many of the blast-resistance genes as possible in the NILs, five pathogen isolates were used to screen lines at each of several generations of backcrossing to the susceptible cultivar CO39 and at each of the subsequent generations of selfing. In spite of this, allelism tests revealed that only five blast-resistance loci were identified among the 22 NILs developed from the four blast-resistance donors (Inukai et al. 1994a). The donor cultivars are resistant to some isolates to which the NILs are susceptible, suggesting that not all of the blast-resistance genes in the donor cultivars were transferred to the NILs (T. Inukai, unpublished). A second limitation of the conventional approach to NIL production is the difficulty of evaluating the extent of return to the recurrent parental genotype. Although the amount of genetic material from the donor is expected to be reduced by half with each generation of backcrossing to the recurrent parent, the observed level of donor DNA may be substantially higher than expected in some cases (Young and Tanksley 1989). For instance, based on RFLP data, the NILs for blast-resistance carried surprisingly high numbers of loci from the donor parents (Yu 1991). Ana-

Disease resistance - Rice blast 9 RFLPs Recombinant inbred lines 9 Pre-isogenic lines
Key words

Communicated by G. Wenzel T. Inukai (~). R. S. Zeigler 9S. Sarkarung - M. Bronson L. V. Dung 9T. Kinoshita. R. J. Nelson Faculty of Agriculture, Hokkaido University, Sapporo 060, Japan

561 lyzed with 4 1 - 9 2 probes, the lines were found to carry 10.9-27.5% of donor loci (Yu 1991). We present here an approach for developing NILs from populations of fixed segregants that have been used for molecular m a p p i n g of blast-resistance genes. All the blast-resistance genes present in a donor cultivar can be detected because every c h r o m o s o m e segment of the d o n o r parent is present in at least one m e m b e r of a p e r m a n e n t population of fixed segregants (either r e c o m b i n a n t inbred lines or doubled-haploid lines), and the population can be analyzed with multiple pathogen isolates. Molecular marker data obtained during the g e n e - m a p p i n g process can be used to select lines carrying blast-resistance genes and few other loci from the donor parent. These two conditions make it possible to select lines carrying single blast-resistance genes and few other loci from the donor parent from a population of fixed segregants. We term such lines "pre-isogenic lines (PILs)" in the sense that they can be used as intermediates in the production of near-isogenic lines (NILs). M o r o b e r e k a n is an African upland japonica cultivar considered to have durable blast-resistance. To better u n d e r s t a n d the genetic basis of this exceptionally useful resistance, Wang et al. (1994) used R F L P markers to map major and m i n o r genes in a cross b e t w e e n Moroberekan and the susceptible indica cultivar CO39. They developed a population of 281 r e c o m b i n a n t inbred lines (RILs), and obtained R F L P data for each of the RILs at 127 loci selected to provide u n i f o r m g e n o m e coverage. Wang et al. (1994) identified and mapped two major genes for blastresistance, Pi-5(t) and Pi-7(t), and ten quantitative trait loci (QTLs) c o n d i t i o n i n g partial resistance to the blast pathogen. They also found evidence for an additional blast-resistance gene in M o r o b e r e k a n ; some of the RILs carrying neither Pi-5(t) nor Pi-7(t) (susceptible to isolate PO6-6) showed resistance to Philippine isolate Ca65 (G. Wang, IRRI, unpublished). In the present study, we selected and further characterized lines from the M o r o b e r e k a n / C O 3 9 population, as a step towards the d e v e l o p m e n t of NILs carrying blast-resistance genes from Moroberekan. IK81-3, IK81-25, V86010 and Ca65 were used in this study. Except for Ca65, these isolates were previously used to differentiate CO39 NILs (Mackilt and Bonman 1992). Isolate Ca65 was recently collected from the IRRI upland screening site at Cavinti, Laguna, The Philippines. Data sets The following data sets were available for a population of 281 Fv RILs derived from a cross between Moroberekan and CO39 (Wang et al. 1994), and were used for the present study: (1) genotypes of each RIL for 127 RFLP loci distributed on the map at intervals of about 20 cM; (2) disease reaction score (0-5) of each RIL to blast isolates PO6-6 and Ca65 (data obtained using monocyclic inoculation tests); (3) diseased leaf area (DLA), susceptible lesion number and lesion size of each RIL (data obtained using polycyclic inoculation tests using blast isolate PO6-6); (4) DLA of each RIL under natural infection at the IRRI upland screening site at Cavinti, Laguna, The Philippines (data obtained from field trials conducted in 1992 and 1994, each with two replications). Inoculation methods For pathogenicity tests, seeds of PILs and CO39 NILs were sown in plastic trays (11? 1 cm). Ten seeds were planted in each of eight rows in each tray with three replications. For F 2 analysis, plastic trays (37?215 cm) were divided equally into five rows, and 30 F2 seeds were uniformly sown in each of four rows in each tray. One row in each tray was divided into two or three more rows, and parental lines and a susceptible check cultivar were sown in each row, with ten seeds per row. For F? analysis, plastic trays were divided equally into ten rows, and F3 line seeds were sown in each of nine rows, with 20 seeds per F 3 line per row. One row in each tray was divided into two more rows, and parental lines were sown in each row, with ten seeds per row. In all experiments, nitrogen fertilizer was applied at 36 g/m 2 as ammonium sulphate, and seedlings were grown in a green house. Seedlings were inoculated 21 days after sowing (at the fifth-to-sixth leaf stage) by the spraying method previously described (Inukai et al. 1994a). The disease reactions were scored about 7 days after inoculation. Selection of PILs To allow selection of lines carrying genes derived from Moroberekan that condition complete resistance to isolate PO6-6 or isolate Ca65, RILs showing complete resistance to isolate PO6-6 or isolate Ca65 (those lines with disease scores of 0-2 to one or both of these isolates) were identified. Among each group with a particular reaction pattern to the two isolates, those lines carrying the fewest RFLP alleles from Moroberekan were selected. The reaction patterns of the selected lines were tested with the five isolates described above, and the lines showing different reaction patterns were selected as PILs and used for the following genetic analysis. To allow selection of PILs carrying Moroberekan genes conditioning partial resistance to blast, RILs not carrying any of the resistance genes to isolates PO6-6 and Ca65, but showing partial resistance to blast in polycyclic tests and field experiments, were identified. From among the lines with less than half the number of susceptible lesions relative to CO39 in polycyclic tests and showing low levels of disease in field experiments, those carrying the fewest Moroberekan alleles for the 127 RFLP loci were selected as PILs. RFLP marker data for loci linked to putative QTLs conditioning partial resistance to blast were used to estimate which QTLs were present in the selected lines. Genetic analysis CO39 was used as a susceptible parent or as a susceptible check cultivar for the F2 and F3 analyses. F2 and F 3 segregation ratios were an-

Materials and methods
Plant materials and pathogen isolates Moroberekan is an African upland.japonica cultivar with durable blast-resistance. CO39 is an indica cultivar that is highly susceptible to most isolates of the rice blast pathogen, Pyricularia grisea. Recombinant inbred lines (RILs), previously developed and designated RILl0, RIL23, RIL29, RILl25, RIL249 and RIL276, were selected from a population of 281 F7 RILs of a cross between Moroberekan and CO39 (Wang et al. I994), based on available markers and phenotypic data. These were considered as putative "pre-isogenic lines (PILs)" for further genetic analysis. The CO39 near-isogenic lines (NILs) C 101LAC (carrying Pi-1), C101A51 (carrying Pi-zS), C 104PKT (carrying Pi-3), C 101PKT (calTyiug Pi-ta) and C 105TTP4L23 (carrying Pi-ta and an additional, unidentified gene) (Mackill and Bonman 1992; Inukai et al. 1994a; Kinoshita et al. 1994) were used as tester lines for genetic analysis. Isolates of the rice blast pathogen P. grisea were obtained from the collection maintained at the Entomology and Plant Pathology Division at the International Rice Research Institute (IRRI) in the Philippines. Isolates PO6-6,

562 alyzed by the chi-square test, and a recombination value was calculated by the maximum-likelihood method. Distances between markers are presented in centimorgans derived using the Kosambi function. RFLP analysis DNA was extracted from the leaves of CO39, RIL29, RIL249 and the F 2 plants using the procedure described by Dellaporta et al. (1984). DNA extracts were digested with the restriction enzymes DraI, EcoRI, EcoRV, HindIII and ScaI. The digested DNAs were subjected to electrophoresis on 1% agarose gels and transferred to BIODYNE B membranes (Pall Corp.) according to the manufacturer's instructions. The DNA clones that were reported to be linked to Pi-5(t) or Pi-7(t) (Tanksley et al. 1992; Wang et al. 1994) were labeled by ECL direct nucleic-acid labelling and detection systems (Amersham Corp.) and used as probes for F 2 segregation analysis. Probe labelling, hybridization and signal detection conditions were done according to the instructions accompanying the ECL direct nucleic-acid labelling and detection systems (Amersham Corp.). o f i n t r o g r e s s e d d o n o r genetic material e x p e c t e d after two or three backcrosses. To d e t e r m i n e the n u m b e r o f genes c o n d i t i o n i n g resistance to isolate PO6-6 in each of these lines, R I L 2 9 and R I L 2 4 9 were each c r o s s e d to CO39, and F 2 p o p u l a t i o n s d e r i v e d from each cross were i n o c u l a t e d with PO6-6. F o r both crosses, the results fit the e x p e c t e d 3:1 ratio for resistant:susceptible progeny, c o n f i r m i n g that each line carried a single gene. To d e t e r m i n e w h e t h e r the two lines carried the same or different genes, a cross was m a d e b e t w e e n RIL29 and RIL249, and again the F 2 plants were inoculated with isolate PO6-6. The ratio o f resistant to susceptible plants fit the 15:1 ratio e x p e c t e d for two i n d e p e n d e n t loci (Table 3). Thus, it was c o n f i r m e d that R I L 2 9 and R I L 2 4 9 each carried single resistance genes to isolate PO6-6, and that those b l a s t - r e s i s t a n c e genes were i n d e p e n dent. W h e n the reaction profiles of the R I L s were c o m p a r e d with those of CO39 NILs, the reaction pattern of RIL29 was f o u n d to be similar to that o f C 1 0 1 L A C c a r r y i n g Pi-1, while the reaction pattern o f R I L 2 4 9 was similar to that o f C 1 0 4 P K T c a r r y i n g Pi-3 (Table 2). To test the hypothesis that the gene in R I L 2 9 was allelic to Pi-1, and that the gene in R I L 2 4 9 was allelic to Pi-3, crosses were m a d e b e t w e e n the two R I L s and the selected C O 3 9 NILs. The cross b e t w e e n R I L 2 9 and C 1 0 1 L A C y i e l d e d no p r o g e n y showing s u s c e p t i b i l i t y to isolate PO6-6, indicating that the b l a s t - r e s i s t a n c e gene in R I L 2 9 was either allelic with or else closely l i n k e d to Pi-1. In the F 2 p o p u l a t i o n d e r i v e d from the cross b e t w e e n R I L 2 4 9 and C 1 0 1 L A C , a 15:1 ratio of resistant and susceptible plants was seen for isolate PO6-6, c o n f i r m i n g that the gene in R I L 2 4 9 is not allelic with or linked to Pi-1 (Table 3). To test a l l e l i s m b e t w e e n the gene in R I L 2 4 9 and Pi-3 in C104PKT, isolate P O 3 - 8 2 - 5 1 was used, b e c a u s e this isolate is i n c o m p a t i b l e to both C 104PKT and RIL249, but c o m p a t i b l e to CO39. W h e n F 2 plants from the cross b e t w e e n R I L 2 4 9 and CO39 were i n o c u l a t e d with this isolate, a 3:1 s e g r e g a t i o n ratio was seen for blast-resistance, indicating that a single gene for resistance to isolate PO382-51 was present in RIL249. A m o n g the 466 F 2 p r o g e n y of the cross b e t w e e n C 1 0 4 P K T and RIL249, no p r o g e n y susceptible to isolate P O 3 - 8 2 - 5 1 were found, indicating that the gene in R I L 2 4 9 c o n d i t i o n i n g resistance to this isolate is allelic, or else c l o s e l y linked, to Pi-3 (Table 3).

Results
PILs with genes c o n d i t i o n i n g c o m p l e t e resistance to P. grisea isolate PO6-6 As r e p o r t e d b y W a n g et al. (1994), M o r o b e r e k a n has at least two genes for c o m p l e t e resistance to P. grisea isolate PO6-6. These loci, d e s i g n a t e d Pi-5(t) and Pi-7(t), were located on c h r o m o s o m e s 4 and 11, r e s p e c t i v e l y (Wang et al. 1994). As a first step t o w a r d s selecting PILs for these blastresistance genes, the RILs that were resistant to isolate PO6-6 and c a r r y i n g less than 20% of M o r o b e r e k a n alleles at R F L P loci were selected from a p o p u l a t i o n of 281 F 7 R I L s of a cross b e t w e e n M o r o b e r e k a n and C O 3 9 (Table 1). Three R I L s (RIL29, R I L l 2 5 and RIL249) satisfying the a b o v e criteria were selected and the reaction patterns o f those RILs to the five test isolates were determined. W h i l e RIL29 and R I L l 2 5 s h o w e d the same reaction patterns to the test isolates, R I L 2 4 9 s h o w e d a different reaction pattern from the other two RILs. The reaction patterns of these RILs were similar to those o f the CO39 N I L s c a r r y i n g Pi-1 and Pi-3, r e s p e c t i v e l y (Table 2). The m o r p h o l o g y of RIL29 was m o r e similar to CO39 than was that o f R I L 125, so the former was selected for further analysis, along with RIL249. RIL29 and R I L 2 4 9 had M o r o b e r e k a n alleles at 17.5 and 10.3% o f the R F L P loci tested, r e s p e c t i v e l y (Table 1). These p r o p o r t i o n s c o r r e s p o n d e d to the amounts

Table 1 Pre-isogenic lines

(PILs) for complete or partial resistance genes to blast, selected from recombinant inbred lines (RILs) of a cross between Moroberekan and CO39

Line no.

Generation

R gene or QTL

No. of RFLP markers from Moroberekan 22/127 11/127 13/127 15.5/127 12/127 12/127

%

RIL29 RILl25 RIL249 RILl0 RIL23 RIL276

F7

F7 Fv F7
F7

F7

R gene to PO6-6 R gene to PO6-6 R gene to PO6-6 R gene to Ca65 QTL linked to RZ398, RG64 (chr. 6) and RG612 (chr. 1) QTL linked to RG64 (chr. 6) and CDO 920 (chr. 1)

17.5 8.7 10.3 12.3 9.5 9.5

563 Table 2 Reaction pattern of pre-isogenic lines (PILs), selected from recombinant inbred lines of a cross between CO39 and Moroberekan, and CO39 near-isogenic lines (NILs) to P. grisea isolates Line
Pi gene a

Reaction to isolates b PO6-6 IK81-3 R R
S S

IK81-25 S S
S S

V86010 R R
I R

Ca65 S S R

RIL29 RIL 125 RIL249 RILl0 CO39 CI01LAC C101A51 C 104PKT C101PKT C 105TTP-4L23

Pi- 7(t) Pi- 7(t) Pi-5(t)? Pi-12(t)

R R
R S

Pi-1 Pi-z 5 Pi-3 Pi-ta Pi-ta, Pi-? c

S
R R R S S

S
R R S R R

S
S R S R R

S
R R S S R

S
-

Wang et al. (1994); Inukai et al. (1994a); Kinoshita et al. (1994) b R=resistant; I=intermediate; S=susceptible c An additional, unidentified gene in C 105TTP-4L23 (Inukai et al. 1994a)

Table 3 Reaction of F 2 populations of crosses between preisogenic lines (PILs) and CO39 near-isogenic lines (NILs) to P. grisea isolates

Cross

Test isolate

No. of F 2 plants observed for each class
R S

Expected ratio (R:S) 3:1 3:1 15 : 1 1:0 15 : 1 3:1 1:0 3: l

Probability

CO39/RIL29 CO39/RIL249 RIL29/RIL249 RIL29/C101LAC RIL249/CO101LAC CO39/RIL249 C 104PKT/RIL249 CO39/RIL10

PO6-6 PO6-6 PO6-6 PO6-6 PO6-6 PO3-82-51 PO3-82-51 V86010

88 61 227 234 225 84 466 66

32 19 11 0 13 32 0 33

0.50-0.75 0.75-0.90 0.25-0.50 0.50-0.75 0.25-0.50 0.05-0.10

Table 4 Observed frequencies of genotypes on R genes to isolate PO6-6 and RFLP markers in crosses CO39/RIL29 and CO39/RIL249 Cross Gene pair Genotype a AABB AABb AAbb CO39 (aabb)/ RIL29(AABB) CO39 (aabb)/ RIL249 (AABB) R gene-RZ537 R gene-RG788 6 (3.4) b 2 (2.7) c 5 (6.9) 10 (12.9) 1 (3.4) 3 (9.3) AaBB 5 (6.9) 5 (4.7) AaBb 20 (13.8) 26 (22.2) Aabb 6 (6.9) 12 (16.1) aaBB 1 (3.4) 3 (2.5) aaBb 7 (6.9) 11 (11.9) aabb 4 (3.4) 9 (8.6) 55 91 0.05-0.10 0.10-0.25 Total Probability

'~A and a represent the genotypes of each complete resistance gene. B and b represent the genotypes of each RFLP marker b Expected value estimated by theoretical ratio 1:2:1:2:4:2:1:2:1 Expected value estimated by observed value. Because the segregation at RG788 locus was significantly skewed (BB:Bb:bb=10:47:34, 22=12.76 " 4), a test for independence was done

W a n g et al. (1994) e s t i m a t e d that P i - 7 ( t ) was f l a n k e d b y the R F L P m a r k e r s R G 1 6 a n d R G 1 0 3 A o n c h r o m o s o m e 11. B e c a u s e the r e c o m b i n a n t i n b r e d p o p u l a t i o n u s e d in this e s t i m a t i o n s h o w e d s k e w e d s e g r e g a t i o n ( W a n g et al. 1994), the c h r o m o s o m a l l o c a t i o n o f P i - 7 ( t ) c o u l d n o t b e determ i n e d precisely. Yu et al. (1991) r e p o r t e d that P i - 1 was l i n k e d to R Z 5 3 6 on c h r o m o s o m e 11, at a d i s t a n c e o f 14.0 cM. M e w et al. (1994) r e p o r t e d that P i - 1 was l i n k e d to R Z 5 3 6 a n d G 1 8 1 [the p r e v i o u s r e g i s t r a t i o n was N p b 181 ( I n o u e et al. 1994)] with d i s t a n c e s o f 7.9 a n d 3.5 cM, respectively. A l t h o u g h R I L 2 9 did n o t carry M o r o b e r e k a n al-

leles at the R G 1 6 , R G 1 0 3 A a n d R Z 7 9 7 loci, this l i n e h a d the M o r o b e r e k a n allele at R Z 5 3 7 , w h i c h is l o c a t e d n e a r R G 1 0 3 A (Fig. 1). To test the h y p o t h e s i s that the b l a s t - r e sistance g e n e in R I L 2 9 is l o c a t e d n e a r R Z 5 3 7 on c h r o m o s o m e 11, the s e g r e g a t i o n o f b l a s t - r e s i s t a n c e g e n e s a n d R Z 5 3 7 was a n a l y z e d for 55 F 3 lines d e r i v e d f r o m the cross b e t w e e n C O 3 9 a n d R I L 2 9 . T h e results s h o w e d that R Z 5 3 7 was l o o s e l y l i n k e d to the b l a s t - r e s i s t a n c e g e n e in R I L 2 9 with d i s t a n c e o f 27.0_+5.5 (SE) c M (Table 4). S i n c e the m a p d i s t a n c e b e t w e e n R Z 5 3 7 a n d G181 was e s t i m a t e d to be a b o u t 30 c M by p r e v i o u s data ( T a n k s l e y et al. 1992; M e w

564 Fig. 1 Graphical genotype of a part of chromosome 11 in RIL29. The black region indicates the segment derived from Moroberekan; white regions indicate segments derived from CO39; the shaded region has not yet been checked. Designations to the right indicate marker names; numbers to the left indicate the map distance s(cM). Order of markers and the map distances were according to Tanksley et al. (1992) and Mew et al. (1994) RZ536 11.4 G181 10.0 RG303 8.0 RGl109 10.1 RZ537 11.7 6.0 RG103A RZ797 RG16 Since the reaction pattern of RIL 10 to the five test isolates was clearly different from that of the existing CO39 NILs, or indeed any of the identified PILs carrying resistance genes to isolate PO6-6 (Table 2), this putative blast-resistance gene was temporarily designated Pi-12(t). RIL23 RIL276 CO39 Moroberekan Table 5 Partial blast-resistance of pre-isogenic lines (PILs) RIL23 and RIL276 Line no. Polycyclic test Lesion no. 47 ~ 41 I00 Lesion size 100c 45 100 DLA b

Field test a
DLA b

30 c 18 100 -

31 c 43 100 1

a In Caliraya experimental station, Laguna, Philippines b Diseased leaf area Relative ratio to CO39 (100)

et al. 1994) (Fig. 1), it appeared that the blast-resistance gene in RIL29 was located in the same region of chromosome 11 as the Pi-1 locus. This conclusion was consistent with the result of the allelism test with C 1 0 1 L A C . Pi-7(t) in RIL29 appeared to be allelic, or else closely linked, to

PILs with QTLs conditioning partial resistance to blast Of the original 281 Moroberekan/CO39 RILs, 138 were susceptible to both isolates PO6-6 and Ca65. A m o n g these, 43 RILs showed fewer than half the number of susceptible lesions than did CO39 in polycyclic inoculation tests (Wang 1992). The D L A of most of these RILs was also less than half that of CO39 in polycyclic tests and field experiments (Wang 1992; Bronson 1994). These lines were considered likely to carry quantitative trait loci (QTLs) conditioning partial resistance to blast. A m o n g these lines, RIL23 and R I L 2 7 6 carried the lowest number of Moroberekan alleles at the R F L P loci analyzed, and were thus selected as PILs (Table 5). While the lesions formed on R I L 2 7 6 were less than half the size of those formed on CO39 in polycyclic tests, the lesions formed on RIL23 were similar to those on CO39 (Table 5). Both RIL23 and RIL276 carried 9.5% Moroberekan alleles (Table 1). RIL23 carried Moroberekan alleles at R F L P loci linked to three different QTLs affecting lesion number and/or DLA: RG612 on c h r o m o s o m e 1, and RZ398 and RG64 on c h r o m o s o m e 6 (Fig. 2). R I L 2 7 6 carried Moroberekan alleles at R F L P loci linked to two QTLs affecting lesion number, D L A and/or lesion size: C D O 9 2 0 on c h r o m o s o m e 1, and RG64 on c h r o m o s o m e 6 (Fig. 2).

Pi-1.
If Pi-5(t) and Pi-7(t) were the only genes in Moroberekan conditioning resistance to isolate PO6-6, then RIL249 would be expected to carry Pi-5(t), because the blastresistance gene in RIL249 was independent of Pi-7(t) in RIL29 (Table 3). Since RIL249 had a Moroberekan allele at the RG788 locus, which was found to be linked to Pi-5(t) (Wang et al. 1994), linkage between the RG788 locus and the blast-resistance gene in RIL249 was analyzed. Ninety one F 3 lines derived from the cross between CO39 and RIL249 were inoculated and scored for reaction to isolate PO6-6. D N A was extracted from the lines, and membrane-bound digested D N A s were probed with RG788. Although the segregation at the RG788 locus in the F 2 population was significantly skewed to the CO39 allele, it was shown that the blast-resistance gene in R l L 2 4 9 was independent of, or else loosely linked to, the RG788 locus (Table 4). This result suggested that Pi-5(t) was actually not linked to the RG788 locus, or else that Moroberekan carried an additional resistance gene to isolate PO6-6.

PIL with a gene conditioning complete resistance to P. grisea isolate Ca65 Twelve of the one-hundred and fifty RILs that were susceptible to isolate PO6-6 were resistant to isolate Ca65, suggesting that Moroberekan carried at least one additional blast-resistance gene (G. Wang, IRRI, unpublished). A m o n g the RILs showing resistance to isolate Ca65, but not to isolate PO6-6, R I L l 0 carried the lowest number of Moroberekan alleles (Table 1), and was selected as a PIL.

Discussion
Near-isogenic lines are useful for studies aimed at characterizing blast-resistance genes, and for studies aimed at characterizing pathogen virulence. NILs for blast-resistance have been developed by conventional backcrossing and selection (Mackitl and B o n m a n 1992; Sasaki et al. 1994). We propose that N I L s could be more easily and efficiently produced by marker-aided selection of lines from

565
Fig. 2 Graphical genotype of chromosomes 1 and 6 in

RILs23 and 276. Black regions indicate segment derived from Moroberekan; white regions indicate segments derived from CO39. Designations to the right indicate marker names; stippled bars to the left represent supporting intervals around the chromosomal regions associated with partial resistance to blast (Wang et al. 1994)

Ii ~
0 cM

CHR 1
- - RG350 RG77 RG236 RG331

CHR 6

CHR I
-~ RG350 RG77 RG331

RGI09 RZ730 RG197 RG101 RG462

IF
I

CHR 6

-~CDO475 __'~RZ516

- - W A X Y

RG109 RZ730 RG197 RGIO1 RG462 CDO920 RZ744 RZ276 RG811

RIL23

El Lesion n u m b e r Diseased leaf area Lesion size

RIL276

populations of fixed segregants that have been used for RFLP mapping. In the present study we demonstrated that a set of PILs carrying individual blast-resistance genes could be selected from a population of recombinant inbreds by marker-aided selection. Three lines carrying distinct major genes were selected based on marker and phenotypic data, and subjected to further genetic analysis. Based on the data available for 127 RFLP loci, each of the selected PILs carried less than 20%, and in some cases nearly 10%, of the genome of the resistant parent, Moroberekan. The extent to which the genetic background of the susceptible parent genome was recovered corresponded to that expected for two or three backcrosses. Based on available RFLP data, the PILs selected in this study do not possess more genetic material from the donor parent than do other near-isogenic lines carrying blast-resistance genes (Yu 1991). It was possible to recover a set of resistant lines with relatively little of the Moroberekan genome because the recombinant inbred population utilized in this study showed skewed segregation, favoring alleles from the susceptible parent (Wang et al. 1994). The RI population represented a set of F 7 lines derived from a cross between an indica and a japonica cultivar. Although 1:1 segregation would be expected with an F 7 population, skewed segregation is often observed with indica/japonica crosses (Oka

1955; McCouch et al. 1988). Averaged across RFLP loci, the frequency of indica (CO39) alleles was 75% in this population; averaged across lines, the frequency of indica alleles was 80% (Wang et al. 1994). Thus, segregation distortion in indica/japonica crosses may be advantageous for the selection of PILs with an indica genetic background. If a recombinant inbred or doubled haploid population showing normal segregation was used, the probability of recovering PILs from the population would be lower, in comparison to a population showing segregation distortion. In this case, the use of a population generated from BC IF1 plants without selection would be an alternative way to reduce the proportion of alleles coming from the donor parent. Although segregation distortion may be advantageous for the selection of PILs, it may reduce the accuracy of gene mapping. The accuracy of gene mapping in the study of Wang et al. (1994), in which the Moroberekan/CO39 RI population was used to identify and locate genes for blastresistance, was affected by this. The results of the initial mapping work should be viewed as establishing an hypothesis about the number of genes effective for the isolates tested, and their approximate locations. For instance, Pi7(t) was estimated by Wang et al. (1994) to reside between RG16 and RG103A on chromosome 1 t. The selection of a line carrying only this gene effective against isolate PO6-

566 6 a l l o w e d more precise genetic analysis, i n v o l v i n g allelism tests and segregation analysis, which led to the conclusion that Pi-7(t) is allelic, or closely linked, to Pi-1. This gene is also present on c h r o m o s o m e 1 1, but is likely to reside at a m o r e distal position, m o r e closely linked to G181. B a s e d on the analysis o f the RI population, W a n g et al. (1994) identified two m a j o r genes effective for isolate PO6-6, and d e s i g n a t e d these as Pi-5(t) and Pi-7(t). In the present study, two genes effective for this isolate were also identified. It is clear that the gene in RIL29 c o r r e s p o n d s to Pi-7(t), and that this gene is a p p a r e n t l y allelic, or closely linked, to Pi-1 in the existing NILs. Further w o r k is, h o w ever, n e e d e d to identify the c h r o m o s o m a l location of the gene in RIL249, and to d e t e r m i n e whether or not this gene is the same as Pi-5(t). The gene in R I L 2 4 9 was a p p a r e n t l y unlinked to RG788, which was found to be linked to Pi5 ( 0 by W a n g et al. (1994). Lines c a r r y i n g the c h r o m o s o mal segment to which Pi-5(t) was m a p p e d are currently under analysis. The RI p o p u l a t i o n utilized in the present study was previously used to m a p both m a j o r genes and Q T L s in M o r oberekan. Twenty R F L P loci defining ten c h r o m o s o m a l segments o f M o r o b e r e k a n have been found to be a s s o c i a t e d with quantitative effects on blast resistance, and the individual contribution o f Q T L s to partial resistance to blast has been e s t i m a t e d (Wang et al. 1994). If a set o f near-isogenic lines carrying i n d i v i d u a l Q T L s can be obtained, it will be p o s s i b l e to s y s t e m a t i c a l l y evaluate the m a g n i t u d e of the effects of i n d i v i d u a l Q T L s relative to a wide range of p a t h o g e n isolates, and to evaluate the associations o f other traits with these c h r o m o s o m a l segments. Such a set o f N I L s w o u l d also be useful for investigating the m e c h a nism of partial resistance to blast. The PILs selected here should be useful as donor parents for d e v e l o p i n g N I L s carrying i n d i v i d u a l QTLs. A l t h o u g h the PILs selected in this study carry r e l a t i v e l y few M o r o b e r e k a n loci, the lines each contain several chrom o s o m e segments from the b l a s t - r e s i s t a n c e donor. These introgressed segments m a y also carry u n r e c o g n i z e d blastresistance genes. The PILs should be b a c k - c r o s s e d to CO39 to r e m o v e i n t r o g r e s s e d segments from M o r o b e r e k a n that are not a s s o c i a t e d with b l a s t - r e s i s t a n c e genes o f interest with the aid o f R F L P analysis. O n l y a few b a c k c r o s s e s will be needed, b e c a u s e the R F L P g e n o t y p e o f each PIL is known, and the c h r o m o s o m a l segments from the resistant parent can be efficiently r e m o v e d by R F L P - b a s e d selection (Tanksley et al. 1989). The purification of the PILs obtained here is now under way. We used the a p p r o a c h p r e s e n t e d in this p a p e r to a n a l y z e only b l a s t - r e s i s t a n c e genes. However, this a p p r o a c h should also be a p p l i c a b l e to the analysis of m a j o r or m i n o r genes a s s o c i a t e d with other traits. The RI p o p u l a t i o n and the m e t h o d e m p l o y e d here for the d e v e l o p m e n t o f N I L s from the RI p o p u l a t i o n will be a p o w e r f u l tool to u n d e r s t a n d the genetic basis of agriculturally i m p o r t a n t traits in rice. Acknowledgements We thank G. Wang for providing the RFLP and phenotypic data of a recombinant inbred lines of a cross between Moroberekan and CO39 used for the selection of pre-isogenic lines. We also thank S. D. Tanksley and S. R. McCouch for providing the RG and RZ clones. This work was supported by grants from the Ministry of Education, Science and Culture, Japan (No. 06760002), the IRRI-Japan Shuttle Research Project, and the Rockefeller Foundation's International Program on Rice Biotechnology.

References
Bronson MR (1994) Identification and characterization of blast-resistance genes from a durably resistant rice cultivar. MS thesis, Colorado State University Dellaporta SL, Wood J, Hicks JB (1984) Maize DNA miniprep. In: M. Russell (ed) Molecular biology of plants. A laboratory course manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, pp 36-37 Inoue T, Zhong HS, Miyao A, Ashikawa I, Monna L, Fukuoka S, Miyadera N, Nagamura Y, Kurata N, Sasaki T, Minobe Y (1994) Sequence-tagged sites (STSs) as standard landmarkers in the rice genome. Theor Appl Genet 89:728-734 Inukai T, Nelson RJ, Zeigler RS, Sarkarung S, Mackill DJ, Bonman JM, Takamure I, Kinoshita T (1994a) Allelism of blast-resistance genes in near-isogenic lines of rice. Phytopathology 84:12781283 Inukai T, Nelson RJ, Zeigler RS, Sarkarung S, Takamure I, Kinoshita T (1994b) Differentiation of pathogenic races of rice blast fungus by using near-isogenic lines with indica genetic background. J Fac Agr Hokkaido Univ 66:27-35 Kinoshita T, Inukai T, Toriyama K (1994) IV. Reports from coordinators. 1. Gene symbols for blast-resistance newly revised. Rice Genet Newslett 11 : 16-18 Kiyosawa S (1967) Inheritance of resistance of the rice variety Pi No. 4 to blast. Japan J Breed 17:165-172 Kiyosawa S (1972) The inheritance of blast-resistance transferred from some indica varieties in rice. Bull Nat Inst Agric Sci D23: 69-96 Mackill DJ, Bonman JM (1992) Inheritance of blast-resistance in near-isogenic lines of rice. Phytopathology 82:746-749 McCouch SR, Kochert G, Yu ZH, Wang ZY, Khush GS, Coffman WR, Tanksley SD (1988) Molecular mapping of rice chromosomes. Theor Appl Genet 76:815-829 Mew TV, Parco SA, Hittalmani S, Inukai T, Nelson RJ, Zeigler RS, Huang N (1994) Fine mapping of major genes for blast-resistance in rice. Rice Genet Newslett 11:126-128 Miyamoto M, Sato M, Ando I, Kodama O, Akatsuka T, Kawasaki S (1993) High resolution mapping of a rice blast-resistance gene Pi-b (in Japanese). Japan J Breed 43 (SuppI 2):220 Oka H (1955) Phylogenetic differentiation of the cultivated rice. XI. Change of gene frequency in hybrid populations of rice. (in Japanese with English summary) Japan J Breed 5:207-212 Sasaki T, Abe S, Matsunaga K, Okamoto E, Tanno K (1994) Breeding of a muItiline cultivar of paddy rice cultivar "Sasanishiki" for blast-resistance (in Japanese). Japan J Breed 44 (Suppl 2): 160 Satoh M, Miyamoto M, Ando I, Saito A, Kawasaki S (1993) RFLP mapping of resistance genes against rice blast disease. II. Pi-ta 2 and Pi-z t (in Japanese). Japan J Breed 43 (Suppl. 2):221 Tanksley SD, Young ND, Paterson AH, Bonierbale MW (1989) RFLP mapping in plant breeding: new tools for an old science. Bio/Technology 7:257-264 Tanksley SD, Causse M, Fulton T, Ahn N, Wang Z, Wu K, Xiao J, Yu Z, Second G, McCouch S (1992) A high-density molecular map of the rice genome. Rice Genet Newslett 9:111-115 Wang G (1992) RFLP mapping of major and minor genes for blastresistance in a durably resistant rice cultivar. PhD dissertation, University of the Philippines at Los Banos Wang G, Mackill DJ, Bonman JM, McCouch SR, Champoux M, Netson RJ (1994) RFLP mapping of genes conferring complete and partial resistance to blast in a durably resistant rice cultivar. Genetics 136:t421-1434 Yokoo M, Kiyosawa S (1970) Inheritance of blast-resistance of the rice variety, Toride 1, selected from the cross Norin 8xTKM. 1. Japan J Breed 20:129-132

567 Young ND, Tanksley SD (1989) Restriction fragment length polymorphism maps and the concept of graphical genotypes. Theor Appl Genet 77:95-101 Yu ZH (1991 ) Molecular mapping of rice (Oryza sativa L.) genes via linkage to restriction fragment length polymorphism (RFLP) markers. PhD dissertation, Cornell University Yu ZH, Mackill D J, Bonman JM, Tanksley SD (1991) Tagging genes for blast-resistance in rice via linkage to RFLP markers. Theor Appl Genet 81:471-476 Zeigler RS, Cuoc LX, Scott RE Bernardo MA, Chen DH, Valent B, Nelson RJ (1995) The relationship between lineage and virulence in Pyricularia grisea in the Philippines. Phytopathology 85: 443-451


相关文章:
更多相关标签: