Bulked segregant analysis confirms the importance of the region near umc89a for days to pollen shed in maize
--Tuberosa, R, Salvi, S, Phillips, RL

The moderate to high heritability of the number of days from planting to pollen shed (DPS) in maize makes it an interesting candidate for a molecular marker analysis of the quantitative trait loci (QTLs) underlying its expression. Previous work carried out at the University of Minnesota identified QTLs for DPS on chromosomes 1, 2, 3, 5, 7, and 8 (Kim, PhD thesis, University of Minnesota, 1992; Phillips et al., Proc. 47 Annu. Corn Sorghum Ind. Res. Conf., pp 135-150, 1992; Parentoni, MS thesis, University of Minnesota, 1993). The most important QTL for DPS was on chr. 8L in the region near umc89a. Surveying a number of unrelated crosses will increase the chances of identifying major QTLs representing valuable targets for gene cloning. Bulked segregant analysis (BSA; Michelmore et al., PNAS 88:9828-9832, 1991) offers an interesting opportunity to investigate specific QTLs in populations for which genetic maps are not available. In this case, RFLP probes tagging QTLs in one population can be tested in different populations segregating for the same trait. Our objective was to use BSA in order to verify whether the chromosomal regions influencing DPS in the work carried out by Phillips and collaborators at the University of Minnesota were also involved in the control of DPS in three F2 populations segregating for DPS. In particular, we wanted to verify the role of the QTL for DPS revealed by umc89.

Three F2 populations (Mt42 x NC254, A679 x NC254, and A86-9 x R225) were evaluated at the St. Paul experimental location of the Department of Agronomy and Plant Genetics, University of Minnesota. The number of F2 plants evaluated was 520 in Mt42 x NC254, 350 in A679 x NC254, and 280 in A86-9 x R225. All F2 plants were scored for DPS and selfed. Among the plants which set seed in each F2 population, the 15 earliest and the 15 latest in DPS were selected. The corresponding F3 progenies were scored for DPS at the experimental location of the Department of Agronomy, University of Bologna. The DNA bulks were obtained by bulking equal amounts of the DNA of the 15 earliest and the 15 latest F2 plants. The DNA of each bulk and that of the parent lines was restricted with two to three restriction enzymes (BamHI, EcoRI, EcoRV, and HindIII). Filters were probed with 14 genomic clones (bnl6.22, bnl6.25, umc6, umc11, umc27, umc30, umc35, umc51, umc54, umc89, umc90, umc113, umc114, and umc124) which showed RFLPs associated with QTLs for DPS in the work of Kim (PhD thesis, University of Minnesota, 1992). Filters were also probed with 24 clones (bnl5.40, bnl5.71, bnl15.21, csu4, csu6, csu13, csu29, csu46, csu61, csu86, csu94, csu109, csu133, csu148, umc15, umc31, umc59, umc84, umc93, umc104, umc119, umc128, umc139, and umc161) mapping to regions not represented by the previous group of clones.

Among the probes which showed polymorphism between parent lines of each cross, marked differences in intensity between the bands of the early- and late-maturity bulks were revealed by csu61, umc11, umc15, umc27, umc30, umc51, umc54, and umc89 (Table 1). These eight probes were all but two (csu61 and umc15) among those reported by Kim (PhD thesis, University of Minnesota, 1992) to be linked to QTLs for DPS in maize crosses different from those herein tested. Differences in band intensity of early and late bulks were not evidenced by any of the remaining probes. A key factor when using BSA for quantitative traits is the correct scoring of the phenotypes. In our study, DNA pools were formed according to the data for DPS of the F2 plants. The major discrepancies between the phenotypic classification of the F2 plants and their F3 progenies involved two A679 x NC254 F3 progenies which were found to be slightly earlier than the average of the early and late F3 progenies, although they derived from F2 plants classified as late. Despite this, Southern analysis with umc89 showed a marked difference in the intensity of the polymorphism between the bulks.

Our results indicate that BSA using RFLPs may be utilized to identify and/or confirm the role of QTLs when previous information is available on their localization in unrelated populations. Five QTLs controlling DPS in the crosses surveyed by Kim (1992) were likely involved in the control of DPS in at least one of the three crosses herein investigated. Due to the duplicate nature of the maize genome, some of the probes negative to BSA and linked to QTLs for DPS in the work of Kim (PhD thesis, University of Minnesota, 1992), may have revealed RFLPs mapping in regions other than those investigated in his work; in such a case, the presence of a QTL for DPS near the duplicate locus is less likely. umc11, umc30, and umc89 revealed differences between the bulks in two crosses. So far, umc89a has been linked to QTLs for DPS in several maize crosses (Abler et al., Crop Sci. 31:267-274, 1991; Phillips et al., Proc. 47 Annu. Corn Sorghum Ind. Res. Conf., pp 135-150, 1992; Zehr et al., TAG 33:903-911, 1992; Stuber et al., Genetics 132:823-839, 1992; Koester et al., Crop Sci. 33:1209-1216, 1993; Lin et al., Genetics 141:391-411, 1995; Ragot et al., Crop Sci. 35:1306-1315, 1995). Genes affecting flowering time are usually divided into two broad categories: i) genes responding to environmental cues, such as temperature and photoperiod, and ii) genes affecting flowering per se, independent of the influence of environmental variables. Although these two types of genes may not be mutually exclusive, the QTL region near umc89a likely contains a gene(s) affecting flowering per se. Coding sequences belonging to the category of flowering genes influenced by the growth environment (e.g. photoperiod) are less likely to maintain their effect over a wide range of environments and genetic backgrounds.

Table 1. Probe-enzyme combinations with marked differences in the intensity of the polymorphic bands between the bulked DNAs of the F2 plants which were classified as early and late in DPS in each cross. The combinations positive to BSA have been underlined.
 

Probe	Chr.	Mt42 x NC254	A679 x NC254	A86-9 x R225
umc11	1	BamHI	HindIII	BamHI
csu61	1	N.P.	N.P.	EcoRI
umc15	4	EcoRI	N.P.	N.P.
umc27	5	HindIII	EcoRV	HindIII
umc54	5	EcoRI	N.P.	EcoRI-HindIII
umc51	5	EcoRI	EcoRI	EcoRI
umc89	8	DraI	EcoRI	EcoRI
umc30	8	DraI	EcoRI	DraI-EcoRI

N.P.: probe-enzyme combination not polymorphic between parent lines.

Return to the MNL 72 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage