The development of nuclear restriction length polymorphism (RFLP) technology has been useful for the genetic analysis of various species and is of potential value in crop improvement as an aid to plant breeders in the selection of desired genotypes. However, RFLP technology is expensive and time consuming and technically difficult to use in some species with large and complex genomes.

The recently developed "Random Amplified Polymorphic DNA" (RAPD) marker system of Williams et al. (Nucl. Acid. Res. 18:6531-6535, 1990), that uses the polymerase chain reaction techniques to generate random amplified DNA markers, holds great promise for quickly placing markers on linkage groups. Accordingly, we have examined in our laboratory the reliability and reproducibility of the RAPD marker system with maize DNA and with a sample of random 10-mers.

The inbreds A7 and B73 were chosen from the maize germplasm collection present at the Experimental Station for Cereal Crops in Bergamo, Italy. The F1 generation and 45 F2 individuals segregating from the cross B73 x A7, which were already characterized by RFLP analysis, were used for the genetic mapping of RAPD markers. Genomic DNA was isolated from about 20 shoots of 10 to 15 day old seedlings derived from each progeny. DNA was purified according to the CTAB method as performed by Saghai-Maroof et al. (Proc. Natl. Acad. Sci. USA 81:8014-8018, 1984). Forty-seven random 10-mers (Operon Technologies Inc., Alameda, CA, USA) were used. The primers had 50-70% G+C content and no internal inverted repeats.

In order to decrease the background and to arrive at reproducible amplification patterns we have optimized the conditions described by Williams et al. (Nucl. Acid. Res. 18:6531-6535, 1990). To determine the effect of primer concentration, we tested three primers (OPA03, OPA04 and OPA05) at 0.1µM, 0.2µM, and 1.0µM final concentration, on genomic DNA extracted from B73, A7, and their F1 hybrid (B73 x A7). In addition, we evaluated three Mg final concentrations (1.5, 2.5, and 3.5mM) combined with OPA03 and OPA04 on the three genotypes. The effect of template concentration was assayed by amplifying in a final volume of 25µl, 2, 4, 10, and 50ng of genomic DNA from the inbred line A7 with the primers OPA02, OPA03 and OPA04. We have also verified hot (Taq-polymerase added as the last reagent to samples already at 94 C) versus cold start and the effect of UV light in reducing artifacts due to contamination with exogen DNA. Final volumes of 25µl and 50µl, with annealing temperatures of 36 C and 37 C, were also assayed. The results obtained suggest that 1) using 1.0µM primers a higher background was obtained, coupled with a lower number of amplification products compared to 0.2µM. Using 0.1µM primer concentration we obtained the same pattern observed with 0.2µM, but fainter bands. 2) Compared to lower concentrations, the use of 3.5mM Mg resulted in a more reproducible amplification of a larger number of distinct bands. 3) The best patterns were obtained using 4-10ng of template DNA in a reaction final volume of 25µl. Larger amounts of DNA gave rise to a non-specific smear. 4) The hot versus the cold start, annealing temperatures of 37 C versus 36 C, and the differences in the final volume had no evident effect. 5) The exposure of reagents to 280nm UV light for 60' significantly decreased the amplification of artifacts observed in the control reactions in our first attempts of producing RAPDs.

The optimal amplification conditions were obtained in a 25µl volume amplifying 4ng of DNA with 0.4U of Taq polymerase (Promega) at a final concentration of 10mM TrisHCl, pH 8.3, 50mM KCl, 3.5mM MgCl2, 200µM dNTPs, 02.µM primer, 0.01% BSA, 0.1% Tryton X-100. To verify the reproducibility of the results all experiments were replicated three to five times. Amplification was conducted with an MJ Programmable Thermal Controller (MJ Research Inc.) for 45 cycles of 1' at 94 C, 1' at 36 C and 2' at 72 C. After the last cycle samples were incubated for 10' at 72 C, and then conserved at 4 C. Samples were loaded on a 1.4% agarose gel, electrophoresed, stained with ethidium bromide, photographed and the distribution of markers among progenies recorded. The 1kb ladder (Bethesda Research Laboratories) was used as molecular weight standard. RAPD markers were named according to the origin (OP) followed by primer identification code and size of the amplification product in base pairs.

Out of 47 primers tested, 35 yielded, on agarose gel, distinct amplification products: the remaining 12 produced a smear. The number of fragments amplified averaged 2.57±1.12, and ranged from 1 to 5. Twelve of the 35 primers revealed polymorphic fragments. These primers produced a total of 17 clearly scorable products which were polymorphic between the two inbred lines B73 and A7 and additive in the F1 genome (Table 1). All non-clear, faint and non-additive bands were excluded from further analysis. In the genotypes evaluated here, the level of polymorphism found with RAPDs was comparable to the level of polymorphism found with RFLP. This suggests that RAPD markers can be used for the construction of genetic maps in maize and other polymorphic crops. Using a 10bp primer, the expected average number of amplified products from the amplification of the maize genome is 9.1 (Williams et al., Methods in Enzymology, Orlando, FL, USA, Academic Press, 1991). In soybean, the authors previously cited observed a higher than expected average number of products and proposed that a 95% homology between primers and annealing sites was sufficient for the PCR amplification. In our experiments, however, the average number of fragments scored was lower than expected. This may indicate that a 100% matching of the primer with the template DNA was necessary to obtain amplification in our laboratory conditions.

The segregation of the 17 RAPD markers was assessed in 45 individuals from the F2 progeny derived from the F1 hybrid B73 x A7, previously analyzed with RFLPs using MAPMAKER computer software (Lander et al., Genomics 1:174-181, 1987). The statistical analysis confirmed that RAPDs behaved as dominant Mendelian factors. Sixteen RAPDs have been located on a 1800cM maize genetic map already covered by 70 evenly distributed RFLP markers. They mapped on chromosomes 1, 2, 3, 4, 5, 6, 7 and 10; interestingly, opm082000, located on the short arm of chromosome 7, mapped 28.0cM distal to pio200581, the most distal mapped probe. This observation, together with the fact that RAPD probes are often not polymorphic and they may identify repetitive DNA sequences, suggests that RAPDs map in regions where polymorphic RFLP markers are rare.

Table 1. Primers used in RAPD analysis, quality of amplified products, maximum number of bands amplified in either parent and number of bands polymorphic between A7 and B73 inbreds.

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage