Analysis of primer-derived, nonspecific amplification products in RAPD-PCR.

Recently, we applied the randomamplified polymorphic DNA polymerase chain reaction (RAPD-PCR) (17), also known as arbitrarily primed PCR (AP-PCR) (16), procedure to fingerprint genomes of various varieties and wild relatives of sugarcane (Saccharum spp.) (4). Unlike restriction fragment length polymorphism (RFLP) (2), RAPD-PCR is a simpler, faster, less laborious and less expensive procedure. However, it has a few disadvantages. Most polymorphic RAPD markers are dominant, and therefore, heterozygotes are not accurately distinguished (17). RAPD markers may not be reproducible between experiments or laboratories (1,6,10,15). Researchers must standardize RAPD conditions (7,10) or use RAPD markers that are reproducible at least in a set environment (5,15). Another problem that, to our knowledge, has largely been ignored in RAPD-PCR literature is the existence of primer-derived, nonspecific amplification products in negative control reactions containing all the reaction components except for a DNA template. Such primer-derived artifacts were observed by Williams et al. in the original RAPD paper (17) and later by Tingey and del Tufo (13), Tingey et al. (14) and Lanham et al. (9). Nonspecific products were also observed in negative control reactions with different pairs of universal rDNA primers (3) and with human immunodeficiency virus (HIV)specific primers SK145/SK431 (7). These artifacts presumably are absent if a genomic DNA template is included in the reaction (7,11,13,14,17). We show that 19 of 20 Operon A Kit primers and 4 of 20 Operon C Kit primers (Operon Technologies, Alameda, CA, USA) produced artifact DNA bands in negative control reactions. We further demonstrate that some of these primerderived DNA products were amplified even in the presence of a DNA template. RAPD-PCR was performed in a PTC-100 Programmable Thermal Controller (Model 60; MJ Research, Watertown, MA, USA) in a 25-μL volume. Unless mentioned otherwise, the reaction mixture contained 10 mM Tris-HCl, pH 8.3, 10 mM KCl, 1.5 mM MgCl2, 0.2 mM each of dATP, dTTP, dGTP and dCTP, 0.4 μM primer, 2 U of Stoffel fragment (Perkin-Elmer, Norwalk, CT, USA), 25 ng of genomic DNA and was overlaid with two drops of mineral oil. The following thermal cycling program was used: 95°C for 5 min; 40 cycles at 93°C for 40 s; 40°C for 30 s and 72°C for 1 min; with a final extension at 72°C for 5 min; with a ramp speed of 2.36 s per 1°C from 93°C to 40°C, 1.73 s per 1°C from 40°C to 72°C, and 2.31 s per 1°C from 72°C to 93°C. Amplified PCR products were resolved by electrophoresis at 5.7 V/cm in 0.7% agarose/0.4% Synergel (Diversified Biotech, Boston, MA, USA) binary gels containing ethidium bromide (EtdBr; 0.5 μg/mL) in 0.5× TBE buffer for 1.5 h and were visualized on a UV transilluminator (Spectronics, Westburg, NY, USA). A total of 40 primers from two 10mer kits (A and C) were tested. All primers have a 60% or 70% G + C content. Without a DNA template, all but OPA-20 of the A kit and four (OPC-02, -05, -08 and -11) of 20 C-kit primers produced polymorphic DNA bands (Figure 1A, OPA-01, -02 and -11 not shown). The sizes of these DNA products ranged from approximately 100 to greater than 2000 bp. The experiment was repeated twice, and polymorphic DNA bands were produced each time. Amplified polymorphic DNA bands were also produced when the annealing temperature was 34°C or when three other brands of DNA polymerase were used: Taq DNA Polymerase from Perkin-Elmer or Promega (Madison, WI, USA) and AmpliTaq DNA Polymerase from Perkin-Elmer (Figure 1B). No amplified DNA products were observed when both DNA template and primer were omitted from the reaction or when the primer was omitted from the RAPDPCR mixture (Figure 1A). Contrary to our finding, OPA-17 did not produce any DNA product in a DNA-minus control in another RAPD study (8). Reagents prepared in our laboratory