Parentage analysis using RAPD PCR.

DNA variation from randomly amplified polymorphic DNA (RAPD) (1, 2) has recently been helpful in addressing a wide range of evolutionary problems including the determination of paternity and maternity (3). It has been greeted with particular enthusiasm because it is fast, technically easy and requires little material (4). By using single 10 base primers of arbitrary sequences to amplify DNA with the polymerase chain reaction (PCR) fingerprint-like bands are generated. Recently, however, concerns have been raised over the suitability of this method for determining parentage because non-parental bands often appear in offspring of known pedigrees (5). Here we report on the low occurrence of non-parental bands in two studies of parents and offspring using RAPD and make some suggestions on its application to avoid false interpretations from non-parental bands. RAPD PCR was carried out in 2 mM MgCl2 0.2-0.8 /iM primer and 10 ng genomic DNA for 45 cycles with an annealing temperature of 35°C (1). Thirty primers with repeatable amplification products yielded on average 6.7 bands each. A total of 76 variable fragments were used to determine the maternity and paternity of experimental broods of wild-caught, communally-breeding burying beetles (Nicrophorus tomentosus: Silphidae). Larvae of a given brood could have two possible mothers and fathers and the smaller possibility of unidentified fathers from previous inseminations. Parentage of 217 larvae from 10 broods was analyzed. In 1450 larvae-reactions, only 25 (0.017) of the reactions produced non-parental bands (data submitted but not shown). Eighty percent of these non-parental bands were present in other individuals in the study population (unrelated adults, N = 40) or siblings; these could have been amplified from DNA of the true father if sperm were from a previous insemination. The unique non-parental bands (0.003 of the amplifications) could have been generated by mutation, or amplified from a contaminant present during DNA extraction, including bacteria and parasites, or from unknown fathers not represented in the sample population. Apparent non-parental bands could also be the result of a PCR artefact in which bands (in the parent) may not be amplified due to competing kinetics of amplification.