Performance of microsatellites for parentage assignment following mass controlled pollination in a clonal seed orchard of loblolly pine (Pinus taeda L.)

Abstract

Mass controlled pollination (MCP), involving large-scale application of pollen on physically isolated female reproductive organs, has been a lower cost alternative to controlled pollination for the commercial production of genetically improved seeds. Nevertheless, rare are the studies that examined the efficacy of operational MCP and no such assessment has been done in loblolly pine to date. The success of MCP was assessed by a microsatellite-based investigation of the realized versus expected parentage of a set of 300 Pinus taeda offspring in 19 families generated in two subsequent rounds of MCP in 2005 and 2006 in a clonal seed orchard in Brazil. Multi-locus combined probability of parentage exclusion both theoretical and realized from actual testing was >99 % for single parent and parent pair testing when using nine or ten markers. Parentage assignments carried out under a maximum likelihood framework revealed a significantly higher success rate of MCP in 2006 (84 %) following technical improvements adopted to minimize pollen contamination and maximize male reproductive success, although significant variability in the correct maternity and full parentage was seen among individual families. The observed patterns of unexpected parentage indicated that this variability likely resulted from mislabeling of clonal ramets of the parents used in the crosses which impacted both maternity and paternity. Preventing pollen contamination will not be sufficient for successful MCP if inaccuracies exist in the identity of the clonal plants that ultimately provide pollen and female strobili, showing that DNA marker auditing and correction of identity of all ramets in a clonal seed orchard should be a standard practice in the operational implementation of MCP.