Submitted to: Botany
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 18, 2009
Publication Date: October 2, 2009
Repository URL: http://hdl.handle.net/10113/37171
Citation: Bell, D.J., Rowland, L.J., Zhang, D., Drummond, F.A. 2009. The spatial genetic structure of lowbush blueberry, Vaccinium angustifolium Ait., in four fields in Maine. Botany. 87:932-946 Interpretive Summary: Lowbush blueberry production, which makes up about 1/3 of the total blueberry production in the U.S., is from managed wild fields in Maine. Individual plants of wild lowbush blueberry are quite variable in terms of yield and we are investigating possible causes of these yield differences, including the genetic structure of wild populations. Lowbush blueberry is pollinated by rented honey bees which tend to fly short distances, thus plants tend to be pollinated by themselves or by near neighbors. Crosses between individual blueberry plants that are too closely related can result in low yields due to inbreeding. If the genetic structure of fields is such that closely related individuals tend to cluster together in patches within fields, then this could explain the low yield of some plants. Here we have used molecular markers to examine genetic relatedness of individuals within and among four wild blueberry fields in Maine, testing whether genetic distance is correlated with physical distance. We found across long distances, that plants from different fields are generally less related than those from the same field. However, across short distances within a field, plants do not tend to cluster in patches of highly related individuals. Therefore, the genetic structure within fields cannot explain these yield differences. This information will be useful to other scientists interested in improving yields in lowbush blueberry.
Technical Abstract: Expressed Sequence Tag-Polymerase Chain Reaction (EST-PCR) molecular markers were used to infer spatial genetic structure (SGS) of four lowbush blueberry (Vaccinium angustifolium Ait.) fields in Maine. Genetic structure was quantified at three spatial scales: 1) within apparent clones (or intrapatch), 2) among clones within a field, and 3) among fields separated by as much as 65 km. Of five clones examined in the intrapatch study, two showed complete genetic homogeneity within a patch, while three showed some band differences at their edges compared to their interiors. These differences at the edges, however, matched adjacent clones, so-called ‘intruders’, from which it was concluded that lowbush blueberry exhibits a tight, phalanx, clonal architecture with no evidence of invasive seedling establishment within clones. No significant correlation between genetic and physical distance was found among clones within fields via several statistical approaches. Significant among field genetic differentiation was found via AMOVA based upon transect samples across four fields ranging from 12.5-65 km apart. Principal Coordinate Analysis (PCA) and spatial autocorrelation (SA) corroborated these findings. Significant positive SA was found at the within field distance class of < 350 m, but decreased to an insignificant value by the first interfield distance of 12.5 km. A special form of SA analysis was employed to detect ‘hotspots’ of genetic similarity between pairs of adjacent clones in two fields. Results indicated that 5 of 23 pairs of clones (21.7%) were genetically similar to each other, while the majority of pairs (18 of 23; 78.3%) showed random and decreasing patterns of genetic similarity. Results are discussed in terms of clonal dynamics including architecture, seedling recruitment, and inferred pollen/seed dispersal distances.