Quantifying pollen transfer between cultivated and wild Cyclopia species in South Africa

Abstract

Gene flow between cultivated populations and their co-occurring wild relatives can result in genetic contamination, with significant consequences in terms of the loss of genetic resources, possible extinction of locally adapted variations, and hybridization. Honeybush, Cyclopia Vent., is an endemic genus in the fynbos biome of South Africa with commercial value in the tea industry. With the growth of the industry, the shift from wild-harvested material to cultivated biomass is promoted to ensure ecological sustainability and avoid over harvesting. However, cultivating indigenous species in their native range may cause pollen-mediated gene flow between cultivated and wild populations, causing depletion of the genetic resources essential for the tea industry of South Africa, threatening local gene pools and possibly resulting in hybridization in the wild. To establish the potential for gene flow through pollen-transfer, the pollinators of four commercially important Cyclopia species; namely Cyclopia genistoides, C. subternata, C. maculata and C. intermedia were identified. Pollinator movement was investigated using mark-release-recapture and radio-tagging. Pollen longevity was determined under field conditions to indicate the distance at which pollen can be transferred. Hand-crossing experiments were conducted to determine the capacity for between and within species crosses in Cyclopia. Commercially important Cyclopia species are pollinated only by carpenter bees, Xylocopa, including Xylocopa capitata, X. flavorufa, X. caffra, X. rufitarsis, X. scioensis and X. sicheli. Carpenter bees are generalist foragers with indiscriminate foraging behaviours on a variety of Cyclopia species. Mark-release-recapture of carpenter bees revealed numerous cases of movement between cultivated and wild Cyclopia populations. In addition, radio-tracking confirmed between-site movement, with a maximum distance of 1194 m travelled in a single foraging bout and daily home range sizes (of up to 23 893 m²) spanning across cultivated and wild Cyclopia populations. Additionally, Cyclopia pollen remains viable for at least five days. Crosses between C. subternata x C. genistoides and C. subternata x C. maculata produced hybrid seeds, revealing the likelihood for hybridization to occur with pollen-transfer between cultivated and wild populations. These results signify the necessity for a protocol guiding the planting of Cyclopia, in order to avoid genetic homogenization and erosion, which is presented in preliminary form for the use by the honeybush tea industry. This protocol considers various barriers to genetic contamination, including the natural range, ploidy, seed dispersal and pollen-flow distances, and seed source. The protocol provides a valuable tool for determining the risk of introgression at individual plantations that can be used not only for Cyclopia, but adapted to other indigenous agricultural flora. In order to ensure the sustainable use of indigenous crops, management in the form of a planting protocol, like the one presented here, is critical.