The impact of urbanization on microbial community structure, pollinator behavior and mating system in the late-flowering plant Linaria vulgaris

Entering the XXIst century, humanity will face enormous challenges to its continued prosperity. This is because the ecosystem in which mankind participates is undergoing planetary scale alterations, giving birth to new unprecedented problems. The growth and changing distribution of human population is a constituent part of this so-called global change, and must be carefully studied in order to give the coming generations a chance for a better future.

One of the processes that strongly affects the global distribution of human populace is urbanization. A strong, stable and longterm trend of migration from rural zones to the cities exists worldwide, and it is predicted that by 2050, 66% of all Earth’s inhabitants will call one city or another their home. Because the well-being and in many instances survival of humans is dependent on a variety of services the ecosystem provides, understanding what impact urbanization has on ecosystems is crucial. That is also the goal of the emerging discipline of urban ecology, and the present dissertation attempts to help developing this novel field. 

Specifically, this thesis tackles issues concerning plant reproduction in cities. Two aspects linked to the pollination process of a plant species  were looked into: their population genetic structure and its nectar microbial communities. A wild plant called yellow toadflax (Linaria vulagaris) was investigated here because, among other reasons, it is common in urban areas (facilitating study), it features a closed flower architecture (so microbes are unlikely to enter its nectar via air) and it is self-incompatible (this type of mating system is expected to react strongly to urban habitat fragmentation).

First, microsatellite genetic markers were developed for L. vulgaris. These were then used to study population genetic structure of yellow toadflax and factors shaping it in city landscapes. In urban environments, the increased amount of sealed surfaces means that plant individuals may suffer from loss of suitable habitat and fragmentation of its remaining patches. Because smaller patches can only sustain smaller plant populations, rare gene alleles are more likely to be lost due to random events. Such alleles may disappear with time, rendering urban populations less genetically diverse. Moreover, because of the fragmentation between the remnant habitat patches, the exchanges of genes between populations is more difficult.

The results of the experiments presented here indicated that L. vulgaris populations were highly differentiated, and that there was no correlation between genetic and geographic distances. This, as hypothesized, suggested low gene flow. A further paternity analysis on the subset of the populations revealed that current pollen flow patterns are probably higher than those estimated from standing genetic variation. This suggested that although pollen does travel between populations, many of the resulting seedlings do not reach sexual maturity. In all likelihood, the seedlings cannot establish themselves in such small patches due to competition with already mature individuals. In other words, it is habitat loss rather than habitat fragmentation that contributes to population differentiation.

Nevertheless, in the study system researched in this dissertation genetic diversity was not lower in urban regions than in agricultural zones. This was contrary to expectations. One possible reason for such indifference may be relative similarity, in terms of environmental impact, of landscapes undergoing intensive agricultural regime, and mildly urbanized zones as exemplified by the city of Leuven (in the vicinity of which the experiments presented in this dissertation were conducted), which according to United Nations classifications is a small city.

Second, nectar microbial communities of L. vulgaris were investigated. Floral nectar yeasts and bacteria are thought to be mainly distributed by insects, the same ones that provide pollination services. Thus, changes in incidence, abundance and structure of pollinator communities might not only impact plant population structure, but also the composition of their nectar microbial communities, too. In the study system investigated here, the microbial incidence in L. vulgaris was high and did not change along the investigated urbanization gradient. The communities were species poor, and no particular species, except for singletons, was present only in urbanizaed or rural zones. The sampling of bacteria may also have been incomplete, as indicated by rarefaction curves. Thus, although the microbial communities showed a nested substructure, indicating that species-poor communities were a subset of species-rich communities, it is difficult to speculate on the relevance of this result.

In search for potential factors responsible different conditions experienced by nectar microbial communities in urban and rural landscapes, the impact of an agricultural fungicide mixture was assessed on nectar yeasts both in vitro and in planta. The results indicated that pesticides hamper the growth of Metschnikowia reukaufii and Metschnikowia gruessii, leading to a decrease of their number in treated plants. However, nectar bacteria, which are presumably competitors of nectar yeasts, do not reach higher abundances as a result of hampered yeast growth.