Bacteriophages affect evolution of bacterial communities in spatially distributed habitats: a simulation study

Abstract

Background: Bacteriophages are known to be one of the driving forces of bacterial evolution. Besides promoting horizontal transfer of genes between cells, they may induce directional selection of cells (for instance, according to more or less resistance to phage infection). Switching between lysogenic and lytic pathways results in various types of (co)evolution in host-phage systems. Spatial (more generally, ecological) organization of the living environment is another factor affecting evolution. In this study, we have simulated and analyzed a series of computer models of microbial communities evolving in spatially distributed environments under the pressure of phage infection.

Results: We modeled evolving microbial communities living in spatially distributed flowing environments. Non-specific nutrient supplied in the only spatial direction, resulting in its non-uniform distribution in environment. We varied the time and the location of initial phage infestation of cells as well as switched chemotaxis on and off. Simulations were performed with the Haploid evolutionary constructor software ( http://evol-constructor.bionet.nsc.ru/ ).

Conclusion: Simulations have shown that the spatial location of initial phage invasion may lead to different evolutionary scenarios. Phage infection decreases the speciation rate by more than one order as far as intensified selection blocks the origin of novel viable populations/species, which could carve out potential ecological niches. The dependence of speciation rate on the invasion node location varied on the time of invasion. Speciation rate was found to be lower when the phage invaded fully formed community of sedentary cells (at middle and late times) at the species-rich regions. This is especially noticeable in the case of late-time invasion. Our simulation study has shown that phage infection affects evolution of microbial community slowing down speciation and stabilizing the system as a whole. This influencing varied in its efficiency depending on spatially-ecological factors as well as community state at the moment of phage invasion.

Fig. 1

The hierarchical scheme of the HEC model (according to [35, 38])

Fig. 2

Simulation of horizontal gene transfer in the HEC (according to [40])

Fig. 3

Spatial organization of the habitat. Mesh of 5х5 nodes. The flow determines the gradient of non-specific substrates. Green arrows depict the directions of diffusion. Red arrows show an example of chemotactic behavior of cells

Fig. 4

Trophic graph of the initial community. N1 – non-specific substrate consumed by all populations (P1, P2, P3) of the community. S1, S2, S3 – specific substrates synthesized by corresponding cells. $P1\stackrel{S2}{\to }P2$ means that cells of P1 population produce S2 substrate, which is consumed by cells of P2 population

Fig. 5

SRI calculated for the first 5000 generations in various nodes

Fig. 6

Average number of species (up) and biomass (down) over the first 5000 generations in various nodes (see also the script in Additional file 3)

Fig. 7

Dependence of SRI on time of invasion and invasion initial localization. Blue plot – early-time (1st generation), orange – middle-time (5000th generation), grey – late-time (6600th generation). X-axis – number of the node of initial invasion, Y-axis – average SRI calculated up to the end of simulation

Fig. 8

Distributions of average values for (a) species richness; (b) SRI; (c) total biomass in spatial nodes. Values were calculated for the first 5000 generations

Fig. 9

Dependence of SRI on time of invasion and invasion initial localization (chemotaxis is on). Blue plot – early-time (1st generation), orange – middle-time (5000th generation) , grey – late-time (6600th generation). X-axis – number of the node of initial invasion, Y-axis – average SRI calculated up to the end of simulation (built according to data from Additional file 5)