Association with a novel protective microbe facilitates host adaptation to a stressful environment

Abstract

Protective symbionts can allow hosts to occupy otherwise uninhabitable niches. Despite the importance of symbionts in host evolution, we know little about how these associations arise. Encountering a microbe that can improve host fitness in a stressful environment may favor persistent interactions with that microbe, potentially facilitating a long-term association. The bacterium Bacillus subtilis protects Caenorhabditis elegans nematodes from heat shock by increasing host fecundity compared to the nonprotective Escherichia coli. In this study, we ask how the protection provided by the bacterium affects the host's evolutionary trajectory. Because of the stark fitness contrast between hosts heat shocked on B. subtilis versus E. coli, we tested whether the protection conferred by the bacteria could increase the rate of host adaptation to a stressful environment. We passaged nematodes on B. subtilis or E. coli, under heat stress or standard conditions for 20 host generations of selection. When assayed under heat stress, we found that hosts exhibited the greatest fitness increase when evolved with B. subtilis under stress compared to when evolved with E. coli or under standard (nonstressful) conditions. Furthermore, despite not directly selecting for increased B. subtilis fitness, we found that hosts evolved to harbor more B. subtilis as they adapted to heat stress. Our findings demonstrate that the context under which hosts evolve is important for the evolution of beneficial associations and that protective microbes can facilitate host adaptation to stress. In turn, such host adaptation can benefit the microbe.

Keywords: Experimental evolution; host adaptation; protective microbes.

Figure 1

Schematic of experimental evolution and assays. (A) Nematodes were passaged on the ancestral B. subtilis (B+ hosts, blue) or on the ancestral E. coli (B– hosts, green), under heat shock (34°C, H+) or no heat shock (20°C, H–) conditions, for 20 generations of selection (40 total generations). After each heat shock, hosts recovered on GFP‐labeled E. coli (gray) to produce offspring. The offspring of these offspring were then placed on fresh plates of their respective bacteria to be heat shocked, starting the next generation. There were five replicate populations for each of the four treatments. (B) To measure host fecundity after 20 generations of selection, we reared hosts from each of the 20 replicate experimental populations and the ancestral population on either the ancestral B. subtilis or E. coli, then heat shocked them at 34°C or left them at 20°C, following the same schedule for one passage of experimental evolution. Two days after the heat shock, we measured the number of offspring per total number of initial adults. To measure B. subtilis colonization, we heat shocked 15 replicate populations (excluding the five B–H– populations) on B. subtilis following the schedule for one passage of experiment evolution. Immediately after heat shock, we washed and crushed nematodes and plated them on media to quantify CFUs in individual hosts.

Figure 2

Fecundity of evolved hosts after 20 generations of selection. The x‐axis indicates the condition under which nematodes evolved. Nematodes from the four experimental treatments were heat shocked at 34°C on (A) B. subtilis or (B) E. coli. Each plate contained roughly 200 nematodes. The data are combined across three rounds. (C) Nematodes from the four experimental treatments were kept at 20°C on B. subtilis (blue points) or E. coli (green points). Each plate contained roughly 150 nematodes. The dotted line and dashed line indicate the average value for the ancestral host on B. subtilis and E. coli, respectively. Error bars indicate the standard errors. Treatments that are not the same letter are significantly different. Note the y‐axes for panels A and B versus panel C differ by almost 20‐fold. Figure S2 shows the distribution of all individual data points for this figure.

Figure 3

Bacillus subtilis colonization in evolved hosts. Evolved nematodes were heat shocked on B. subtilis, washed, and individually crushed to quantify within‐host bacterial colonization. The x‐axis indicates the condition under which nematodes evolved. (A) The number of colony‐forming units (CFUs) in each nematode. (B) The proportion of nematodes harboring at least one CFU. Each data point is the average of 10 nematodes from each replicate population from experimental evolution. The data are combined across four rounds. The dotted line indicates the average value for the ancestral host. Error bars indicate the standard errors. ^** P< 0.01^{, *} P < 0.05.

Figure 4

Host fecundity versus B. subtilis colonization when heat shocked after generation 20. Fecundity is plotted against CFUs per host. Error bars indicate standard errors.