Trade-offs constrain adaptive pathways to T6 survival

Abstract

Many microbial communities are characterized by intense competition for nutrients and space. One way for an organism to gain control of these resources is by eliminating nearby competitors. The Type VI Secretion System (T6) is a nano-harpoon used by many bacteria to inject toxins into neighboring cells. While much is understood about mechanisms of T6-mediated toxicity, little is known about the ways that competitors can defend themselves against this attack, especially in the absence of their own T6. Here we use directed evolution to examine the evolution of T6 resistance, subjecting eight replicate populations of Escherichia coli to T6 attack by Vibrio cholerae. Over ~500 generations of competition, the E. coli evolved to survive T6 attack an average of 27-fold better than their ancestor. Whole genome sequencing reveals extensive parallel evolution. In fact, we found only two pathways to increased T6 survival: apaH was mutated in six of the eight replicate populations, while the other two populations each had mutations in both yejM and yjeP. Synthetic reconstruction of individual and combined mutations demonstrate that yejM and yjeP are synergistic, with yejM requiring the mutation in yejP to provide a benefit. However, the mutations we identified are pleiotropic, reducing cellular growth rates, and increasing susceptibility to antibiotics and elevated pH. These trade-offs underlie the effectiveness of T6 as a bacterial weapon, and help us understand how the T6 shapes the evolution of bacterial interactions.

Multiplicity of Type 6 Secretion System toxins limits the evolution of resistance

“…Third, our models focus on individual level traits, neglecting the potential for collective resistance mechanisms 27,28 . While our experimental parameters are designed to select for resistance at the single cell level (by using high cell density and high attacker:target cell ratio, so that a maximum number of target cells are individually subject to T6SS attack 34 , we cannot say which of our evolved isolates have a collective component to their T6SS resistance. Clearly both individual-and collective level defences exist against the T6SS 22 , and an interesting avenue of future research would be to compare the circumstances that lead to one form of resistance or another.…”

“…Finally, we considered the inclusion of fitness penalties associated with T6SS resistance (Fig. S6), which have been demonstrated in previous work by MacGillivray et al 34 . Including a fitness reduction proportional to toxin insensitivity further suppressed resistance evolution against multi-toxin attackers, since multi-resistance then became inherently more costly than single-toxin resistance.…”

“…Next, we sought to test this prediction empirically, using experimental evolution. Inspired by the recent work of MacGillivray and colleagues 34 , we conducted a serial transfer experiment tracking resistance evolution by Escherichia coli against Acinetobacter baylyi strains bearing a T6SS with either single or multiple toxins. Shown in Fig.…”

“…In the last 7 years, multiple alternate forms of defence have been reported, demonstrating T6SS protection via biofilm 27 or microcolony formation 28 , cell envelope stress responses 29 , capsule formation 30,31 , toxin-specific defence pathways 32 , or through loss of susceptibility factors 33 . In some cases, these resistance mechanisms can obviate the need for immunity proteins altogether 29 ; recent work has also demonstrated microbes' potential to evolve spontaneous resistance to T6SS attacks 34 . However, because these studies have generally focused on resistance mechanisms against T6SS attackers with fixed toxin arsenals, the role of specific toxins or toxin combinations in determining weapon susceptibility and evolutionary responses has remained thus far uncharacterised.…”

Spatial constraints and stochastic seeding subvert microbial arms race

“…Rock-paper-scissors game theory can lead to antagonism being a poor strategy. Further, strains can evolve various resistance mechanisms [38][39][40][41]. In a different vein, a toxin may not be effective against a novel competitor.…”