Synthetic Denitrifying Communities Reveal a Positive and Dynamic Biodiversity-Ecosystem Functioning Relationship during Experimental Evolution

Abstract

Biodiversity is vital for ecosystem functions and services, and many studies have reported positive, negative, or neutral biodiversity-ecosystem functioning (BEF) relationships in plant and animal systems. However, if the BEF relationship exists and how it evolves remains elusive in microbial systems. Here, we selected 12 Shewanella denitrifiers to construct synthetic denitrifying communities (SDCs) with a richness gradient spanning 1 to 12 species, which were subjected to approximately 180 days (with 60 transfers) of experimental evolution with generational changes in community functions continuously tracked. A significant positive correlation was observed between community richness and functions, represented by productivity (biomass) and denitrification rate, however, such a positive correlation was transient, only significant in earlier days (0 to 60) during the evolution experiment (180 days). Also, we found that community functions generally increased throughout the evolution experiment. Furthermore, microbial community functions with lower richness exhibited greater increases than those with higher richness. Biodiversity effect analysis revealed positive BEF relationships largely attributable to complementary effects, which were more pronounced in communities with lower richness than those with higher richness. This study is one of the first studies that advances our understanding of BEF relationships and their evolutionary mechanisms in microbial systems, highlighting the crucial role of evolution in predicting the BEF relationship in microbial systems. IMPORTANCE Despite the consensus that biodiversity supports ecosystem functioning, not all experimental models of macro-organisms support this notion with positive, negative, or neutral biodiversity-ecosystem functioning (BEF) relationships reported. The fast-growing, metabolically versatile, and easy manipulation nature of microbial communities allows us to explore well the BEF relationship and further interrogate if the BEF relationship remains constant during long-term community evolution. Here, we constructed multiple synthetic denitrifying communities (SDCs) by randomly selecting species from a candidate pool of 12 Shewanella denitrifiers. These SDCs differ in species richness, spanning 1 to 12 species, and were monitored continuously for community functional shifts during approximately 180-day parallel cultivation. We demonstrated that the BEF relationship was dynamic with initially (day 0 to 60) greater productivity and denitrification among SDCs of higher richness. However, such pattern was reversed thereafter with greater productivity and denitrification increments in lower-richness SDCs, likely due to a greater accumulation of beneficial mutations during the experimental evolution.

Keywords: Shewanella; biodiversity-ecosystem functioning relationship; complementarity effect; experimental evolution; selection effect; synthetic microbial community.

FIG 1

The overall relationship between the community function and species richness of synthetic denitrifying communities (SDCs). (a) A significant positive relationship was found for species richness and productivity (OD₆₀₀) using Spearman’s correlation. (b) A significant positive relationship was also observed between species richness and denitrification using Spearman’s correlation.

FIG 2

Relationships between species richness and the community function along the evolution experiment. (a) Significant positive linear relationships were found between richness and productivity (OD₆₀₀) at the 3rd day [slope = 0.015; F_{(1, 28)} = 20.70; P < 0.001], 30th day [slope = 0.014; F_{(1, 28)} = 12.04; P = 0.002], and 60th day [slope = 0.012; F_{(1, 28)} = 24.55; P < 0.001]. (b) Relationships between species richness and denitrification rate were significantly positive at the 3rd day [slope = 0.051; F_{(1, 28)} = 17.68; P < 0.001] and 30th day [slope = 0.015; F_{(1, 28)} = 7.81; P = 0.009].

FIG 3

The temporal dynamics of synthetic denitrifying community composition during the evolution experiment. Log-transferred absolute abundances of individual Shewanella species/strains in SDCs were shown as mean ± standard deviation (SD) (n = 6). (a) Four-species SDCs. (b) Twelve-species SDCs. Species abbreviations are as follows: S12, S. decolorationis S12; CN-32, S. putrefaciens CN-32; 400, S. frigidimarina NCIMB 400; W3_18_1, Shewanella sp. W3_18_1; PV-4, S. loihica PV-4; EP1, S. marisflavi EP1; YQH10, S. mangrovi YQH10; SB2B, S. amazonensis SB2B; BAA-318, S. fidelis ATCC BAA-318; OS155, S. baltica OS155; HAW-EB3, S. sediminis HAW-EB3; HAW-EB4, S. halifaxensis HAW-EB4.

FIG 4

The overall relationship between biodiversity effects of richness (a) or time (b) on synthetic denitrifying communities. Significant relationships were observed for net biodiversity and richness [F_{(1, 109)} = 29.55; P < 0.001], complementarity effect and richness [F_{(1, 109)} = 22.62; P < 0.001], and selection effect and richness [F_{(1, 109)} = 18.12; P < 0.001]; significant relationships were also observed for biodiversity and time [F_{(1, 109)} = 10.88; P = 0.001] and complementarity effect and time [F_{(1, 109)} = 10.14; P = 0.002]. Biodiversity effects were estimated by additive partitioning equation $\Delta Y=N\overline{\Delta \mathit{RY}}\overline{N}+N\mathrm{cov}\left(\Delta \mathit{RY},M\right)$, where ΔY reflects net biodiversity effect, $N\overline{\Delta \mathit{RY}}\overline{N}$ reflects complementarity effect, and $N\mathrm{cov}\left(\Delta \mathit{RY},M\right)$ reflects selection effect (49).

FIG 5

Relationships between richness and slopes of biodiversity effect versus time. (a) Complementarity. (b) Selection. (c) Net biodiversity effect.

FIG 6

The relationship between the relative yield total of each community and richness (a) or time (b). The relative yield total (RYT) was calculated by summing the relative yield of all species in a mixture, representing the overall community interaction, and the relative yield (RY) was for each species in their corresponding monocultures.

FIG 7

A schematic representation of possible mechanisms of biodiversity-ecosystem functioning relationships of synthetic microbial communities along the evolution experiment. A greater increase rate of complementarity and net biodiversity effects was observed at low-richness SDCs than at high-richness SDCs, and such asymmetrical changes of biodiversity effects could result in a flattened biodiversity-ecosystem functioning relationship in high-richness SDCs due to their lower increase rates along the evolution experiment. This less important role of selection effect along the long-term evolution experiment may be due to the increasing productivity of low-richness SDCs (especially the monoculture of poor competitors), which are less or not associated with competitive ability, while the increasing contribution of complementary effect in the biodiversity-ecosystem functioning relationship over the long-term may be due to the higher resource utilization efficiency of diverse SDCs.