Co-inoculations of bacteria and mycorrhizal fungi often drive additive plant growth responses

Abstract

Controlled greenhouse studies have shown the numerous ways that soil microbes can impact plant growth and development. However, natural soil communities are highly complex, and plants interact with many bacterial and fungal taxa simultaneously. Due to logistical challenges associated with manipulating more complex microbiome communities, how microbial communities impact emergent patterns of plant growth therefore remains poorly understood. For instance, do the interactions between bacteria and fungi generally yield additive (i.e. sum of their parts) or nonadditive, higher order plant growth responses? Without this information, our ability to accurately predict plant responses to microbial inoculants is weakened. To address these issues, we conducted a meta-analysis to determine the type (additive or higher-order, nonadditive interactions), frequency, direction (positive or negative), and strength that bacteria and mycorrhizal fungi (arbuscular and ectomycorrhizal) have on six phenotypic plant growth responses. Our results demonstrate that co-inoculations of bacteria and mycorrhizal fungi tend to have positive additive effects on many commonly reported plant responses. However, ectomycorrhizal plant shoot height responds positively and nonadditively to co-inoculations of bacteria and ectomycorrhizal fungi, and the strength of additive effects also differs between mycorrhizae type. These findings suggest that inferences from greenhouse studies likely scale to more complex field settings and that inoculating plants with diverse, beneficial microbes is a sound strategy to support plant growth.

Keywords: bacteria-mycorrhizal fungi interactions; bioinoculants; meta-analysis; microbial ecology; tripartite interactions.

Figure 1

Distribution of organisms used in AM studies. The composition of bacterial (A), fungal (B), and plant (C) genera of studies used in our meta-analysis is shown. The x-axis displays the six plants responses that were analyzed in our study. The abbreviations are as follows: MC = mycorrhizal fungi colonization percentage, RL = plant root length, RW = plant root weight, SH = plant shoot height, SW = plant shoot weight, and TPW = total plant weight. The total number of inoculants used in each analysis is denoted above each stacked bar. See Supplementary Table S1 for additional information about the selected studies. (D) Number of unique genera across studies.

Figure 2

Distribution of organisms used in EcM studies. The composition of bacterial (A), fungal (B), and plant (C) genera of studies used in our meta-analysis is shown. The x-axis displays the six plants responses that were analyzed in our study. The abbreviations are as follows: MC = mycorrhizal fungi colonization percentage, RL = plant root length, RW = plant root weight, SH = plant shoot height, SW = plant shoot weight, and TPW = total plant weight. The total number of inoculants used in each analysis is denoted above each stacked bar. See Supplementary Table S1 for additional information about the selected studies.

Figure 3

Biomass plant response effect sizes of single and co-inoculations of bacteria and mycorrhizal fungi. Changes in effect sizes (y-axis) for total plant weight (A and B), plant shoot weight (C and D), and plant root weight (E and F) are shown for both AM fungi and EcM fungi across different inoculation types (x-axis). The P-values for each comparison are provided, where P<.05 is considered a significant difference. Study information can be found in Figs 1 and 2 and Supplementary Table S1. The linear regression model outputs are listed in Table 1, Supplementary Tables S2, and S3.

Figure 4

Plant shoot height and root length effect size comparisons of single and co-inoculations of bacteria and mycorrhizal fungi. Changes in effect sizes (y-axis) for plant shoot height (A and B) and plant root length (C and D) are shown for both AM fungi and EcM fungi across different inoculation types (x-axis). The P-values for each comparison are provided, where P<.05 is considered a significant difference. Study information can be found in Figs 1 and 2 and Supplementary Table S1. The linear regression model outputs are listed in Table 1, Supplementary Tables S2, and S3.

Figure 5

Mycorrhizal fungi plant root colonization percentage effect size comparisons of single and co-inoculations of bacteria and mycorrhizal fungi. Changes in effect sizes (y-axis) for the colonization of AM fungi (A) and EcM fungi (B) across different inoculation types (x-axis) are shown. The P-values for each comparison are provided, where P<.05 is considered a significant difference. Study information can be found in Figs 1 and 2 and Supplementary Table S1. The linear regression model outputs are listed in Table 1, Supplementary Tables S2, and S3.