Expression-environment associations in transcriptomic heat stress responses for a global plant lineage.

The increasing frequency and severity of heatwaves will intensify stress on plants. Given regional variation in heatwave exposure and expected differences in thermal tolerance between species it is unlikely that all plant species will be affected equally by climate change. However, little is currently known about variation in the responses of plants to heat stress, or how those responses differ among closely related species adapted to different environments. Here we quantify the response of 17 Acacia species (175 RNA-seq libraries), from across Australia's diverse biomes, to a multi-day experimental heatwave treatment to identify variation in transcriptomic and physiological responses to heat stress. Genes with known heat response functions showed consistent responses across Acacia species. Up to 10% of all genes and over 100 gene families showed significant clinal variation in the magnitude of their expression plasticity across species. Specifically, gene families linked to the temperature stress response were overrepresented among significant relationships with home range temperature conditions. Gene expression responses seen on the first day of the heatwave were more frequently associated with home range climates, while expression responses by day four were more commonly related to photosystem II acclimation. Comparative transcriptomics on non-model species has the potential to provide key information on stress response plasticity, especially when linked with our understanding of model species. Our study indicates that the pressing challenge to identifying potentially vulnerable species to climate change could be benefited by the further exploration of clinal variation in transcriptome plasticity.

Negative effects of nitrogen override positive effects of phosphorus on grassland legumes worldwide.

Anthropogenic nutrient enrichment is driving global biodiversity decline and modifying ecosystem functions. Theory suggests that plant functional types that fix atmospheric nitrogen have a competitive advantage in nitrogen-poor soils, but lose this advantage with increasing nitrogen supply. By contrast, the addition of phosphorus, potassium, and other nutrients may benefit such species in low-nutrient environments by enhancing their nitrogen-fixing capacity. We present a global-scale experiment confirming these predictions for nitrogen-fixing legumes (Fabaceae) across 45 grasslands on six continents. Nitrogen addition reduced legume cover, richness, and biomass, particularly in nitrogen-poor soils, while cover of non-nitrogen-fixing plants increased. The addition of phosphorous, potassium, and other nutrients enhanced legume abundance, but did not mitigate the negative effects of nitrogen addition. Increasing nitrogen supply thus has the potential to decrease the diversity and abundance of grassland legumes worldwide regardless of the availability of other nutrients, with consequences for biodiversity, food webs, ecosystem resilience, and genetic improvement of protein-rich agricultural plant species.

Generalized mutualisms promote range expansion in both plant and ant partners.

Mutualism improves organismal fitness, but strong dependence on another species can also limit a species' ability to thrive in a new range if its partner is absent. We assembled a large, global dataset on mutualistic traits and species ranges to investigate how multiple plant-animal and plant-microbe mutualisms affect the spread of legumes and ants to novel ranges. We found that generalized mutualisms increase the likelihood that a species establishes and thrives beyond its native range, whereas specialized mutualisms either do not affect or reduce non-native spread. This pattern held in both legumes and ants, indicating that specificity between mutualistic partners is a key determinant of ecological success in a new habitat. Our global analysis shows that mutualism plays an important, if often overlooked, role in plant and insect invasions.

Nonsymbiotic legumes are more invasive, but only if polyploid.

Both mutualism and polyploidy are thought to influence invasion success in plants, but few studies have tested their joint effects. Mutualism can limit range expansion when plants cannot find a compatible partner in a novel habitat, or facilitate range expansion when mutualism increases a plant's niche breadth. Polyploids are also expected to have greater niche breadth because of greater self-compatibility and phenotypic plasticity, increasing invasion success. For 847 legume species, we compiled data from published sources to estimate ploidy, symbiotic status with rhizobia, specificity on rhizobia, and the number of introduced ranges. We found that diploid species have had limited spread around the globe regardless of whether they are symbiotic or how many rhizobia partners they can host. Polyploids, by contrast, have been successfully introduced to many new ranges, but interactions with rhizobia constrain their range expansion. In a hidden state model of trait evolution, we also found evidence of a high rate of re-diploidization in symbiotic legume lineages, suggesting that symbiosis and ploidy may interact at macroevolutionary scales. Overall, our results suggest that symbiosis with rhizobia limits range expansion when legumes are polyploid but not diploid.

Priority effects alter interaction outcomes in a legume-rhizobium mutualism.

Priority effects occur when the order of species arrival affects the final community structure. Mutualists often interact with multiple partners in different orders, but if or how priority effects alter interaction outcomes is an open question. In the field, we paired the legume Medicago lupulina with two nodulating strains of Ensifer bacteria that vary in nitrogen-fixing ability. We inoculated plants with strains in different orders and measured interaction outcomes. The first strain to arrive primarily determined plant performance and final relative abundances of rhizobia on roots. Plants that received effective microbes first and ineffective microbes second grew larger than plants inoculated with the same microbes in the opposite order. Our results show that mutualism outcomes can be influenced not just by partner identity, but by the interaction order. Furthermore, hosts receiving high-quality mutualists early can better tolerate low-quality symbionts later, indicating that priority effects may help explain the persistence of ineffective symbionts.

Environmental variation impacts trait expression and selection in the legume-rhizobium symbiosis.

PREMISE: The ecological outcomes of mutualism are well known to shift across abiotic or biotic environments, but few studies have addressed how different environments impact evolutionary responses, including the intensity of selection on and the expression of genetic variance in key mutualism-related traits. METHODS: We planted 30 maternal lines of the legume Medicago lupulina in four field common gardens and compared our measures of selection on and genetic variance in nodulation, a key trait reflecting legume investment in the symbiosis, with those from a previous greenhouse experiment using the same 30 M. lupulina lines. RESULTS: We found that both the mean and genetic variance for nodulation were much greater in the greenhouse than in the field and that the form of selection on nodulation significantly differed across environments. We also found significant genotype-by-environment (G × E) effects for fitness-related traits that were generated by differences in the rank order of plant lines among environments. CONCLUSIONS: Overall, our results suggest that the expression of genotypic variation and selection on nodulation differ across environments. In the field, significant rank-order changes for plant fitness potentially help maintain genetic variation in natural populations, even in the face of directional or stabilizing selection.

Novel model-based clustering reveals ecologically differentiated bacterial genomes across a large climate gradient.

A pervasive challenge in microbial ecology is understanding the genetic level where ecological units can be differentiated. Ecological differentiation often occurs at fine genomic levels, yet it is unclear how to utilise ecological information to define ecotypes given the breadth of environmental variation among microbial taxa. Here, we present an analytical framework that infers clusters along genome-based microbial phylogenies according to shared environmental responses. The advantage of our approach is the ability to identify genomic clusters that best fit complex environmental information whilst characterising cluster niches through model predictions. We apply our method to determine climate-associated ecotypes in populations of nitrogen-fixing symbionts using whole genomes, explicitly sampled to detect climate differentiation across a heterogeneous landscape. Although soil and plant host characteristics strongly influence distribution patterns of inferred ecotypes, our flexible statistical method enabled us to identify climate-associated genomic clusters using environmental data, providing solid support for ecological specialisation in soil symbionts.

More partners, more ranges: generalist legumes spread more easily around the globe.

How does mutualism affect range expansion? On the one hand, mutualists might thrive in new habitats thanks to the resources, stress tolerance or defence provided by their partners. On the other, specialized mutualists might fail to find compatible partners beyond their range margins, limiting further spread. A recent global analysis of legume ranges found that non-symbiotic legumes have been successfully introduced to more ranges than legumes that form symbioses with rhizobia, but there is still abundant unexplained variation in introduction success within symbiotic legumes. We test the hypothesis that generalist legumes have spread to more ranges than specialist legumes. We used published data and rhizobial 16S rRNA sequences from GenBank to quantify the number of rhizobia partners that associate with 159 legume species, spanning the legume phylogeny and the globe. We found that generalist legumes occur in more introduced ranges than specialist legumes, suggesting that among mutualists, specialization hinders range expansions.

Standing genetic variation in host preference for mutualist microbial symbionts.

Many models of mutualisms show that mutualisms are unstable if hosts lack mechanisms enabling preferential associations with mutualistic symbiotic partners over exploitative partners. Despite the theoretical importance of mutualism-stabilizing mechanisms, we have little empirical evidence to infer their evolutionary dynamics in response to exploitation by non-beneficial partners. Using a model mutualism-the interaction between legumes and nitrogen-fixing soil symbionts-we tested for quantitative genetic variation in plant responses to mutualistic and exploitative symbiotic rhizobia in controlled greenhouse conditions. We found significant broad-sense heritability in a legume host's preferential association with mutualistic over exploitative symbionts and selection to reduce frequency of associations with exploitative partners. We failed to detect evidence that selection will favour the loss of mutualism-stabilizing mechanisms in the absence of exploitation, as we found no evidence for a fitness cost to the host trait or indirect selection on genetically correlated traits. Our results show that genetic variation in the ability to preferentially reduce associations with an exploitative partner exists within mutualisms and is under selection, indicating that micro-evolutionary responses in mutualism-stabilizing traits in the face of rapidly evolving mutualistic and exploitative symbiotic bacteria can occur in natural host populations.

Herbivory eliminates fitness costs of mutualism exploiters.

A common empirical observation in mutualistic interactions is the persistence of variation in partner quality and, in particular, the persistence of exploitative phenotypes. For mutualisms between hosts and symbionts, most mutualism theory assumes that exploiters always impose fitness costs on their host. We exposed legume hosts to mutualistic (nitrogen-fixing) and exploitative (non-nitrogen-fixing) symbiotic rhizobia in field conditions, and manipulated the presence or absence of insect herbivory to determine if the costly fitness effects of exploitative rhizobia are context-dependent. Exploitative rhizobia predictably reduced host fitness when herbivores were excluded. However, insects caused greater damage on hosts associating with mutualistic rhizobia, as a consequence of feeding preferences related to leaf nitrogen content, resulting in the elimination of fitness costs imposed on hosts by exploitative rhizobia. Our experiment shows that herbivory is potentially an important factor in influencing the evolutionary dynamic between legumes and rhizobia. Partner choice and host sanctioning are theoretically predicted to stabilize mutualisms by reducing the frequency of exploitative symbionts. We argue that herbivore pressure may actually weaken selection on choice and sanction mechanisms, thus providing one explanation of why host-based discrimination mechanisms may not be completely effective in eliminating nonbeneficial partners.

Consistently high heat tolerance acclimation in response to a simulated heatwave across species from the broadly distributed Acacia genus.

When leaves exceed their thermal threshold during heatwaves, irreversible damage to the leaf can accumulate. However, few studies have explored short-term acclimation of leaves to heatwaves that could help plants to prevent heat damage with increasing heatwave intensity. Here, we studied the heat tolerance of PSII (PHT) in response to a heatwave in Acacia species from across a strong environmental gradient in Australia. We compared PHT metrics derived from temperature-dependent chlorophyll fluorescence response curves (T-F 0 ) before and during a 4-day 38°C heatwave in a controlled glasshouse experiment. We found that the 15 Acacia species displayed surprisingly large and consistent PHT acclimation responses with a mean tolerance increase of 12°C (range, 7.7-19.1°C). Despite species originating from diverse climatic regions, neither maximum temperature of the warmest month nor mean annual precipitation at origin were clear predictors of PHT. To our knowledge, these are some of the largest measured acclimation responses of PHT from a controlled heatwave experiment. This remarkable capacity could partially explain why this genus has become more diverse and common as the Australian continent became more arid and suggests that the presence of Acacia in Australian ecosystems will remain ubiquitous with climate change.

Environmental stress leads to genome streamlining in a widely distributed species of soil bacteria.

Bacteria have highly flexible pangenomes, which are thought to facilitate evolutionary responses to environmental change, but the impacts of environmental stress on pangenome evolution remain unclear. Using a landscape pangenomics approach, I demonstrate that environmental stress leads to consistent, continuous reduction in genome content along four environmental stress gradients (acidity, aridity, heat, salinity) in naturally occurring populations of Bradyrhizobium diazoefficiens (widespread soil-dwelling plant mutualists). Using gene-level network and duplication functional traits to predict accessory gene distributions across environments, genes predicted to be superfluous are more likely lost in high stress, while genes with multi-functional roles are more likely retained. Genes with higher probabilities of being lost with stress contain significantly higher proportions of codons under strong purifying and positive selection. Gene loss is widespread across the entire genome, with high gene-retention hotspots in close spatial proximity to core genes, suggesting Bradyrhizobium has evolved to cluster essential-function genes (accessory genes with multifunctional roles and core genes) in discrete genomic regions, which may stabilise viability during genomic decay. In conclusion, pangenome evolution through genome streamlining are important evolutionary responses to environmental change. This raises questions about impacts of genome streamlining on the adaptive capacity of bacterial populations facing rapid environmental change.

Four Complete Genome Sequences for Bradyrhizobium sp. Strains Isolated from an Endemic Australian Acacia Legume Reveal Structural Variation.

Bradyrhizobium sp. strains were isolated from root nodules of the Australian legume, Acacia acuminata (Fabaceae). Here, we report the complete genome sequences of four strains using a hybrid long- and short-read assembly approach. The genome sizes range between ∼7.1 Mbp and ∼8.1 Mbp, each with one single circular chromosome. Whole-genome alignments show extensive structural rearrangement.

Symbiosis limits establishment of legumes outside their native range at a global scale.

Microbial symbiosis is integral to plant growth and reproduction, but its contribution to global patterns of plant distribution is unknown. Legumes (Fabaceae) are a diverse and widely distributed plant family largely dependent on symbiosis with nitrogen-fixing rhizobia, which are acquired from soil after germination. This dependency is predicted to limit establishment in new geographic areas, owing to a disruption of compatible host-symbiont associations. Here we compare non-native establishment patterns of symbiotic and non-symbiotic legumes across over 3,500 species, covering multiple independent gains and losses of rhizobial symbiosis. We find that symbiotic legume species have spread to fewer non-native regions compared to non-symbiotic legumes, providing strong support for the hypothesis that lack of suitable symbionts or environmental conditions required for effective nitrogen-fixation are driving these global introduction patterns. These results highlight the importance of mutualisms in predicting non-native species establishment and the potential impacts of microbial biogeography on global plant distributions.

A composite guanyl thiourea (GTU), dicyandiamide (DCD) inhibitor improves the efficacy of nitrification inhibition in soil.

This study investigated whether applying dicyandiamide (DCD) and guanyl thiourea (GTU) in conjunction with urea improves the efficacy of nitrification inhibition relative to traditional fertiliser application of urea or urea + DCD. Urea at a rate of 100 mg N kg(-1) soil was applied to soil microcosms (high nutrient tenosol and low nutrient hydrosol) which were treated with either no inhibitor (urea-only); 15 mg DCD kg(-1) soil or 15 mg DCD kg(-1) soil plus 21 mg GTU kg soil(-1). Mineral N (NH4(+) & NO3(-)) concentrations, potential nitrification rates (PNR) and abundances of ammonia oxidising bacteria (AOB) were measured over time. After 100-days incubation, ∼73 mg N kg(-1) soil was found as NH4(+) when urea + DCD + GTU were applied to the tenosol. NH4(+) concentrations were lower (11-32 mg N kg(-1) soil) when urea or urea + DCD were applied. This suggests that the application of GTU in conjunction with DCD elongated the effects of nitrification inhibition. In both soils, PNRs were faster and AOB abundances (gene copies g(-1) soil) were higher when urea was applied without nitrification inhibitors. There were, however, no differences in PNR or AOB abundances in either soil type when 'urea + DCD' or 'urea + DCD + GTU' were applied. The results indicate that the application of GTU with DCD may extend nitrification inhibition in certain soil types. This finding has the potential to improve the efficacy of commercially available and widely used inhibitors such as DCD.

Short-term fertilizer application alters phenotypic traits of symbiotic nitrogen fixing bacteria.

Fertilizer application is a common anthropogenic alteration to terrestrial systems. Increased nutrient input can impact soil microbial diversity or function directly through altered soil environments, or indirectly through plant-microbe feedbacks, with potentially important effects on ecologically-important plant-associated mutualists. We investigated the impacts of plant fertilizer, containing all common macro and micronutrients on symbiotic nitrogen-fixing bacteria (rhizobia), a group of bacteria that are important for plant productivity and ecosystem function. We collected rhizobia nodule isolates from natural field soil that was treated with slow-release plant fertilizer over a single growing season and compared phenotypic traits related to free-living growth and host partner quality in these isolates to those of rhizobia from unfertilized soils. Through a series of single inoculation assays in controlled glasshouse conditions, we found that isolates from fertilized field soil provided legume hosts with higher mutualistic benefits. Through growth assays on media containing variable plant fertilizer concentrations, we found that plant fertilizer was generally beneficial for rhizobia growth. Rhizobia isolated from fertilized field soil had higher growth rates in the presence of plant fertilizer compared to isolates from unfertilized field soil, indicating that plant fertilizer application favoured rhizobia isolates with higher abilities to utilize fertilizer for free-living growth. We found a positive correlation between growth responses to fertilizer and mutualism benefits among isolates from fertilized field soil, demonstrating that variable plant fertilizer induces context-dependent genetic correlations, potentially changing the evolutionary trajectory of either trait through increased trait dependencies. Our study shows that short-term application is sufficient to alter the composition of rhizobia isolates in the population or community, either directly though changes in the soil chemistry or indirectly through altered host legume feedbacks, and is potentially a strong selective agent acting on natural rhizobia populations.

Reduced plant competition among kin can be explained by Jensen's inequality.

Plants often compete with closely related individuals due to limited dispersal, leading to two commonly invoked predictions on competitive outcomes. Kin selection, from evolutionary theory, predicts that competition between relatives will likely be weaker. The niche partitioning hypothesis, from ecological theory, predicts that competition between close relatives will likely be stronger. We tested for evidence consistent with either of these predictions by growing an annual legume in kin and nonkin groups in the greenhouse. We grew plant groups in treatments of symbiotic nitrogen fixing bacteria differing in strain identity and composition to determine if differences in the microbial environment can facilitate or obscure plant competition patterns consistent with kin selection or niche partitioning. Nonkin groups had lower fitness than expected, based on fitness estimates of the same genotypes grown among kin. Higher fitness among kin groups was observed in mixtures of N-fixing bacteria strains compared to single inoculations of bacteria strains present in the soil, which increased fitness differences between kin and nonkin groups. Lower fitness in nonkin groups was likely caused by increased competitive asymmetry in nonkin groups due to genetic differences in plant size combined with saturating relationships with plant size and fitness- i.e. Jensen's inequality. Our study suggests that microbial soil symbionts alter competitive dynamics among kin and nonkin. Our study also suggests that kin groups can have higher fitness, as predicted by kin selection theory, through a commonly heritable trait (plant size), without requiring kin recognition mechanisms.

Estimating uncertainty in multivariate responses to selection.

Predicting the responses to natural selection is one of the key goals of evolutionary biology. Two of the challenges in fulfilling this goal have been the realization that many estimates of natural selection might be highly biased by environmentally induced covariances between traits and fitness, and that many estimated responses to selection do not incorporate or report uncertainty in the estimates. Here we describe the application of a framework that blends the merits of the Robertson-Price Identity approach and the multivariate breeder's equation to address these challenges. The approach allows genetic covariance matrices, selection differentials, selection gradients, and responses to selection to be estimated without environmentally induced bias, direct and indirect selection and responses to selection to be distinguished, and if implemented in a Bayesian-MCMC framework, statistically robust estimates of uncertainty on all of these parameters to be made. We illustrate our approach with a worked example of previously published data. More generally, we suggest that applying both the Robertson-Price Identity and the multivariate breeder's equation will facilitate hypothesis testing about natural selection, genetic constraints, and evolutionary responses.