(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Diverse signatures of convergent evolution in cactus-associated yeasts [1]

['Carla Gonçalves', 'Department Of Biological Sciences', 'Vanderbilt University', 'Nashville', 'Tennessee', 'United States Of America', 'Evolutionary Studies Initiative', 'Associate Laboratory Institute For Health', 'Bioeconomy', 'Ucibio Applied Molecular Biosciences Unit']

Date: 2024-10

Many distantly related organisms have convergently evolved traits and lifestyles that enable them to live in similar ecological environments. However, the extent of phenotypic convergence evolving through the same or distinct genetic trajectories remains an open question. Here, we leverage a comprehensive dataset of genomic and phenotypic data from 1,049 yeast species in the subphylum Saccharomycotina (Kingdom Fungi, Phylum Ascomycota) to explore signatures of convergent evolution in cactophilic yeasts, ecological specialists associated with cacti. We inferred that the ecological association of yeasts with cacti arose independently approximately 17 times. Using a machine learning–based approach, we further found that cactophily can be predicted with 76% accuracy from both functional genomic and phenotypic data. The most informative feature for predicting cactophily was thermotolerance, which we found to be likely associated with altered evolutionary rates of genes impacting the cell envelope in several cactophilic lineages. We also identified horizontal gene transfer and duplication events of plant cell wall–degrading enzymes in distantly related cactophilic clades, suggesting that putatively adaptive traits evolved independently through disparate molecular mechanisms. Notably, we found that multiple cactophilic species and their close relatives have been reported as emerging human opportunistic pathogens, suggesting that the cactophilic lifestyle—and perhaps more generally lifestyles favoring thermotolerance—might preadapt yeasts to cause human disease. This work underscores the potential of a multifaceted approach involving high-throughput genomic and phenotypic data to shed light onto ecological adaptation and highlights how convergent evolution to wild environments could facilitate the transition to human pathogenicity.

Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: JLS is a scientific advisor for WittGen Biotechnologies. JLS is an advisor for ForensisGroup Inc. AR is a scientific consultant for LifeMine Therapeutics, Inc. All other authors have declared that no competing interests exist.

Funding: This work was supported by the National Science Foundation (Grants DEB-2110403 to CTH and DEB-2110404 to AR). Research was also supported by DOE Great Lakes Bioenergy Research Center, funded by BER Office of Science (Grant DE-SC0018409 to CTH), USDA National Institute of Food and Agriculture (Hatch Projects 1020204 and 7005101, to CTH), an H. I. Romnes Faculty Fellowship, supported by the Office of the Vice Chancellor for Research and Graduate Education with funding from the Wisconsin Alumni Research Foundation (to CTH). Research was also supported by the National Institutes of Health/National Institute of Allergy and Infectious Diseases Grant (R01AI153356 to AR) and the Burroughs Wellcome Fund (to AR). Research was also supported by the National Key R&D Program of China (Grant 2022YFD1401600 to XXS), the National Science Foundation for Distinguished Young Scholars of Zhejiang Province (Grant LR23C140001 to XXS), and the Fundamental Research Funds for the Central Universities (Grant 226-2023-00021 to XXS). Research was partially supported by the National Institutes of Health (Grant T32HG002760-16 to JFW) and a National Science Foundation Grant Postdoctoral Research Fellowship in Biology (1907278 to JFW). Research was also supported by Fundação para a Ciência e a Tecnologia, in the scope of the project UIDP/04378/2020 and UIDB/04378/2020 of the Research Unit on Applied Molecular Biosciences - UCIBIO and the project LA/P/0140/2020 of the Associate Laboratory Institute for Health and Bioeconomy - i4HB, and grants PTDC/BIA-EVL/0604/2021 (to CG) and the Federation of European Microbiological Societies (FEMS grant FEMS-GO-2019-537 to CG). JLS is a Howard Hughes Medical Institute Awardee of the Life Sciences Research Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2024 Gonçalves et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Benefiting from the wealth of genomic and phenotypic data available for nearly all known yeast species described in the subphylum Saccharomycotina [ 51 ] and cross-referencing with the ecological data available from the cactus-yeast-Drosophila system [ 35 – 37 , 52 – 60 ], we employed a high-throughput framework to detect signatures of convergent evolution in 17 independently evolved lineages of cactophilic yeasts. Using a machine learning algorithm, we uncovered distinctive phenotypic traits enriched among cactus-associated yeasts, including the ability to grow at high (≥37°C) temperatures. We found that thermotolerance might be related to distinctive rates of evolution in functions impacting the integrity of the cell envelope, some of which are under positive selection in distantly related cactophilic clades. Gene family analyses identified gene duplication and HGT events involving plant cell wall–degrading enzymes in distinct clades, suggesting adaptations associated with feeding on plant material. These results reveal that convergence to cactophily by distinct lineages of Saccharomycotina yeasts was accomplished through diverse evolutionary mechanisms acting on distinct genes, although some of these are involved in similar biological functions. Interestingly, we found that several cactus-associated yeasts and close relatives have been reported as emerging opportunistic human pathogens raising the hypothesis that fungi inhabiting certain wild environments may be preadapted for opportunistic pathogenicity. More broadly, we advocate for a methodological framework that couples diverse lines of genomic, phenotypic, and ecological data with multiple analytical approaches to investigate the plurality of evolutionary mechanisms underlying ecological adaptation.

Contrasting with Drosophila, where cactophily is largely found within the monophyletic D. repleta group [ 41 ], molecular phylogenetic analyses revealed that cactophilic yeasts belong to phylogenetically distinct clades, indicating that association with cacti evolved multiple times, independently, in the Saccharomycotina [ 37 ]. While relevant ecological and physiological information of cactus-associated yeasts is available [ 35 – 37 , 39 ], the genetic changes that facilitated the convergent evolution of multiple yeast lineages to the cacti environment are unknown.

In Drosophila, the adoption of cacti as breeding and feeding sites evolved around 16 to 21 million years ago (Mya) and is considered one of the most extensive and successful ecological transitions within the genus [ 41 ]. Cactophilic Drosophila thrive across a wide range of cacti species that differ in the profiles of toxic metabolites they produce—Opuntia species, commonly called prickly pear cactus, generally contain fewer toxic metabolites than columnar cacti species [ 40 ] and are likely the ancestral hosts [ 41 ]. The distinctive characteristics of the cacti environment [ 42 ] seemingly selected for adaptive traits across cactophilic Drosophila, such as high resistance to heat, desiccation, and toxic alkaloid compounds produced by certain types of cacti [ 43 – 48 ]. These traits are likely associated with several genomic signatures (for instance, positive selection, gene duplications, HGT) impacting multiple functions, such as water preservation or detoxification [ 45 – 50 ].

Saccharomycotina yeasts are ecologically diverse, occupy diverse ecosystems [ 31 ], and vary considerably in their degree of ecological specialization ranging from cosmopolitan generalists to ecological specialists. For instance, Sugiyamaella yeasts are mostly isolated from insects [ 32 ] and most Tortispora species have been almost exclusively found in association with cacti plants [ 33 ]. The cactus environment accommodates numerous yeast species rarely found in other niches [ 34 – 37 ]. Moreover, cactophilic yeasts are part of a model ecological system involving the tripartite relationship between cactus, yeast, and Drosophila [ 34 , 36 , 38 , 39 ]. Cactophilic yeasts use necrotic tissues of cacti as substrates for growth [ 40 ] while serving as a food source to cactophilic Drosophila. Cactophilic flies (and other insects) play, in turn, a crucial role in the yeast’s life cycle by acting as vectors [ 34 ].

Fungi exhibit very high levels of evolutionary sequence divergence [ 25 ]; the amino acid sequence divergence between the baker’s yeast Saccharomyces cerevisiae and the human commensal and opportunistic pathogen Candida albicans, both members of subphylum Saccharomycotina (one of the 3 subphyla in Ascomycota, which is one of the more than 1 dozen fungal phyla), is comparable to the divergence between humans and sponges [ 26 ]. Due to their very diverse genetic makeups, convergent phenotypes arising in fungi might involve distinct genetic determinants and/or mechanisms, including HGT [ 27 , 28 ], a far less common mechanism among animals (but see [ 29 , 30 ]).

Independently evolved phenotypes often share the same genetic underpinnings (parallel molecular evolution) [ 4 – 7 ], such as similar mutations in specific genes [ 4 , 5 , 8 ], but can also arise through distinct molecular paths and by distinct evolutionary mechanisms [ 9 , 10 ], such as gene duplications [ 11 – 13 ], gene losses [ 14 – 16 ], or horizontal gene transfer (HGT) events [ 17 , 18 ]. Molecular signatures of convergence can also be inferred from independent shifts in overall evolutionary rates [ 19 ] and examined at higher hierarchical levels of molecular organization, such as functions or pathways [ 20 ]. For instance, comparing rates of evolution across distantly related animal lineages could pinpoint convergent slowly evolving genes involved in adaptive functions [ 21 ] or convergent rapidly evolving genes indicating parallel relaxed constraints acting on dispensable functions [ 22 ]. Parallel molecular changes are common across all domains of life [ 9 ], but their occurrence can be reduced by mutational epistasis or the polygenic nature of some phenotypic traits [ 10 , 23 , 24 ], particularly when studying convergence in distantly related organisms.

Convergent evolution, the repeated evolution of similar traits among distantly related taxa, is ubiquitous in nature and has been documented across all domains of life [ 1 – 3 ]. Convergence typically arises when organisms occupy similar ecological niches or encounter similar conditions and selective pressures; facing similar selective pressures, organisms from distinct lineages often evolve similar adaptations.

Results

Yeast cactophily likely evolved independently 17 times We examined the ecological association of yeasts with the cacti environment across a dataset of 1,154 strains from 1,049 yeast species. Yeast-cacti associations vary substantially in their strengths [35]. Some cactus-associated yeast species are considered cosmopolitan, being commonly isolated from cacti but also other environments (henceforth referred to as transient), whereas others are strictly cactophilic, defined as those almost exclusively isolated from cacti (S1 Table; note that this classification is based on the available ecological information, which may be impacted by sampling bias and other sampling issues—it is possible that strictly cactophilic species could also be found in other, yet unsampled, environments). We observed that strictly cactophilic species are found across almost the entire Saccharomycotina subphylum spanning from the Trigonopsidales (i.e., Tortispora spp.) [33] to the Saccharomycetales (i.e., Kluyveromyces starmeri) [52] (Figs 1 and S1). Using the yeast phylogeny and distribution of (strict) cactophily in an ancestral state reconstruction, we inferred a total of 17 origins for the evolution of cactophily (S2 Fig). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. Yeast cactophily originated repeatedly and at different times. (top) Genome-wide–based phylogeny of the subphylum Saccharomycotina [51] depicting the different types of ecological association of strictly cactophilic yeast species with the cacti environment: (necrotic) cacti tissues, which include both Opuntia spp. and columnar cacti, cacti flowers, and cacti-visiting insects. Cactophilic lineages and well-known yeasts, such as Saccharomyces spp. and Candida albicans, are noted on the phylogeny. (a, b, and c) Subtrees of 3 cactophilic clades: (a) Tortispora, (b) Starmera, and (c) Pichia. Estimated times of origin, as determined in [51], for the emergence of cactophily are represented for these 3 example clades. The data underlying this Figure can be found in https://doi.org/10.6084/m9.figshare.24114381. https://doi.org/10.1371/journal.pbio.3002832.g001 Cactophily is found in single species belonging to different orders, but it also involves nearly an entire genus (i.e., Tortispora). Specifically, 7 of the 17 instances of cactophily evolution involve clades containing 2 or more species while the remaining 10 involve single species, suggesting that different taxa evolved this ecological association at different times (Figs 1 and S1). For instance, cross-referencing relaxed molecular clock analyses of the yeast phylogeny [51] with ancestral state reconstructions suggests that cactophily in Tortispora likely emerged twice, once in the most recent common ancestor (MRCA) of T. starmeri/T. phaffi around 47 Mya, and in the MRCA of T. caseinolytica/T. mauiana/T. ganteri around 11 Mya (Fig 1). An alternative hypothesis would place the emergence of cactophily in the MRCA of the Tortispora genus around 180 Mya; however, this hypothesis is inconsistent with the estimated origin of the Cactaceae family (35 Mya) [61]. Cactophily in the genus Starmera and in the Pichia cactophila clade emerged more recently, most likely around 12 and 3 Mya, respectively (Fig 1). We also observed that cactus-associated yeasts seemingly exhibit significant niche partitioning (S1 Table). For example, T. ganteri is typically isolated from columnar cacti, while its close relative T. caseinolytica is more commonly found in Opuntia spp. [33]. Furthermore, P. cactophila is considered a generalist cactophilic yeast, being widely distributed across a wide range of cacti species [37], while closely related P. heedii has been predominantly found in association with certain species of columnar cacti [39]. However, many species (for instance, T. starmeri or P. insulana) alternate between the 2 types of cacti [33,58], similar to some Drosophila species [41]. Other strictly cactophilic species, such as Wickerhamiella cacticola or Kodamaea nitidulidarum, are associated with cacti flowers and/or flower-visiting insects, like beetles, and not with necrotic cacti tissues [59,60,62].

Detecting signatures of convergent evolution in cactophilic yeasts We envision 3 distinct scenarios that may capture how different yeast lineages convergently evolved cactophily (Fig 2A): Scenario I: Convergent phenotypes and genotypes Selective pressures associated with the cacti environment (for instance, high temperature, desiccation, or presence of toxic compounds) favor similar phenotypic traits that evolved through the same genomic mechanisms; Scenario II: Convergent phenotypes through divergent genotypes Selective pressures associated with the cacti environment favor similar phenotypic traits across cactophilic species, but different evolutionary mechanisms (for instance, gene duplication, HGT) and/or genes contribute to phenotypic convergence of different lineages. In this scenario, similar phenotypes emerge through distinct evolutionary trajectories; Scenario III: Divergent phenotypes and genotypes Distinct phenotypic landscapes are explored by distinct clades when thriving in the same environment (niche partitioning); different evolutionary mechanisms contribute to these phenotypes. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Alternative scenarios for the evolution of yeast cactophily and methodological framework employed in this study. (A) Methodological framework for investigating signatures of convergent evolution of yeast cactophily. Association with cacti evolved multiple times in yeasts from distinct genetic backgrounds, which are represented by distinct shades of gray. Convergent adaptation to the cacti environment might have involved convergence at both phenotypic and genomic levels (Scenario I), phenotypic convergence through distinct molecular paths (Scenario II), or lack of convergence at both phenotypic and genotypic levels (Scenario III). Each of these scenarios is tested using a methodological framework (B) involving (a) machine learning approach wherein a model is trained to distinguish cactophilic from non-cactophilic yeasts from genomic and phenotypic features; (b) gene family analyses to find evidence of gene duplications, losses, and HGT that might have occurred in cactophilic yeasts; and (c) evaluation of signatures of positive selection and of changes in relative evolutionary rates (evolving faster or slower) in branches leading to cactophilic clades. https://doi.org/10.1371/journal.pbio.3002832.g002 To explore which scenario(s) best reflect(s) the process of yeast adaptation to the cacti environment, we developed a framework for identifying signatures of adaptation and convergence from high-throughput genomic and phenotypic data [51] (Fig 2B). First, we employed a random forest (RF) classifier to identify phenotypic and genetic commonalities that distinguish cactophilic from non-cactophilic yeasts. We would expect that our RF classifier would yield highly accurate predictions for Scenario I, intermediate accuracy for Scenario II, and lack of accuracy for Scenario III. For instance, in Scenario I, where association with cacti would involve the evolution of similar phenotypes encoded by the same genomic paths, we expect that the accuracy of prediction obtained would be near 100% as both genomic and phenotypic features would be shared by all cactophilic species. In contrast, Scenario III, which implies that the distinct cactophilic lineages experienced distinct changes and display distinct phenotypes, we would expect that our accuracy of prediction would be close to random (i.e., 50%) because no genetic or phenotypic feature would be predictive of cactophily. Second, we inspected patterns of gene presence/absence due to gene duplication, and HGT. Third, we investigated genome-wide evolutionary rates to detect signatures of convergence in evolutionary rates and of positive selection in individual genes (Fig 2B), which have been also frequently implicated in adaptive evolution [5,19,21,22,63]. Specifically, for the detection of convergent evolutionary rates, we adopted an approach that identifies genes with evolutionary rate (i.e., number of amino acid substitutions per site) shifts across a phylogeny including multiple cactophilic and non-cactophilic species and correlates these shifts with the independent emergence of cacti association [64]. The analyses of gene family and evolutionary rates allow us to identify which genomic features (genes) and mechanisms (duplication, HGT, or altered evolutionary rates) may be associated with the common phenotypic features identified in the machine learning analyses. Under Scenario I, we would expect to find similar mechanisms (positive selection, rapid/slow evolutionary rates, HGT, or duplication) impacting the same phenotype across cactophilic clades. Under Scenario II, we would expect to find distinct genetic mechanisms/genes affecting the same phenotypes across cactophilic clades. Under Scenario III, we would expect an absence of shared mechanisms, genes, and phenotypes across clades. We applied this methodological framework to study convergence in ecological specialization in yeasts but note that it can be applied more generally to study the process of convergent or adaptive evolution.

HGT and duplication of cell wall–degrading enzymes in cactophilic yeasts We next looked for genes that might be implicated in cactophily by examining gene duplication and HGT across 3 groups that contained 2 or more cactophilic lineages. We constructed 3 distinct datasets (S3 Table) containing multiple cactophilic species and closest non-cactophilic relatives within the Lipomycetales/Dipodascales/Trigonopsidales orders (referred to as LDT group, including Tortispora spp., Dipodascus australiensis, Magnusiomyces starmeri, Myxozyma mucilagina, and Myxozyma neglecta), Phaffomycetales (including Starmera spp. and Phaffomyces spp.), and Pichiales (including 2 distinct Pichia spp. cactophilic clades). We focused on gene duplications and HGT, as gene losses are usually not reliably estimated due to annotation and sampling issues or inaccurate gene family clustering [71]. By inspecting orthogroups that uniquely contained cactophilic species, we found a gene encoding a pectate/pectin lyase was uniquely found in P. eremophila (strictly cactophilic) and P. kluyveri (transient), which are commonly isolated from rotting cacti tissues. Sequence similarity searches across the entire dataset of 1,154 yeast genomes confirmed that this gene is absent from all other species. Pectate lyases are extracellular enzymes involved in pectin hydrolysis and plant cell wall degradation. Consistent with this function, these enzymes are mostly found among plant pathogens and plant-associated fungi and bacteria [72,73], and their activity has only been reported in a handful of Saccharomycotina species [74]. Phylogenetic analyses showed that the 2 yeast sequences are nested deeply within a clade of bacterial pectin lyases (Fig 4A). The most closely related sequence belongs to Acinetobacter boissieri [75], which has been frequently isolated from plants and flowers, and to Xanthomonas and Dickeya, 2 genera of plant pathogenic bacteria [76,77]. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Duplication and horizontal gene transfer of plant cell wall-degrading enzymes in cactophilic species. (A) Phylogenetic tree of closest related sequences to pectin lyases from P. eremophila and P. kluyveri. (on the right) Pruned tree highlighting the ecological association of bacteria species harboring the closest related pectin lyase sequences to P. eremophila/P. kluyveri proteins. Prediction of subcellular localization [78,79] is shown in the panel below. (B) Distribution of rhamnogalacturonan endolyases across the 1,154 yeast genomes (presence in black and absence in gray). (on the right) Pruned phylogenetic tree of yeast rhamnogalacturonan endolyase and closest relatives, highlighting the duplication events in cactophilic Phaffomyces species. A BLASTp search against the yeast dataset of 1,154 proteomes was performed, and all significant hits were retrieved (e-value cutoff e−3). Phylogeny was constructed in IQ-TREE v2.0.6 (-m TEST, -bb 1,000) [80,81]. Branch support (bootstrap > = 90) is represented as black circles. Branches are colored according to taxonomy as indicated in the key. Branches clustering other putative rhamnogalacturonan endolyase sequences in the Saccharomycotina (total of 33) were collapsed. In P. opuntiae, there are 2 additional partial sequences (g001496.m1 and g001773.m1, 163 amino acids) that only partially overlap (from 39 to 163 overlapping amino acids) with the remaining nearly complete sequences (g002229.m1: 468 amino acids and g001049.m1: 610 amino acids). Prediction of subcellular localization according to SignalP and Deeploc [78,79] is shown in the panel below. The data underlying this Figure can be found in https://doi.org/10.6084/m9.figshare.24114381. https://doi.org/10.1371/journal.pbio.3002832.g004 Using gene tree–species tree reconciliation analyses implemented in GeneRax [71], we next examined genes with evidence of duplication in at least 1 cactophilic species belonging to each group, while excluding events of duplication in non-cactophilic species belonging to each of the 3 lineages/groups inspected (S4 Table). Duplication of another gene involved in plant cell wall degradation, encoding a rhamnogalacturonan endolyase (K18195), was detected in the cactophilic P. antillensis, P. opuntiae, and Candida coquimbonensis (Phaffomycetales; S4 Table). These species contained 2 copies of this gene compared to their closest relatives, which contained only one (Fig 4B). These enzymes are responsible for the extracellular cleavage of pectin [82], which, along with cellulose, is one of the major components of plant cell wall. Pectin lyase [74] activities have been rarely reported among yeasts; therefore, we assessed the distribution of the rhamnogalacturonan endolyase across the 1,154 proteomes using a BLASTp search (e-value cutoff e−3) and found that it displayed a patchy distribution, being found in fewer than 60 species (Fig 4B). Consistent with their function, HGT-derived pectin lyases and rhamnogalacturonan endolyases were predicted to localize to the extracellular space based on primary sequence analyses [78] (Fig 4). Pectin lyase enzymatic activity was previously detected in P. kluyveri strains associated with coffee fermentation [73,83], suggesting that the identified HGT-derived pectin lyase is likely responsible for this activity.

Genes involved in maintenance of the cell envelope show evidence of codon optimization To infer the transcriptional activity of cactophilic yeasts, we determined gene-wise relative synonymous codon usage (gw-RSCU), a metric that measures biases in codon usage that have been shown to be associated with expression level [107], and examined the top-ranked genes (95th percentile) in cactophilic species (S8 Table). Top-ranked genes include many encoding ribosomal proteins and histones, which are known to be highly expressed and codon-optimized in S. cerevisiae [108]. We noticed that the chitin deacetylase gene CDA2 and genes involved in ergosterol biosynthesis (namely, ERG2, ERG5, ERG6, and ERG11) were among the genes that fell within the 95th percentile rank for gw-RSCU in multiple cactophilic species. To ascertain whether these genes also show signatures of codon optimization in closely related non-cactophilic species, we determined their respective gw-RSCU percentile ranks. While no clear pattern was observed for ERG genes (these genes were also highly ranked for gw-RSCU in non-cactophilic species), we observed that CDA2 is particularly highly ranked in Phaffomyces, Starmera, and Pichia clades compared to their closest relative non-cactophilic species (S5 Fig). CDA2 also showed evidence for positive selection in both Pichia clades and Starmera, suggesting that distinctive synonymous and/or nonsynonymous might have resulted from translational selection for optimized codons due to higher gene expression.

Convergent evolution of yeast cactophily occurred via the independent acquisition of the same phenotypic traits through mostly distinct genetic changes (Scenario II) Our results suggest that the evolution of cactophily across the Saccharomycotina are most consistent with Scenario II (phenotypic convergence associated with distinct genetic underpinnings; Fig 2). The intermediate accuracy in predicting cactophily obtained with the RF classifier, using both genomic and metabolic features together or separately, and the general overlap between the top features identified across the distinct RF runs, suggests that only some features (both genomic and metabolic) are common to all species, while others may be species-specific. It may also suggest that some common phenotypic features to most cactophilic species (such as thermotolerance) might have evolved through distinct genetic trajectories (Scenario II). This was further supported by the results obtained from the evolutionary rates analyses where we found mostly distinct genes involved in similar functions (for instance, heat stress response, ergosterol biosynthesis) under selection or accelerated evolution across cactophilic clades. Nevertheless, some exceptions where the same genes were involved (for instance, CDA2), offering support for Scenario I, did exist. Finally, we found distinct mechanisms (HGT and gene duplication) acting on distinct genes (encoding pectin degrading enzymes) involved in similar functions (plant cell wall degradation); these findings were also most consistent with Scenario II.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002832

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.