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



Phenotypic plasticity evolves at multiple biological levels in response to environmental predictability in a long-term experiment with a halotolerant microalga [1]

['Christelle Leung', 'Cefe', 'Université De Montpellier', 'Cnrs', 'Ephe', 'Ird', 'Université Paul Valéry Montpellier', 'Montpellier', 'Daphné Grulois', 'Leandro Quadrana']

Date: 2023-03

We analyzed 9 populations of the halotolerant microalga D. salina (strain CCAP 19/15) that have evolved under regimes of randomly fluctuating environments, with controlled and variable predictability [5,24]. During experimental evolution, lines derived from a single ancestral population were exposed to randomly fluctuating salinity, with changes every 3 or 4 days (with about one generation per day), for a total of c. 500 generations. Salinity had a normal distribution over time, with the same mean (2.4 M NaCl) and standard deviation (1 M NaCl) across treatments, but variable autocorrelation, and, hence, variable predictability [5,24]. Previous morphological analysis of 32 of these lines revealed that reduced morphological plasticity has evolved in lines that experienced less predictable environments [5]. To further characterize the molecular basis of this evolution, we set out to determine whether the plasticity of DNA methylation and gene expression levels evolved for a subset of these lines that experienced 3 different predictability treatments (3 lines per treatment): low (ρ2 = 0), intermediate (ρ2 = 0.25), and high (ρ2 = 0.81) predictability, where ρ is the stationary (long-term) temporal autocorrelation of salinity time series. All populations were subsequently subjected to a 10-day acclimation step at intermediate salinity ([NaCl] = 2.4 M), to ensure they had similar physiological states and population densities before the phenotypic and molecular plasticity assays. They were then placed for 24 h at 2 salinities near the extremes of their historical range ([NaCl] = 0.8 M and 4.0 M), to assess their degree of plasticity in DNA methylation, gene expression, and individual cell morphology (Fig 1). At any of these levels, an effect of salinity is indicative of phenotypic plasticity, an effect of the evolutionary treatment ρ2 denotes evolution, while ρ2 × salinity interaction indicates evolution of plasticity.

The experiment included 3 steps: (i) long-term experimental evolution (left); (ii) acclimation at a constant, intermediate salinity (middle); and (iii) plasticity assays at high versus low salinity (right). Each colored time series on the left represents an actual realization of salinity fluctuations for one of the populations used in this study, with the color denoting the treatment of stationary (i.e., expected long-term) temporal autocorrelation (ρ). At the end of the experiment, cells were harvested for DNA methylation, gene expression, and cell morphology analyses.

2.1 DNA methylation and gene expression plasticity evolved in response to environmental predictability

DNA methylation can contribute to phenotypic differentiation by influencing gene regulation [18, 25–28]. To assess whether experimental evolution of D. salina may lead to epigenetic differentiation, we performed whole-genome bisulfite sequencing (WGBS) for all samples, yielding a total of 1.23 × 109 150 bp paired-end raw reads, and an estimated average depth of coverage of 43.76 × (s.d. 3.14 ×) per sample (S1 Table). After data filtering, we carried out our methylation analyses on an average of 7.56 × 107 (s.d. 8.28 × 106) cytosines per samples at the CpG context (S1 Table), where methylations are most prevalent in this species [23]. Redundancy analyses (RDA) based on overall CpG methylation revealed a significant effect of evolutionary treatments ρ2 (R2 adj = 4.66%; P = 0.005) on DNA methylation (Fig 2A). Fig 2A, which is a constrained ordination that maximizes variation in DNA methylation that can be explained by salinity and environmental predictability, suggests that lines from highly predictable environments displayed higher salinity differences than those from lowly predictable environments. However, we did not detect a significant marginal effect of salinity (R2 adj = 0.79%; P = 0.356) or ρ2 × salinity interaction (R2 adj = 0.30%; P = 0.562) on the overall DNA methylation pattern, suggesting the absence of overall DNA methylation changes in response to salinity or a lack of power to detect small variation being explained.

That we did not detect significant differentiation at the whole epigenome level does not preclude the existence of more localized epigenetic differences along the genome, so we also investigated regional changes in DNA methylation at a finer scale, by considering nonoverlapping 100 bp windows, hereafter denoted differentially methylated regions (DMRs). We detected 227 DMRs (with FDR < 0.05 under Benjamini–Hochberg (BH) adjustment of P values and |diff-Methylation| > 20%) among the evolutionary treatments, as summarized by their environmental predictability ρ2 (Fig 2B). We also detected ρ2-specific DMRs between salinities within each evolutionary treatment, among the 14,357 total 100 bp regions (Fig 2B). Interestingly, populations that evolved under less predictable environmental fluctuations displayed the least number of DMRs between salinities (n = 3), as compared to populations from intermediate (n = 78) or high (n = 29) environmental predictability (Fig 2B), indicative of reduced epigenetic plasticity. We then assessed whether changes in DNA methylation patterns in response to a given environmental challenge involved similar genomic regions in the different evolved lines. Comparison of the list of DMRs between salinities revealed no overlap across evolutionary conditions (Fig 2B), suggesting that evolution of plastic epigenetic responses involved modifications of methylations in distinct genomic regions in different treatments.

We next investigated variation in gene expression, by analyzing 32,718 transcripts through RNA-sequencing (RNA-seq). We obtained 5.62 × 108 150 bp paired-end raw reads in total (S2 Table). As with DNA methylation, we detected a significant effect of the evolutionary treatments ρ2 (R2 adj = 9.79%; P < 0.001) on gene expression at the whole-transcriptome level. However, the assay salinity now explained the greatest part of variation in gene expression (R2 adj = 32.91%; P < 0.001), indicating pervasive and highly significant plasticity. The ρ2 × salinity interaction was also significant (R2 adj = 4.30%; P = 0.029) (Fig 3A), indicating evolution of transcriptional plasticity. At a more local level, analyses of differentially expressed (DE) transcripts yielded similar results: The numbers of transcripts that were significantly DE (Likelihood ratio test, FDR < 0.05 and |Log 2 FC| > 1) were highest for contrasts between salinities (n = 4,283), followed by evolutionary treatments ρ2 (n = 1,315), and, finally, ρ2 × salinity interaction (n = 837). As for DMRs, populations that evolved in less predictable environments displayed fewer DE transcripts than populations that evolved in highly predictable environments (Fig 3B, n = 2,638, 2,844, and 4,086 for ρ2 = low, intermediate, and high, respectively). However, in contrast to what we found for DNA methylation, we observed substantial overlap among DE transcripts identified between salinities for the different evolved lines (Fig 3B), indicating that the plastic response to salinity largely involved transcriptional regulation of a common pool of genes, regardless of their evolutionary trajectory. Nonetheless, we still detected some ρ2-specific DE transcripts between salinities (Fig 3B).

To confirm that differences in gene expression (and, to a lesser extent, DNA methylation) among salinities were due to plasticity, rather than resulting from putative strong selection taking place in the polymorphic populations during the short duration of the assay, we also sequenced isogenic populations—founded from a single, presumably haploid cell from the evolved populations—subjected to different salinities. Despite the genetic homogeneity of these population, we confirmed the salinity effect on both DNA methylation and gene expression (S1 Fig). Furthermore, we did not detect any significant differences in either DNA methylation (P = 0.423) or gene expression (P = 0.520) between isogenic lines and the experimental populations they originated from (S2 Fig).

We next performed Gene Ontology (GO) enrichment analysis, to assess gene functions involved in D. salina response to salinity changes and its evolution, using the functional genome annotation constructed in Leung and colleagues [23]. Transcript functional annotation analysis revealed that DE transcripts between salinities mostly involved genes associated with cellular components and molecular functions involving the chloroplast and protein transport (Fig 4). While no GO term enrichment was detected for DE transcripts among evolutionary treatments (for FDR < 0.1), we were found that ρ2 × salinity interaction effects on gene expression essentially involved gene functions associated with different metabolic processes within the GO category “biological process” (Fig 4).

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Evolution and plasticity of DNA methylation. (A) Variation of DNA methylation patterns across evolved populations. RDA plot performed on DNA methylation patterns according to evolution conditions (colors) and salinity during plasticity assay (shapes). (B) Distribution of DMRs. Venn diagram describing the number of DMRs (q-value < 0.05 and |diff-Methylation| > 20%) across evolutionary conditions (Predictability, light gray), and between salinities within each of the 3 evolutionary conditions: low (green), intermediate (blue), and high (red) predictability of environmental changes. The raw data underlying this figure are available in the Figshare repository https://doi.org/10.6084/m9.figshare.21905670. DMR, differentially methylated region; RDA, redundancy analysis. https://doi.org/10.1371/journal.pbio.3001895.g002

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Evolution and plasticity of gene expression. (A) Variation of gene expression levels across evolved populations. RDA plot performed on gene expression levels according to evolution conditions (colors) and salinity during plasticity assay (shapes). (B) Distribution of DE transcripts. Venn diagram describing the numbers of DE transcripts across evolutionary conditions (Predictability, light gray) identified by performing LRT (FDR < 0.05) as implemented in DESeq2, and between salinities (Wald test, FDR < 0.05 after BH adjustment and |log 2 FC| > 1) for 3 evolutionary conditions: low (green), intermediate (blue), and high (red) predictability of environmental changes. The raw data underlying this figure are available in the Figshare repository https://doi.org/10.6084/m9.figshare.21905670. DE, differentially expressed; LRT, likelihood-ratio test; RDA, redundancy analysis. https://doi.org/10.1371/journal.pbio.3001895.g003

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3001895

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.