Case Study of the Response of N6-Methyladenine DNA Modification to Environmental Stressors in the Unicellular Eukaryote Tetrahymena thermophila

Abstract

Rediscovered as a potential epigenetic mark, N⁶-methyladenine DNA modification (6mA) was recently reported to be sensitive to environmental stressors in several multicellular eukaryotes. As 6mA distribution and function differ significantly in multicellular and unicellular organisms, whether and how 6mA in unicellular eukaryotes responds to environmental stress remains elusive. Here, we characterized the dynamic changes of 6mA under starvation in the unicellular model organism Tetrahymena thermophila. Single-molecule, real-time (SMRT) sequencing reveals that DNA 6mA levels in starved cells are significantly reduced, especially symmetric 6mA, compared to those in vegetatively growing cells. Despite a global 6mA reduction, the fraction of asymmetric 6mA with a high methylation level was increased, which might be the driving force for stronger nucleosome positioning in starved cells. Starvation affects expression of many metabolism-related genes, the expression level change of which is associated with the amount of 6mA change, thereby linking 6mA with global transcription and starvation adaptation. The reduction of symmetric 6mA and the increase of asymmetric 6mA coincide with the downregulation of AMT1 and upregulation of AMT2 and AMT5, which are supposedly the MT-A70 methyltransferases required for symmetric and asymmetric 6mA, respectively. These results demonstrated that a regulated 6mA response to environmental cues is evolutionarily conserved in eukaryotes. IMPORTANCE Increasing evidence indicated that 6mA could respond to environmental stressors in multicellular eukaryotes. As 6mA distribution and function differ significantly in multicellular and unicellular organisms, whether and how 6mA in unicellular eukaryotes responds to environmental stress remains elusive. In the present work, we characterized the dynamic changes of 6mA under starvation in the unicellular model organism Tetrahymena thermophila. Our results provide insights into how Tetrahymena fine-tunes its 6mA level and composition upon starvation, suggesting that a regulated 6mA response to environmental cues is evolutionarily conserved in eukaryotes.

Keywords: 6mA; Tetrahymena thermophila; starvation; unicellular eukaryote.

FIG 1

DNA 6mA level was dramatically reduced during starvation. (A) IF staining of DNA 6mA in logarithmically vegetative (Veg) and starved cells. S3 to S72 represent 3 to 72 h after starvation. (B) Statistical analysis of 6mA IF signal intensity in panel A. Cell images were processed by ImageJ (Veg, n = 112; S3, n = 111; S6, n = 111; S12, n = 105; S24, n = 112; S48, n = 105; S72, n = 109). Data are presented as box plots (from top to bottom: maximum, first quartile, median, third quartile, and minimum). Student's t test was performed (***, P < 0.001; **, P < 0.01). Note that there was no significant difference in cells starved for 12 to 72 h.

FIG 2

The genome-wide distribution of 6mA was affected in starved cells. (A) 6mA density (6mA/A) is dramatically reduced in starved cells (orange) compared to vegetative cells (blue) across the 180 non-rDNA chromosomes. Chromosomes are arranged by length, from shortest to longest. (B) Classification of 6mA sites according to their sequence preference in vegetative (blue) and starved (orange) cells. The left and right panels represent 6mA site number and percentage (a particular class of 6mA/all 6mA), respectively. See Table 1 for details. (C) 6mA distribution across the shortest (left) and longest (right) chromosomes. The height of lines represents the 6mA density (percent). Note that the 6mA density of the shortest chromosome is much greater than that of the longest chromosome. (D) Overall behaviors of individual 6mA sites in vegetative and starved cells, showing results from SMRT sequencing data. (E) The methylation status change of selected conversion sites (C1 to C4; symmetric in Veg cells and unmethylated in S24 cells) and retain site (R1; symmetric in both Veg and S24 cells) was confirmed by DpnI/DpnII digestion and qPCR analysis. The Ct value difference (Veg − S24) of conversion sites was larger than zero, while that of the retain site was close to zero. (F) Sequence logos for 6mA (at position 0) in vegetative and starved cells.

FIG 3

Methylation level of 6mA was dramatically changed in starved cells. (A) Classification of 6mA sites according to their methylation level (L1, 0 to 20%; L2, 20 to 40%; L3, 40 to 60%; L4, 60 to 80%; L5, 80 to 100%) in vegetative and starved cells. See Table 1 for details. (B) Density plots of 6mA distribution, according to methylation levels on Watson (x axes) or Crick (y axes) strands in vegetative (top) and starved (bottom) cells. Note that in starved cells, the highly methylated symmetric 6mA is decreased while the highly methylated asymmetric 6mA is increased. (C) Venn diagram of highly methylated 6mA (L5) and asymmetric 6mA showing that highly methylated asymmetric 6mA increased in starved cells, compared to vegetative cells. (D) Statistics of methylation level of symmetric, asymmetric and non-ApT 6mA. Only the number of the L5 asymmetric 6mA was doubled in starved cells compared to vegetative cells.

FIG 4

The increase/reduction of 6mA affected gene expression during starvation. (A) Composite analysis of 6mA distribution on the gene body of vegetative (blue) and starved (orange) cells. Genes were scaled to unit length and extended to each side by 1 unit length. Density was calculated as 6mA amount at a certain position/total 6mA amount. (B) The 6mA amount change (S24 − Veg) in the 1 kb downstream of TSS in highly regulated genes and starvation-responding genes presented a positive correlation with their expression level changes (log₂ fold change). (C) GBrowse snapshot of two genes clearly showing that the expression level of the gene with 6mA reduction was significantly reduced in starved cells. (D) Expression levels of putative 6mA methyltransferases at different time points during vegetative growth and starvation, shown by microarray signals (59). For vegetative cells, Ll, Lm, and Lh correspond to ∼1 × 10⁵ cells/ml, ∼3.5 × 10⁵ cells/ml, and ∼1 × 10⁶ cells/ml, respectively. For starvation, ∼2 × 10⁵ cells/ml were collected at 0, 3, 6, 9, 12, 15, and 24 h, referred to as S0, S3, S6, S9, S12, S15, and S24. (E) Expression level change of putative 6mA methyltransferases, shown by both RNA-Seq and qRT-PCR analysis.

FIG 5

Nucleosome positioning was influenced by 6mA change in starved cells. (A) Distribution profiles of 6mA (top) and nucleosome (bottom) around TSS in vegetative (blue) and starved (orange) cells. (B) Nucleosome positioning degree in vegetative (blue) and starved (orange) cells. Degrees of positioning were calculated for the +1 to +5 nucleosomes. (C) 6mA distribution relative to the nucleosome dyad in vegetative (blue) and starved cells (orange). The violin plots show the density of 6mA between neighboring nucleosome dyads, grouped by methylation levels. Note that the highly methylated 6mA was increased in starved cells. (D) Nucleosome positioning degree of genes that gained (red) or lost (green) highly methylated (L5) asymmetric 6mA in the 1 kb downstream of the TSS in starved cells. Degrees of positioning were calculated for the +1 nucleosome (left) and +1, +2, and +3 nucleosomes (right).