tRNAHis 5-methylcytidine levels increase in response to several growth arrest conditions in Saccharomyces cerevisiae

Abstract

tRNAs are highly modified, each with a unique set of modifications. Several reports suggest that tRNAs are hypomodified or, in some cases, hypermodified under different growth conditions and in certain cancers. We previously demonstrated that yeast strains depleted of tRNA(His) guanylyltransferase accumulate uncharged tRNA(His) lacking the G(-1) residue and subsequently accumulate additional 5-methylcytidine (m(5)C) at residues C(48) and C(50) of tRNA(His), due to the activity of the m(5)C-methyltransferase Trm4. We show here that the increase in tRNA(His) m(5)C levels does not require loss of Thg1, loss of G(-1) of tRNA(His), or cell death but is associated with growth arrest following different stress conditions. We find substantially increased tRNA(His) m(5)C levels after temperature-sensitive strains are grown at nonpermissive temperature, and after wild-type strains are grown to stationary phase, starved for required amino acids, or treated with rapamycin. We observe more modest accumulations of m(5)C in tRNA(His) after starvation for glucose and after starvation for uracil. In virtually all cases examined, the additional m(5)C on tRNA(His) occurs while cells are fully viable, and the increase is neither due to the GCN4 pathway, nor to increased Trm4 levels. Moreover, the increased m(5)C appears specific to tRNA(His), as tRNA(Val(AAC)) and tRNA(Gly(GCC)) have much reduced additional m(5)C during these growth arrest conditions, although they also have C(48) and C(50) and are capable of having increased m(5)C levels. Thus, tRNA(His) m(5)C levels are unusually responsive to yeast growth conditions, although the significance of this additional m(5)C remains unclear.

FIGURE 1.

Post-transcriptional modifications of mature tRNA^His. (A) Secondary structure of mature S. cerevisiae tRNA^His. Post-transcriptional modifications are indicated. G₋₁ is added to the 5′ end of tRNA^His by tRNA^His guanylyltransferase (Thg1). Under normal growth conditions, Trm4 methylates position 49 to form 5-methylcytidine (m⁵C, circle). When Thg1 is depleted, Trm4 also methylates adjacent cytidines at positions 48 and 50 (squares). (B) Representative HPLC traces of tRNA^His nucleosides. Wild-type (BY4741) and trm4-Δ cells were grown to log phase in SD complete media, and tRNA^His was purified, digested with P1 nuclease, phosphatase-treated, and separated by reverse-phase HPLC. Insets compare the wild type vs. trm4-Δ m⁵C peak (left), with the Am peak (right) as a control.

FIGURE 2.

Temperature-sensitive strains have additional m⁵C on tRNA^His when grown at nonpermissive temperature. (A) Assessment of viable cell titer following growth at 37°C. (B) m⁵C levels in tRNA^His and tRNA^Val(AAC) during growth at 37°C. tRNA^His and tRNA^Val(AAC) were purified from BY4741, thg1-Y146H, and fcp1-1 strains, and modification levels were measured by HPLC. m⁵C levels for tRNA^His (black) and tRNA^Val(AAC) (gray) during temperature-sensitive strain growth at 37°C were plotted. m⁵C values during growth at 24°C were plotted as the 37°C 0-h time point.

FIGURE 3.

tRNA^His m⁵C levels increase when BY4741 is grown to stationary phase. tRNA^His was purified from BY4741 grown in SD complete media in mid-log phase and in stationary phase for the indicated number of days. Nucleoside modifications were analyzed by HPLC, and m⁵C (black), Am (dark gray), and m¹G + Gm (light gray) values are plotted over time. Other control modifications (dihydrouridine and pseudouridine) were unchanged during the time course (data not shown). The increase in m⁵C levels between log phase growth and Day 1 of stationary phase is highly reproducible (Day 0 log phase, m⁵C = 0.86 ± 0.03, n = 6; Day 1 stationary phase, m⁵C = 1.53 ± 0.11, n = 7, mean ± standard deviation) (see Table 3).

FIGURE 4.

Trm4-cMORF levels do not increase during starvation. BY4741 and Trm4-cMORF strains were grown in SD complete, SD – His, or SD – Leu media and grown for 24 h at 30°C. Crude extracts were prepared at 0, 7, and 24 h, and 30 μg of protein were loaded in each lane and resolved by SDS-PAGE, followed by Western blotting. Anti-HA antibody was used to detect the cMORF tag on Trm4. BY4741 crude extract was used as a negative control. The same blot was incubated with anti-enolase antibody as a loading control.

FIGURE 5.

Analysis of tRNA^His aminoacylation under different starvation conditions. RNA from the wild-type (BY4741) strain grown in the indicated media was isolated in acidic conditions to preserve aminoacylation, and 2 μg of RNA were resolved on an acidic gel, analyzed by Northern blot for tRNA^His, tRNA^Arg(ICG), and 5S rRNA, and quantified using ImageQuant software. A control sample was treated with base (+) to detect migration of deacylated tRNA. (Solid arrows) Aminoacyl-tRNA, (dotted arrows) deacyl-tRNA. Each percent aminoacylation value is the mean ± standard deviation (SD) from two independent cultures. A representative Northern blot from one of the two independent cultures is shown.

FIGURE 6.

m⁵C increases specifically on tRNA^His during starvation. tRNA^His (black), tRNA^Val(AAC) (dark gray), and tRNA^Gly(GCC) (light gray) were purified from BY4741 grown in the indicated starvation conditions. Nucleoside modifications were analyzed by HPLC, and m⁵C values are plotted over time for each tRNA.