Global analysis of mRNA decay in Halobacterium salinarum NRC-1 at single-gene resolution using DNA microarrays

Abstract

RNA degradation is an important factor in the regulation of gene expression. It allows organisms to quickly respond to changing environmental conditions by adapting the expression of individual genes. The stability of individual mRNAs within an organism varies considerably, contributing to differential amounts of proteins expressed. In this study we used DNA microarrays to analyze mRNA degradation in exponentially growing cultures of the extremely halophilic euryarchaeon Halobacterium salinarum NRC-1 on a global level. We determined mRNA half-lives for 1,717 open reading frames, 620 of which are part of known or predicted operons. Under the tested conditions transcript stabilities ranged from 5 min to more than 18 min, with 79% of the evaluated mRNAs showing half-lives between 8 and 12 min. The overall mean half-life was 10 min, which is considerably longer than the ones found in the other prokaryotes investigated thus far. As previously observed in Escherichia coli and Saccharomyces cerevisiae, we could not detect a significant correlation between transcript length and transcript stability, but there was a relationship between gene function and transcript stability. Genes that are known or predicted to be transcribed in operons exhibited similar mRNA half-lives. These results provide initial insights into mRNA turnover in a euryarchaeon. Moreover, our model organism, H. salinarum NRC-1, is one of just two archaea sequenced to date that are missing the core subunits of the archaeal exosome. This complex orthologous to the RNA degrading exosome of eukarya is found in all other archaeal genomes sequenced thus far.

FIG. 1.

³H-radiolabeled uridine was added to cultures in mid-exponential growth phase at time point zero. After 12 min 50 μg (□), 100 μg (▵), or 200 μg (×) of actinomycin D/ml was added, no inhibitor was added to a control culture (⋄). RNA synthesis was monitored by quantifying TCA-precipitable tritium in a scintillation counter. The arrow indicates the time point of actinomycin D addition.

FIG. 2.

(A) Half-lives for 1,717 mRNAs were evaluated by DNA microarray analysis. A total of 44% of all mRNAs in this analysis had half-lives of 8 or 9 min, and 79% had half-lives of between 8 and 12 min. Fewer than 2% of the mRNAs had half-lives of 18 min or more. The mean overall half-life was 10.1 min. (B) About 83% of the plasmid-encoded mRNAs had half-lives of 8 or 9 min (□); the mRNA half-lives of chromosomal genes were more widespread (▪). The mean half-life of chromosome-encoded mRNAs was 10.6 min, whereas the mean half-life of plasmid-encoded mRNAs was 8.7 min.

FIG. 3.

Distribution of mRNA half-lives according to functional classes in H. salinarum NRC-1 as annotated in the Halolex database. The percentages of mRNAs ranging from 0 to 40% were plotted against the half-lives ranging from 5 to ≥18 min. The numbers of genes present per function class are indicated on the right of each diagram. Panels: -, spurious ORF; AA, amino acid metabolism; CHY, conserved, hypothetical protein; CIM, central intermediary metabolism; COM, coenzyme metabolism; CP, cellular processes; EM, energy metabolism; HY, hypothetical protein; ISH, transposases and ISH-encoded proteins; LIP, lipid metabolism; MIS, miscellaneous; NOF, no function assigned to experimentally identified protein; NUM, nucleotide metabolism; REG, gene regulation; RRR, replication, repair, recombination; SIG, signal transduction; TC, transcription; TL, translation; TP, small molecule transport.

FIG. 4.

mRNA half-lives of all evaluated mRNAs in relation to their transcript length for monocistronic genes (•) and polycistronic operons (□). In case of the polycistronic messages the shown half-life is the average half-life of the individual segments of each message.

FIG. 5.

Northern blots for six mRNAs representing a wide range of mRNA half-lives. The 16S rRNA hybridizations shown below the Northern blots were used as a control of the total RNA loaded on the gels. The half-lives evaluated by Northern blot are shown in Table 1.