Skip to main content
NCBI home page
Nucleic Acids Research logoLink to Nucleic Acids Research
. 2002 Dec 15;30(24):5436–5443. doi: 10.1093/nar/gkf699

Individual gvp transcript segments in Haloferax mediterranei exhibit varying half-lives, which are differentially affected by salt concentration and growth phase

Andreas Jäger, Regina Samorski, Felicitas Pfeifer 1, Gabriele Klug a
PMCID: PMC140082  PMID: 12490712

Abstract

The mc-gvp genes for gas vesicle formation in Haloferax mediterranei are transcribed from two promoters located in front of the mc-gvpA and mc-gvpD genes. The different transcripts originating from both promoters show different abundances dependent on salt concentration in the medium and growth phase. Here we show that the half-lives of these transcripts differ significantly and that the small gvp transcripts exhibit higher stabilities than the larger gvp transcripts. While the stability of most gvp transcripts is independent of the salt concentration in the medium, the gvpA mRNA decays about twice as fast in cultures grown at 18% salt compared to cultures grown at 25% salt. The stability of the 0.45 kb transcript population derived from the 5′ part of the gvpD gene depends on the growth phase of the culture. Thus, differences in mRNA stability contribute to the salt-dependent and growth phase-dependent abundance of gvp transcripts. This implies that, like in bacteria and eukarya, mRNA processing contributes to regulated gene expression in archaea.

INTRODUCTION

The moderately halophilic archaeon Haloferax mediterranei synthesizes gas vesicles during stationary phase when grown in the presence of 17–30% (w/v) salt. Cultures grown in 15% salt medium are gas-vesicle free. The gas vesicles consist of two proteins, the hydrophobic GvpA constituting the vesicle wall and the hydrophilic GvpC located on the outer surface. Including gvpA and gvpC, 14 genes are involved in gas vesicle formation and are clustered in the chromosomal mc-vac region (1). The gvp genes are arranged as two units which are transcribed in opposite directions: gvpACNO and gvpDEFGHIJKLM. A number of various sized transcripts originate from the gvp genes, which show significantly different abundances (2) (Fig. 1). The amounts of the gvp transcripts are affected by the growth phase and also by the salt concentration in the medium (2,3). The GvpD and GvpE proteins encoded within the gvpDEFGHIJKLM unit are involved in the regulation of gvp expression. GvpE is a transcriptional activator and required for the high activity of the mcA and mcD promoter during stationary growth (2,4,5), whereas GvpD has a repressor function (6–8). However, it was not known whether growth phase and salt-dependent expression of the gvp genes are exclusively due to transcriptional regulation or whether these parameters also affect the stability of gvp mRNAs. In bacteria, gene regulation is not restricted to regulated transcription but often occurs at the post-transcriptional level.

Figure 1.

Figure 1

Open in a new tab

Schematic map of the mc-gvp operon of H.mediterranei. The transcripts and their relative abundances are indicated by arrows. The transcripts are drawn as predicted by size on northern blots as described by Röder and Pfeifer (2). The designation for the transcripts is given and whenever the estimated transcript size implies that the 3′ end is located within a reading frame this is indicated by an apostrophe behind the last gene. The start and stop codons for gvpJ/gvpI, gvpI/gvpH, gvpH/gvpG and gvpC/gvpN overlap. The intergenic region between gvpN and gvpO comprises 24 nt, the intergenic region between gvpC and gvpA 76 nt. All other intergenic regions comprise 1–6 nt. The DNA fragments used as probes in northern blot hybridization are indicated below the genes and the transcriptional starts are indicated by arrows. gvpA and gvpC encode the structural proteins of gas vesicles.

Bacterial polycistronic mRNAs are frequently processed into mRNA segments with varying stabilities, allowing differential expression of genes transcribed from the same promoter (reviewed in 9,10). External factors like oxygen tension, temperature or nutrient availability can affect the stability of transcripts or certain mRNA segments of polycistronic transcripts (reviewed in 11). Many investigations have focused on the mechanisms of mRNA decay in bacteria. The existence of a high molecular weight protein complex with a major role in mRNA decay, the degradosome, has been detected in the γ-proteobacterium Escherichia coli (12,13) and also in the α-proteobacterium Rhodobacter capsulatus (14). The central component of this complex is the endoribonuclease RNase E which is associated with other proteins, like helicases, the 3′→5′ exonuclease polynucleotide phosphorylase, polyphosphate kinase, enolase, or transcription termination factor Rho (12–16). The compositions of the RNA degrading complexes from E.coli and R.capsulatus are different and may also vary under different growth conditions. RNase E and the degradosome can attack single stranded RNA 5′ ends harboring a monophosphate (17) and then successively cleave in an overall 5′→3′ direction. The decay of other mRNAs is initiated by internal cleavage mediated by RNase E or the degradosome at AU-rich sequences, and is followed by further 3′→5′ decay mediated by exonucleases and by further attack of the unprotected 5′ end by RNase E. Hairpin loop structures at the 5′ ends of RNAs can protect against an RNase E mediated attack and stabilize downstream mRNA segments. Internal hairpin loop structures or structures localized at the 3′ end can protect against 3′→5′ exonucleases and therefore stabilize upstream mRNA segments (reviewed in 9,10,18). Short polyadenylate tails are present in certain bacterial mRNAs (reviewed in 19), leading to decreased mRNA stability (20,21).

Most eukaryal mRNAs have methylated guanosine caps at the 5′ end and a long poly(A) tail at the 3′ end. mRNA decay in Saccharomyces cerevisiae occurs by initial deadenylation and decapping and subsequent 5′→3′ exonucleolytic degradation (reviewed in 22). Eukaryotes also harbor a high molecular weight RNA degrading machinery, the exosome (23,24). The eukaryotic exosome consists of several paralogous proteins known or predicted to possess exonuclease activity (containing RNase PH, RNase II or RNase D domains), RNA binding proteins containing the S1 domain, and more loosely associated helicases and adapter proteins (23–25).

Almost nothing is known about the mechanisms involved in the decay of mRNA in archaea. Archaeal mRNAs are not capped, some mRNAs carry short poly(A) tails (26–28). The function of several archaeal RNases has been determined. All of these RNases are involved in the processing and maturation of stable RNAs, like RNase P (29), tRNA splicing endonuclease (30,31), rRNA processing endonucleases (32) and RNase H (33). An RNase E like activity was found in extracts of Haloarcula marismortui (34) suggesting that some similarities to the mechanisms of mRNA decay in eubacteria may exist. Archaeal genome sequences do not contain any homologs to bacterial rne genes encoding RNase E. Recently an archaeal counterpart of the eukaryotic exosome was predicted by comparing the gene order in completely sequenced archaeal genomes complemented by sequence profile analysis. The genome of Halobacterium sp., however, lacks most of the genes encoding homologs to eukaryotic exosome proteins (35).

Here, we analyzed the decay of the gvp mRNAs from H.mediterranei as a model system for an archaeal polycistronic mRNA in order to learn more about the role of mRNA decay in gene regulation in archaea. Our data reveal that the transcripts derived from the mc-vac region of H.mediterranei exhibit very different half-lives. Furthermore, the half-lives of some but not all gvp transcripts are affected by the growth phase or the salt concentration in the medium. Thus, different levels of individual gvp transcripts and a variation of transcript levels under diverse growth conditions are not only due to transcriptional regulation but also due to different rates of mRNA decay.

MATERIALS AND METHODS

Strains and growth conditions

Cultivation of H.mediterranei (DSMZ 1411T) and Haloferax volcanii (WFD11) was performed as described previously (2). The media contained either 18 or 25% NaCl. Haloferax volcanii strains harboring plasmids were grown in medium supplemented with novobiocin (0.2 µg/ml final concentration) or mevinolin derived from Lovastatin (provided by Merck/ Sharp and Dohme, 6 µg/ml final concentration). For inhibition of transcription, actinomycin D was added to a final concentration of 100 µg/ml.

In vivo labeling of RNA

Cultures were grown to early exponential phase (OD600nm = 0.45) or to early stationary phase (OD600nm = 2.8). Radio labeled uridine ([5,6-3H], 32 Ci/mmol, Amersham Biosciences) was added to a final concentration of 100 µCi/ml for exponential growing cultures or 300 µCi/ml for stationary phase cultures at time point 0. Samples of 20 µl were collected at several time points after addition of uridine and mixed with 500 µl of unlabeled cells. Cold trichlor acetic acid (TCA) was added to a final concentration of 11.5% (w/v). After incubation on ice for 5 min, the samples were centrifuged and the TCA-precipitated counts were quantified in a Liquid Scintillation counter (Beckman).

RNA isolation and northern blot analysis

Total RNA was isolated according to the protocol of Nieuwlandt et al. (36) followed by two phenol/chloroform extraction steps. After ethanol precipitation and resuspension in 180 µl buffer A (20 mM NaPO4, pH 6.5, 1 mM EDTA), 20 µl DNase-buffer (200 mM Na–acetate, pH 4.5, 180 mM MgCl2, 100 mM NaCl) and DNase I (15–20 U, Life Technologies) was added followed by incubation for 30 min at room temperature. Twenty microliters 250 mM EDTA, pH 7.0, was added followed by phenol/chloroform extraction and ethanol precipitation. The RNA was dissolved in 50 µl buffer A, 7 µg of RNA per lane was run on a 1% (w/v) agarose 2.2 M formaldehyde gel, and transferred to nylon membrane (Pall Biodyne B) by vacuum pressure blotting (Vacuum Blotter, Appligene) according to the manufacturer’s recommendations.

gvpD or gvpA specific DNA fragments were radiolabeled with [α-32P]dCTP using nick translation (Nick translation kit, Amersham Biosciences) and purified on micro columns (Probe Quant G-50, Amersham Biosciences). Aliquots of 2 × 106 c.p.m. were used per hybridization reaction. The signals were quantified using a phosphorimaging system (Molecular Imager FX, BioRad) and the appropriate software (Quantity One, BioRad).

Plasmid construction

To construct plasmid pGvpDE2.0 a 2045 bp DNA fragment was amplified using chromosomal DNA from H.mediterranei as a template. We used primers GvpDupNco (5′-CAT CCATGGCCCCACCAAAC-3′, introduces NcoI site) and GvpE2.0Kpn (5′-GGAGGTACCTTTCGTCGAGC-3′, introduces KpnI site) for the PCR. The resulting fragment was cut by NcoI and KpnI and inserted next to the fdx promoter of plasmid pJAS35 (7). Transformation of H.volcanii was performed as described previously (2).

RESULTS

The half-lives of the gvp transcripts in H.mediterranei inversely correlate to their sizes

Northern blot analysis detected multiple RNA species transcribed from the gvp promoters (Fig. 1) (2). Transcripts of 0.45 kb gvpD’, 1.3 kb gvpD’, 2.0 kb gvpDE’, 3.0 kb gvpDEF and ∼7 kb gvpDEFGHIJKLM originate from the mcD promoter located upstream of gvpD. The 0.32 kb gvpA, 1.8 kb gvpACN’, 2.4 kb gvpACN’ and 3.0 kb gvpACNO mRNAs originate from the mcA promoter located upstream of gvpA. Except for the 7 kb gvpD-M RNA, all of these transcripts show higher abundance during late exponential and stationary growth phase compared to mid exponential growth phase. Cells grown in 25% salt medium contain higher levels of these mRNAs than cells grown in 18% salt medium (2).

The mRNA amounts detected by northern blot analysis represent steady state levels determined by the rate of transcription as well as by the rate of mRNA decay. To investigate whether differences in mRNA half-lives contribute to the growth stage and salt-dependent expression of the gvp genes, we determined the half-lives of the most abundant gvp mRNAs under different growth conditions after blocking transcription by the addition of actinomycin D. The archaeal RNA polymerase activity is sensitive to heparin and actinomycin D but resistant to specific inhibitors of the bacterial RNA polymerase such as rifampicin (37). Recently the inhibition of transcription by actinomycin D was demonstrated for the hyperthermophilic archaeon Sulfolobus solfataricus (38).

In order to test the virtue of actinomycin D on transcription in Haloferax, we analyzed its effect on the incorporation of radiolabeled uridine in H.mediterranei cultures. Radiolabeled uridine was added to cultures either in early exponential or in early stationary growth phase (OD600nm = 0.45 or 2.8). Twelve minutes after the addition of uridine, actinomycin D was added to the cultures, and the TCA-precipitable counts, reflecting the in vivo rate of total RNA synthesis, were quantified over a total period of 3 h. A linear increase of the incorporated label was observed in exponential growth phase when no inhibitor was added (Fig. 2B). In early stationary phase, the incorporation of label was linear for ∼45 min and then slowed down (Fig. 2C). As shown in Figure 2 final concentrations of actinomycin D of 100 µg/ml were sufficient to efficiently block RNA synthesis. These concentrations of actinomycin D led to a transient retardation of cell growth, but cultures showed normal doubling times 90 min after the addition of the inhibitor (Fig. 2A).

Figure 2.

Figure 2

Open in a new tab

Effect of actinomycin D on growth and RNA synthesis in H.mediterranei. (A) Growth curves of cultures, which remained untreated (closed circles) or after addition of actinomycin D (open circles) at the indicated time. (B and C) Radiolabeled uridine was added to the cultures in exponential growth phase (B) (OD600nm = 0.45) or in stationary phase (C) (OD600nm = 2.8) at time point 0. 100 µg/ml (open circles) of actinomycin D were added 12 min after the addition of uridine while no actinomycin D was added to a control culture (closed circles) and RNA synthesis was followed by measuring the TCA-precipitable counts. The time points of actinomycin D addition are marked by arrows.

First, we analyzed the decay of gvp RNAs from the mc-vac region of H.mediterranei grown in 25% salt medium to an optical density of ∼3.2 at 660 nm (early stationary phase). We determined half-lives for the 0.45 kb gvpD’, the 1.3 kb gvpD’ and the 2.0 kb gvpDE transcripts of 10.2 ± 2 min, 6.8 ± 1 min and 5.5 ± 1 min, respectively (Fig. 3 and Table 1). The 0.45 kb gvpD’ and the 1.3 kb gvpD’ transcripts are visible as rather broad bands on northern blots, indicating that they arise from a population of RNA molecules with heterogenous 5′ or 3′ ends. The presence of multiple 3′ ends has been shown before for the 0.45 kb gvpD’ transcript (2). The 2.0 kb gvpDE mRNA is a broad but distinct band right below the position of the 23S rRNA. The ‘3.0 kb’ gvpDEFG mRNA is a diffuse smear right above the position of the 23S rRNA. We therefore did not determine the half-life for this mRNA population. Considering the migration in relation to the 23S rRNA it is conceivable that part of the ‘3.0 kb’ band arises from unspecific attachment of longer or smaller mRNAs to the 23S rRNA, as often observed on northern blots of total bacterial RNA. This assumption is supported by the data of Röder and Pfeifer (2). When a probe homologous to the gvpL gene was used, light bands were detected at these positions, although this probe should only detect the 7.0 kb gvpDEFGHIJKLM transcript.

Figure 3.

Figure 3

Open in a new tab

Determination of half-lives of the mc-gvpD specific mRNAs. (A) Total RNA from H.mediterranei grown in the presence of 25% salt to an OD600nm of 3.2 was analyzed by northern blot hybridization using a gvpD specific DNA probe. Samples were taken at different time points after addition of actinomycin D. The dashed bars indicate the positions of the 23S and 16S rRNAs, respectively. Numbers at the left mark the approximate sizes in kb. (B) Quantification of the RNA signals shown in (A) by phosphorimaging and deduced half-lives. Squares, 2.0 kb RNA, half-life 5.0 min; triangles, 1.3 kb RNA, half-life 7.5 min; circles, 0.45 kb RNA, half-life 11.0 min. The half-lives given in the text are the average of at least three independent experiments.

Table 1. Half-lives of gvp transcripts in Haloferax determined by northern blot analysis after addition of actinomycin D.

  H.mediterranei H.volcanii (A+D+E) H.volcanii (pgvpDE2.0)
mRNA species 25% salt, OD 3.2 18% salt, OD 3.2 25% salt, OD 2.2 18% salt, OD 2.2 25% salt, OD 0.8 25% salt, OD 1.6 25% salt, OD 3.2
0.45 kb gvpD 10.2 ± 2.0 min 10.8 ± 2.0 min     not detectable 9.0 ± 2.0 min 21.0 ± 4.0 min
1.3 kb gvpD 6.8 ± 1.0 min 7.4 ± 1.0 min          
2.0 kb gvpDE 5.5 ± 1.0 min 4.5 ± 1.0 min     11.0 ± 3.0 min 9.0 ± 1.5 min 10.0 ± 1.5 min
0.32 kb gvpA 80.0 ± 9.0 min 39.0 ± 7.0 min 160 ± 15 min 90 ± 8 min      
3.0 kb gvpACNO 11.5 ± 2.0 min 12.5 ± 2.5 min          
Open in a new tab

The numbers represent the mean of at least three independent experiments and the maximal deviation is indicated.

We also determined the half-lives of mRNAs transcribed from the mcA promoter located in front of gvpA. Haloferax mediterranei was grown in the presence of 25% salt and cells were collected at an optical density of ∼3.2 at 660 nm. We determined half-lives for the 0.32 gvpA, the 1.8 kb gvpACN’, the 2.4 kb gvpACN’ and the 3.0 gvpACNO mRNAs of 80 ± 9 min, 20 ± 5 min, 16 ± 3 min and 11.5 ± 2 min, respectively (Fig. 4; results for 0.32 and 3.0 kb see Table 1). While the 0.32 kb gvpA mRNA migrates as a relative distinct band, the other mRNAs are represented by more diffuse smears. The ‘1.8 kb’ gvpACN’ mRNA migrates just above of the position of the 16S rRNA. The ‘2.4 kb’ gvpACN’ mRNA migrates below the position and the ‘3.0 kb’ gvpACNO mRNA just above the position of the 23S rRNA. Thus, the position of these mRNA bands may again indicate that part of these signals arise from unspecific attachment of longer or shorter mRNAs to the ribosomal RNA. While the ‘1.8 kb’ and the ‘2.4 kb’ mRNA bands can not be distinguished from the unspecific bands around the rRNA, the 3.0 kb mRNA is rather visible as a distinct band. Since it is likely that the signals of the ‘1.8 kb’ and ‘2.4 kb’ bands partly arise from mRNAs of different sizes, which attach to the rRNA, we will not consider these RNA bands for our further investigations.

Figure 4.

Figure 4

Open in a new tab

Determination of half-lives of the mc-gvpA specific mRNAs. (A and B) Total RNA from H.mediterranei grown at the indicated salt concentrations to an OD600nm of ∼3 was analyzed by northern blot hybridization using a gvpA specific DNA probe. Samples were taken at different time points after addition of actinomycin D. The dashed bars indicate the positions of the 23S and 16S rRNAs, respectively. Numbers at the left mark the approximate sizes in kb. Different time points for RNA isolation and different exposure times were chosen in order to quantify the bands corresponding to the different transcripts. (C) Quantification of the 3.0 kb RNA band from (A) and the 0.32 kb RNA band from (B) by phosphorimaging and deduced half-lives. Open triangles, 3.0 kb RNA, 18% salt, half-life 14.0 min; closed triangles, 3.0 kb RNA, 25% salt, half-life 13.5 min; open circles, 0.32 RNA, 18% salt, half-life 44 min; closed circles, 0.32 kb RNA, 25% salt, half-life 82 min. The half-lives given in the text are the average of at least three independent experiments.

Our analysis revealed an inverse correlation of mRNA size and stability. Thus, the higher stabilities contribute to the higher abundance of the smaller mRNA species.

Salt concentration affects the half-lives of the 0.32 kb gvpA mRNA in H.mediterranei and H.volcanii ADE transformants

The amounts of all mc-gvp transcripts in H.mediterranei are increased when cells are grown at high salt concentration. To determine whether a different stability of gvp mRNAs contributes to the salt-dependent regulation of gvp gene expression we also determined RNA half-lives in cultures grown at 18% salt. This was the minimal salt concentration which allowed reliable detection and quantification of most gvp transcripts. Cells of H.mediterranei were harvested in early stationary growth phase. Half-lives of the mcD derived transcripts 0.45 kb gvpD’, 1.3 kb gvpD’ and 2.0 kb gvpDE’ were 10.8 ± 2 min, 7.4 ± 1 min and 4.5 ± 1 min, respectively (Table 1). These values show no significant differences to the half-lives determined in 25% salt medium indicating that salt does not affect the stability of RNAs transcribed from the promoter upstream of gvpD. The half-life of the 0.32 kb gvpA mRNA was determined to be 39.0 ± 7 min during growth in 18% salt medium. This is <50% of the half-life we determined during growth at 25% salt (Fig. 4 and Table 1). Thus, the salt concentration clearly affects the stability of the 0.32 kb gvpA mRNA and contributes to the salt-dependent expression of gvpA. There was, however, no significant effect of salt on the stability of the 3.0 kb gvpACNO mRNA (Fig. 4 and Table 1).

To test whether this salt-dependent stability of the gvpA mRNA is specific for H.mediterranei the decay of the gvpA mRNA was analyzed in the gas vesicle negative strain H.volcanii containing a plasmid harboring the mc-gvpA and mc-gvpDE genes (ADE construct) (2). The 0.32 kb gvpA mRNA was considerably more stable in H.volcanii than in H.mediterranei and decayed with a half-life of ∼90 min in 18% salt medium in late exponential growth phase (Table 1). This value compares to the half-life we determined for the gvpA mRNA in H.mediterranei grown in 25% salt medium. A salt concentration of 25% in the medium increased the half-live of the gvpA mRNA even further to ∼160 min (Table 1). Despite the relative high errors in determining these long half-lives the differences are significant enough to allow the conclusion that the decay of the gvpA mRNA is also salt dependent in H.volcanii.

Differential RNA processing contributes to the growth phase dependent expression of the 0.45 kb gvpD mRNA

The abundance of the gvp mRNAs is not only affected by salt concentration but also by growth phase (2). To test whether different mRNA stabilities contribute to this growth phase dependent regulation, we analyzed the half-lives of the gvp mRNAs from H.mediterranei grown in the presence of 25% salt during mid exponential growth (OD600nm ∼0.8).

We determined a half-life of 5.6 ± 1 min for the 1.3 kb gvpD’ transcript (data not shown). The average of this half-life is slightly shorter than the average determined during early stationary phase. Considering the relative errors of the half-life determinations this difference is not significant. The half-lives of the 0.45 kb gvpD’ and the 2.0 kb gvpDE’ transcripts were not determined since their amounts were too low for a reliable quantification.

During early exponential growth the signal for the 2.0 kb mRNA was clearly more intense than the signal for the 0.45 kb mRNA, whereas during early stationary growth phase the signal of the 0.45 kb gvpD’ transcript showed higher intensity than the signal of the 2.0 kb mRNA (data not shown). This strongly suggests that the growth phase dependent differences in the levels of the 0.45 kb gvpD’ transcript are due to post-transcriptional processes. To prove this assumption we constructed plasmid pGvpDE2.0, which contains the first 1940 nt of the gvpDE genes under the control of the fdx promoter of plasmid pJAS35 (7). This promoter originates from the ferredoxin gene of Halobacterium salinarum. The activity of the fdx promoter is high during exponential growth and drops during stationary growth phase, independent of the salt concentration (D. Gregor and F. Pfeifer, unpublished data). Haloferax volcanii was transformed with plasmid pGvpDE2.0. Actinomycin D was added to cultures in mid exponential growth phase (OD600nm = 0.8–0.9), late exponential growth phase (OD600nm = 1.6) or in early stationary phase (OD600nm = 3.0–3.2) and the gvp transcripts were analyzed at various time points by northern analysis (Fig. 5). Due to the use of the fdx promoter relatively high amounts of gvp RNAs were detected in H.volcanii throughout growth, allowing the half-life determination. The ‘0.45 kb’ gvpD’ transcript population occurred again as a broad band. The abundances of the band around 0.45 kb and of the 2.0 kb band were clearly dependent on growth phase. The half-life of the 2.0 kb gvpDE mRNA in H.volcanii grown to an OD600nm of 3.2 was determined to be 10.0 ± 1.5 min (Fig. 5), which was significantly longer than determined for the same mRNA species transcribed from the H.mediterranei chromosome (5.5 min, see Table 1). In mid exponential growth the half-life of the 2.0 kb mRNA was 11.0 ± 3 min. Thus the higher abundance of the 2.0 kb mRNA in early growth stage is not due to higher stability of this mRNA species.

Figure 5.

Figure 5

Open in a new tab

Determination of half-lives of the mc-gvpD specific mRNAs. (A) Total RNA isolated from H.volcanii (pGvpDE2.0) grown in the presence of 25% salt to growth phase as indicated was analyzed by northern blot hybridization using a gvpD specific DNA probe. Samples were taken at different time points after addition of actinomycin D. Numbers at the right mark the approximate sizes in kb. (B) Quantification of the RNA signals shown in (A) by phosphorimaging and deduced half-lives. Open triangles, 2.0 kb RNA, OD600nm = 1.6, half-life 10.5 min; closed triangles, 2.0 kb RNA, OD600nm = 3.2, half-life 9.0 min; open circles, 0.45 kb RNA, OD600nm = 1.6, half-life 9.5 min; closed circles, 0.45 kb RNA smear, OD600nm = 3.2, half-life 22.0 min. The half-lives given in the text are the average of at least three independent experiments.

The 0.45 kb RNA band decreased with a half-life of 21 ± 4 min in early stationary growth phase (OD600nm = 3.2) which is again longer than observed for the 0.45 kb RNA population transcribed from the H.mediterranei chromosome (Fig. 4 and Table 1). During late exponential growth (OD600nm = 1.6) the half-life of this RNA population was 9.0 ± 2.0 min, while almost no 0.45 kb RNA population was detectable during mid-exponential growth (OD600nm = 0.8). These results imply that a higher RNA stability contributes to the higher abundance of the 0.45 kb RNA population at later stages in growth.

DISCUSSION

To investigate whether mRNA processing contributes to regulated gene expression in archaea, we analyzed the decay of various gvp transcripts in two Haloferax species under different growth conditions. The half-lives determined for the individual gvp transcripts vary from 4 to 80 min in H.mediterranei, whereas in H.volcanii significantly longer half-lives were observed. The most stable and most abundant transcript was the gvpA mRNA encoding the major gas vesicle structural protein. For most transcripts the stability correlates with its relative abundance. Thus, differences in the rate of decay contribute to the different abundances of gvp transcripts. Little is known about the stability of mRNA species in archaea. In an earlier study the growth of Methanococcus vannielii was inhibited by addition of bromoethanesulphonate or by the removal of hydrogen and the decrease of some mRNA species was quantified by northern blot analysis (39). The half-lives of the secY, mcr, Mva and argG transcripts varied from 7 to 57 min, which is in the same range as determined for the gvp transcripts in Haloferax in this report. Recently, the stability of the dgh1, tfb1, tfb2, sod, gln1 and malA transcripts was determined in S.solfataricus after blocking transcription with actinomycin D (38). While the tfb1 and sod transcripts exhibited half-lives of at least 2 h, the other transcripts decayed with half-lives between 6.3 and 37 min, indicating again a large range of half-lives for individual transcripts.

To date, the origin of the different sized gvp mRNAs is not known. The individual gvp mRNAs could either originate from partial termination of transcription at certain positions or by mRNA processing. It is conceivable that initially full length mRNAs are produced, which are then degraded by an exosome-like protein complex in a 3′→5′ direction. A slow-down of this exonucleolytic process at highly structured mRNA regions would explain the occurrence of different sized mRNA species. Secondary structures near the 3′ end, which protect against 3′→5′ exonucleases have been demonstrated for different bacterial mRNAs (40–42). The fact that the sizes and half-lives of the gvp mRNAs correlate inversely is in accordance with this model. At this time, however, we can not exclude the participation of endoribonucleases in gvp mRNA processing. If endonucleases were involved in rate-limiting cleavage which is then followed by fast endo- or exonucleolytic decay, one would have to postulate faster attack of the more 3′ gvp mRNA segments. This also holds true if the different sized gvp mRNAs originate from partial transcriptional termination.

Most importantly, our study reveals that mRNA processing contributes to gene regulation in archaea as known for bacteria and eukarya. The salt concentration clearly affected the stability of the 0.32 kb gvpA mRNA, which was almost reduced to 50% when cells were grown in 18% salt medium compared to 25% salt. In contrast the stability of all other gvp mRNAs was unaffected. It is conceivable that an mRNA secondary structure located at the 3′ end of the 0.32 kb gvpA mRNA protects against 3′→5′ exonucleases and is stabilized by a high salt concentration. The 3′ end of the gvpA transcript maps to a cytosine preceded by a stretch of five uridines (UUUUUC) located 64 nt downstream of the second stop codon of the gvpA reading frame and suggested to function as a terminator of transcription (43). A stem–loop structure (10 nt of complementary sequence separated by a 10 nt loop) is found within the 64 nt terminal mRNA sequence. In bacteria mRNA secondary structures can function as both, i.e. as transcriptional terminators and in protection against 3′→5′ exonucleases (44). If the gvp mRNA was exclusively degraded by 3′→5′ exonucleases, and pausing at RNA secondary structures would generate the different sized gvp RNAs, one would expect a similar effect of salt on the stability of such structures and consequently on the decay of other gvp transcripts as well. However, the differential effect of salt on the stability of the individual gvp transcripts suggests the involvement of different mechanisms in the rate-limiting initiation of their decay.

The abundance of the gvp mRNAs in H.mediterranei is strongly affected by growth phase (2). These transcripts occur predominantly during the stationary growth phase, and the high amount of gvp mRNAs is mainly due to the activator protein GvpE that activates both mc-vac promoters (2,5). The amount of transcripts formed during exponential growth was too low to determine half-lives. Thus, the gvpD gene and part of gvpE were inserted in the expression vector pJAS35 and expressed under fdx promoter control in H.volcanii transformants. This promoter is mainly active during exponential growth and not affected by salt concentration (45) (D. Gregor and F. Pfeifer, unpublished data). The half-lives of the 0.45 kb gvpD’ and the 2.0 kb gvpDE mRNA were almost twice as long as observed for the same RNA populations in H.mediterranei. Since it is unlikely that the higher amount of the gvp mRNAs due to expression from a plasmid leads to higher stability, these data imply species-specific differences in the efficiency of mRNA decay. It is also known for bacteria that the half-life of certain mRNAs is different when expressed in different species (46). The half-life of the 2.0 kb mRNA did not show a significant growth phase dependence, while the 0.45 kb transcript population indicated a higher stability during the stationary growth phase than in exponential growth, correlating well with the higher abundance of the 0.45 kb RNA in the stationary growth phase. Thus, there is no general growth phase-dependent effect on mRNA decay in Haloferax. The high abundance of the 0.45 kb RNA population in stationary growth phase might be due to an early termination in the gvpD coding region. A stronger transcriptional termination during stationary growth phase would contribute to the lower amount of the 2.0 kb mRNA, and also lead (together with the increased stability of the small transcript) to the abundant 0.45 kb RNA. The multiple 3′ ends determined for the 0.45 kb mRNA population do not map close to U-rich sequences (2). How ever, little is known about the mechanisms of transcriptional termination in archaea to date.

The 0.45 kb gvpD’ transcripts do not encompass a complete reading frame and are thus not involved in the production of a Gvp protein required for gas vesicle synthesis. Whether this transcript has a function in gas vesicle formation remains to be seen—the main purpose of the formation of this transcript by an early transcript termination might be the reduction of the gvpEFGHIJKLM gene expression despite of the activation of the mcD promoter by GvpE.

Taken together our analyses suggest that different processes are responsible for the generation of the individual gvp RNAs. These may include partial transcriptional termination and possibly complex mechanisms in mRNA processing as observed for bacterial polycistronic operons (reviewed in 9–11). Overall, our studies demonstrated that mRNA stability and processing contributes to regulated gene expression in archaea. Therefore, the mechanisms involved in mRNA processing in archaea deserve further attention.

Acknowledgments

ACKNOWLEDGEMENTS

We thank Peter Zimmermann for transforming plasmid pGvpDE2.0 into H.volcanii and for valuable advice. Lovastatin was kindly provided by Merck/Sharp and Dohme. This work was supported by Deutsche Forschungs gemeinschaft (Kl563/12-1 and Pf165/7-1) and by Fonds der Chemischen Industrie.

REFERENCES

  • 1.Englert C., Krüger,K., Offner,S. and Pfeifer,F. (1992) Three different but related gene clusters encoding gas vesicles in halophilic Archaea. J. Mol. Biol., 227, 586–592. [DOI] [PubMed] [Google Scholar]
  • 2.Röder R. and Pfeifer,F. (1996) Influence of salt on the transcription of the gas-vesicle genes of Haloferax mediterranei and identification of the endogenous transcriptional activator gene. Microbiology, 142, 1715–1723. [DOI] [PubMed] [Google Scholar]
  • 3.Englert C., Horne,M. and Pfeifer,F. (1990) Expression of the major gas vesicle gene in the halophilic archaebacterium Haloferax mediterranei is modulated by salt. Mol. Gen. Genet., 222, 225–232. [DOI] [PubMed] [Google Scholar]
  • 4.Krüger K., Hermann,T., Armbruster,V. and Pfeifer,F. (1998) The transcriptional activator GvpE for the halobacterial gas vesicle genes resembles a basic region leucine-zipper regulatory protein. J. Mol. Biol., 279, 761–771. [DOI] [PubMed] [Google Scholar]
  • 5.Gregor D. and Pfeifer,F. (2001) Use of a halobacterial bgaH reporter gene to analyse the regulation of gene expression in halophilic archaea. Microbiology, 147, 1745–1754. [DOI] [PubMed] [Google Scholar]
  • 6.Englert C., Wanner,G. and Pfeifer,F. (1992) Functional analysis of the gas vesicle gene cluster of the halophilic archaeon Haloferax mediterranei defines the vac-region boundary and suggests a regulatory role for the gvpD gene or its product. Mol. Microbiol., 6, 3543–3550. [DOI] [PubMed] [Google Scholar]
  • 7.Pfeifer F., Offner,S., Krüger,K., Ghahraman,P. and Englert,C. (1994) Transformation of the halophilic archaea and investigation of gas vesicle synthesis. Syst. Appl. Microbiol., 16, 569–577. [Google Scholar]
  • 8.Pfeifer F., Zotzel,J., Kurenbach,B., Röder,R. and Zimmermann,P. (2001) A p-loop motif and two basic regions in the regulatory protein GvpD are important for the repression of gas vesicle formation in the archaeon Haloferax mediterranei. Microbiology, 147, 63–73. [DOI] [PubMed] [Google Scholar]
  • 9.Grunberg-Manago M. (1999) Messenger RNA stability and its role in control of gene expression in bacteria and phages. Annu. Rev. Genet., 33, 193–227. [DOI] [PubMed] [Google Scholar]
  • 10.Rauhut R. and Klug,G. (1999) mRNA degradation in bacteria. FEMS Microbiol. Rev., 23, 353–370. [DOI] [PubMed] [Google Scholar]
  • 11.Takayama K. and Kjelleberg,S. (2000) The role of mRNA stability during bacterial stress responses and starvation. Environ. Microbiol., 2, 355–365. [DOI] [PubMed] [Google Scholar]
  • 12.Carpousis A.J., Van Houwe,G., Ehretsmann,C. and Krisch,H.M. (1994) Copurification of E.coli RNase E and PNPase: evidence for a specific association between two enzymes important in RNA processing and degradation. Cell, 76, 889–900. [DOI] [PubMed] [Google Scholar]
  • 13.Py B., Causton,H., Mudd,E.A. and Higgins,C.F. (1994) A protein complex mediating mRNA degradation in Escherichia coli. Mol. Microbiol., 14, 717–729. [DOI] [PubMed] [Google Scholar]
  • 14.Jäger S., Fuhrmann,O., Heck,C., Hebermehl,M., Rauhut,R. and Klug,G. (2001) An mRNA degrading complex in Rhodobacter capsulatus. Nucleic Acids Res., 29, 4581–4588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Py B., Higgins,C.F., Krisch,H.M. and Carpousis,A.J. (1996) A DEAD-box RNA helicase in the Escherichia coli RNA degradosome. Nature, 381, 169–172. [DOI] [PubMed] [Google Scholar]
  • 16.Blum E., Carpousis,A.J. and Higgins,C.F. (1997) Polyphosphate kinase is a component of the Escherichia coli RNA degradosome. Mol. Microbiol., 27, 387–398. [DOI] [PubMed] [Google Scholar]
  • 17.Mackie G.A. (1998) Ribonuclease E is a 5′-end-dependent endonuclease. Nature, 395, 720–723. [DOI] [PubMed] [Google Scholar]
  • 18.Steege D.A. (2000) Emerging features of mRNA decay in bacteria. RNA, 6, 1079–1090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Sakar N. (1997) Polyadenylation of mRNA in prokaryotes. Annu. Rev. Biochem., 66, 173–197. [DOI] [PubMed] [Google Scholar]
  • 20.Xu F., Lin-Chao,S. and Cohen,S.N. (1993) The Escherichia coli pcnB gene promotes adenylylation of antisense RNAI of ColE1-type plasmids in vivo and degradation of RNAI decay intermediates. Proc. Natl Acad. Sci. USA, 90, 6756–6760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.O’Hara E.B., Chechanova,J.A., Ingle,C.A., Kushner,Z.R., Peters,E. and Kushner,S.R. (1995) Polyadenylation helps regulate mRNA decay in Escherichia coli. Proc. Natl Acad. Sci. USA, 92, 1807–1811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Mitchell P., and Tollervey,D. (2000) mRNA stability in eukaryotes. Curr. Opin. Genet. Dev., 10, 193–198. [DOI] [PubMed] [Google Scholar]
  • 23.Mitchell P., Petfalski,E., Shevchenko,A., Mann,M. and Tollervey,D. (1997) The exosome: a conserved eukaryotic RNA processing complex containing multiple 3′→5′ exoribonucleases. Cell, 91, 457–466. [DOI] [PubMed] [Google Scholar]
  • 24.Decker C.J. (1998) The exosome: a versatile RNA processing machine. Curr. Biol., 8, R238–R240. [DOI] [PubMed] [Google Scholar]
  • 25.van Hoof A. and Parker,R. (1999) The exosome: a proteasome for RNA? Cell, 99, 347–350. [DOI] [PubMed] [Google Scholar]
  • 26.Brown T.D. and Reeve,J.N. (1985) Polyadenylated non-capped RNA from the archaebacterium Methanococcus vannielii. J. Bacteriol., 162, 909–917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Brown T.D. and Reeve,J.N. (1986) Polyadenylated RNA isolated from the archaebacterium Halobacterium halobium. J. Bacteriol., 166, 686–688. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Volkl P., Markiewicz,P., Baikalov,C., Fitz-Gibbon,S., Stetter,K.O. and Miller,J.H. (1996) Genomic and cDNA sequence tags of the hyperthermophilic archaeon Pyrobaculum aerophilum. Nucleic Acids Res., 24, 4373–4378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Pannucci J.A., Haas,E.S., Hall,T.A., Harris,J.K. and Brown,J.W. (1999) RNase P RNAs from some Archaea are catalytically active. Proc. Natl Acad. Sci. USA, 96, 7803–7808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Lykke-Andersen J. and Garret,R.A. (1997) RNA–protein interaction of an archaeal homotetrameric splicing endoribonuclease with an exceptional evolutionary history. EMBO J., 16, 6290–6300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Li H., Trotta,C.R. and Abalson,J. (1998) Crystal structure and evolution of a transfer RNA splicing enzyme. Science, 280, 279–284. [DOI] [PubMed] [Google Scholar]
  • 32.Russell A.G., Ebhardt,H. and Dennis,P.P. (1999) Substrate requirements for a novel archaeal endonuclease that cleaves within the 5′ external transcribed spacer of Sulfolobus acidocaldarius precursor rRNA. Genetics, 152, 1373–1385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Muroya A., Tsuchiya,D., Ishikawa,M., Haruki,M. Kanaya,S. and Morikawa,K. (2001) Catalytic center of an archaeal type 2 ribonuclease H as revealed by X-ray crystallographic and mutational analysis. Protein Sci., 10, 707–714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Franzetti B., Sohlberg,B., Zaccai,G. and von Gabain,A. (1997) Biochemical and serological evidence for an RNase E-like activity in halophilic archaea. J. Bacteriol., 179, 1180–1185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Koonin E.V., Wolf,Y.I. and Aravind,L. (2001) Prediction of the archaeal exosome and its connection with the proteasome and the translation and transcription machineries by a comparative-genomic approach. Genome Res., 11, 240–252. [DOI] [PMC free article] [PubMed]
  • 36.Nieuwlandt D.T., Palmer,J.R., Armbruster,D.T., Kuo,Y.-P., Oda,W. and Daniels,C.J. (1995) A rapid procedure for the isolation of RNA from Haloferax volcanii. In Robb,F.T. and Place,A.R. (eds), Archaea: A Laboratory Manual. Cold Spring Harbor Laboratory Press, New York, NY, pp. 161–162.
  • 37.Sippel A. and Hartmann,G. (1968) Mode of action of rafamycin on the RNA polymerase reaction. Biochim. Biophys. Acta, 157, 218–219. [DOI] [PubMed] [Google Scholar]
  • 38.Bini E., Dikshit,V., Dirksen,K., Drozda,M. and Blum,P. (2002) Stability of mRNA in the hyperthermophilic archaeon Sulfolobus solfataricus. RNA, 9, 1129–1136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Hennigan A.N. and Reeve,J.N. (1994) mRNAs in the methanogenic archaeon Methanococcus vannielii: numbers, half-lives and processing. Mol. Microbiol., 11, 655–670. [DOI] [PubMed] [Google Scholar]
  • 40.Newbury S.F., Smith,N.H. and Higgins,C.F. (1987) Differential mRNA stability controls relative gene expression within a polycistronic operon. Cell, 51, 1131–1143. [DOI] [PubMed] [Google Scholar]
  • 41.Klug G., Adams,C.W., Belasco,J.G., Dörge,B. and Cohen,S.N. (1987) Biological consequences of segmental alterations in mRNA stability: effects of the deletion of the intercistronic hairpin loop region of the Rhodobacter capsulatus puf operon. EMBO J., 6, 3515–3520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Chen C.Y., Beatty,J.T., Cohen,S.N. and Belasco,J.G. (1988) An intercistronic stem–loop structure functions as an mRNA decay terminator necessary but insufficient for puf mRNA stability. Cell, 52, 609–619. [DOI] [PubMed] [Google Scholar]
  • 43.Englert C. and Pfeifer,F. (1993) Analysis of gas vesicle gene expression in Haloferax mediterranei reveals that GvpA and GvpC are both gas vesicle structural proteins. J. Biol. Chem., 268, 9329–9336. [PubMed] [Google Scholar]
  • 44.Klug G. and Cohen,S.N. (1990) Combined action of multiple hairpin loop structures and sites of rate-limiting endonucleolytic cleavage determine differential degradation rates of individual segments within polycistronic puf operon mRNA. J. Bacteriol., 172, 5140–5146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Offner S., Hofacker,A., Wanner,G. and Pfeifer,F. (2000) Eight out of fourteen gvp genes are sufficient for formation of gas vesicles in halophilic archaea. J. Bacteriol., 182, 4328–4336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Klug G., Jock,S. and Rothfuchs,R. (1992) The rate of decay of Rhodobacter capsulatus specific puf mRNA segments is differentially affected by RNase E activity in E.coli. Gene, 121, 95–102. [DOI] [PubMed] [Google Scholar]

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press

RESOURCES