Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. 2007 Feb 27; 104(9): 3135–3140.
Published online 2007 Feb 20. doi:  10.1073/pnas.0611611104
PMCID: PMC1805545

Regulation of a glutamyl-tRNA synthetase by the heme status


Glutamyl-tRNA (Glu-tRNA), formed by Glu-tRNA synthetase (GluRS), is a substrate for protein biosynthesis and tetrapyrrole formation by the C5 pathway. In this route Glu-tRNA is transformed to δ-aminolevulinic acid, the universal precursor of tetrapyrroles (e.g., heme and chlorophyll) by the action of Glu-tRNA reductase (GluTR) and glutamate semialdehyde aminotransferase. GluTR is a target of feedback regulation by heme. In Acidithiobacillus ferrooxidans, an acidophilic bacterium that expresses two GluRSs (GluRS1 and GluRS2) with different tRNA specificity, the intracellular heme level varies depending on growth conditions. Under high heme requirement for respiration increased levels of GluRS and GluTR are observed. Strikingly, when intracellular heme is in excess, the cells respond by a dramatic decrease of GluRS activity and the level of GluTR. The recombinant GluRS1 enzyme is inhibited in vitro by hemin, but NADPH restores its activity. These results suggest that GluRS plays a major role in regulating the cellular level of heme.

Keywords: gene expression, protein biosynthesis, tetrapyrrole biosynthesis

In addition to their crucial role in aminoacyl-tRNA formation during protein synthesis (1) aminoacyl-tRNA synthetases have acquired additional functions including regulation of gene expression, RNA splicing, or cytokine activity (2, 3).

Glutamyl-tRNA (Glu-tRNA) synthetase (GluRS), which forms Glu-tRNA by the esterification of glutamate to the 3′ end of tRNA, is found in different organisms in two forms: a discriminating enzyme (D-GluRS) that recognizes exclusively tRNAGlu and a relaxed specificity nondiscriminating enzyme that recognizes both tRNAGln and tRNAGlu (ND-GluRS). This ND-GluRS produces Glu-tRNAGln in the many organisms (most bacteria and all archaea and eukaryotic organelles) that lack GluRS; the mischarged Glu-tRNAGln is then transformed by a tRNA-dependent amidotransferase to Gln-tRNAGln (4). Genomewide analysis revealed that many bacteria contain duplicated GluRSs (GluRS1, GluRS2) (5). In addition, some organisms contain a truncated version of GluRS (GluQRS) that is involved in the modification of tRNAAsp (6, 7). Thus, the GluRS enzymes form a family of diverse and versatile proteins that synthesize multiple Glu-tRNAs with different functions.

Apart from its role in protein biosynthesis, Glu-tRNAGlu is also the substrate for the biosynthesis of tetrapyrroles (heme, chlorophyll, and other derivatives) by the C5 pathway found in plants, archaea, and most bacteria (810). Thus, the glutamate moiety of Glu-tRNA is reduced by the Glu-tRNA reductase (GluTR) to glutamate semialdehyde (GSA) (11). GSA is then converted to δ amino levulinic acid (ALA) by the transamination of the ammonium group from position 1 to position 2 by GSA aminomutase (12, 13). ALA is transformed to tetrapyrroles by a universally conserved pathway (14).

The biosynthesis of tetrapyrroles is a finely tuned process because of the many functions and the potential toxicity of its products. Heme plays a central role in the regulation of tetrapyrrole biosynthesis as a regulator of ALA formation. Several lines of evidence indicate that either the intracellular concentration and/or the activity of GluTR are regulated by heme. Conditional proteolysis occurs in a direct heme-dependent process in Salmonella (15) or through the interaction with FLU/FLP proteins in plants (16, 17). Increased transcription of hemA (the gene encoding GluTR) also controls the level of GluTR in Pseudomonas (18). Alternatively, feedback regulation of GluTR activity by heme (19) is also a mechanism that regulates the formation of ALA. Because of these regulatory features and the fact that in model organisms only a small fraction of Glu-tRNA is transformed to tetrapyrroles, GluTR is regarded as the first enzyme committed to tetrapyrrole biosynthesis.

Heme is the cofactor of cytochromes in respiratory chains (20). This porphyrin molecule also plays roles in iron homeostasis (21), protection against reactive oxygen species (14), and as cofactor of PAS (common domain in period circadian protein, aryl hydrocarbon receptor nuclear translocator, single-mind protein) and globin-like proteins for bacterial signal transduction (22). A fine balance between the content of protoheme and iron to form active heme is critical for cell viability because an excess of any of the two components might be toxic to cells because of oxidative damage of biomolecules (23).

Acidithiobacillus ferrooxidans, an acidophilic γ-proteobacterium that uses ferrous ion (Fe2+), sulfur (S0), or reduced sulfur compounds as electron donors (24), is a suitable model to study the molecular mechanisms involved in the regulation of heme content. In fact, A. ferrooxidans growing in different substrates is able to modulate the content of cytochromes, reaching up to 10% of the total proteins in cells grown in Fe2+ medium (24, 25). Our working hypothesis is that some of the components of the complex Glu-tRNA formation system from A. ferrooxidans might have a regulatory role in the biosynthesis of heme. This system is formed by several tRNAs and two paralog GluRS enzymes; GluRS1 is specific for tRNAs with augmented D stem (tRNAGlu and tRNAGln), and GluRS2 preferentially aminoacylates a tRNAGln, which has only 3 bp at the D stem (5, 26). The fact that A. ferrooxidans Glu-tRNAGlu3 is not substrate for GluTR suggests that some components of the system are committed to specific roles in either protein or heme biosynthesis (27). The data presented here reveal that the levels of both GluRS1 and GluTR are regulated by the heme status. Strikingly, the activity of GluRS1 was also modified under heme excess. These data led us to propose that GluRS1 and GluTR in A. ferrooxidans both are committed to the regulation of heme biosynthesis.


Aminoacylation of tRNAGlu Correlates with the Demand of Heme.

Cytochromes represent up to 10% of total protein from A. ferrooxidans grown in Fe2+, most of them being higher than in cells grown in S0 (24, 25). To test whether the level of cytochromes correlates with the heme content, we measured total heme in cells grown under different culture conditions (Table 1). The concentration of heme was 1.8-fold higher in cells grown in Fe2+ than in S0. An additional increase of heme content to 2.8-fold was obtained when the cells were grown in Fe2+ medium supplemented with ALA (Fe2+/ALA), the universal precursor of heme biosynthesis. The maximum heme concentration in A. ferrooxidans was up to 9-fold higher than in Escherichia coli grown in LB medium (55.1 pmol/mg of protein). Despite the higher amount of heme produced in A. ferrooxidans grown in Fe2+/ALA, there was no effect on the cell growth rates (data not shown), suggesting that a mechanism exists that increases the cells' tolerance to the eventual damage of biomolecules by heme.

Table 1.
Effect of growth conditions on the GluRS activity and heme content of extracts from A. ferrooxidans

Considering the high and variable amounts of heme produced in A. ferrooxidans cultured under different conditions, we also predicted that variable amounts of Glu-tRNA could be used for heme biosynthesis. Because of the lower level of heme in cells grown in medium containing S0 compared with Fe2+, a lower utilization of Glu-tRNA as precursor of heme biosynthesis would be expected. Similarly, the presence of ALA in the medium should permit the cells to bypass the utilization of Glu-tRNA as precursor for its biosynthesis. Thus, fully aminoacylated tRNAGlu is expected in the cells where the demand for Glu-tRNA is lower (i.e., in S0 or Fe2+/ALA). Conversely, in cells grown in Fe2+, less tRNAGlu should be aminoacylated. To test these predictions, the extent of aminoacylation of specific substrate tRNAs of each GluRS was evaluated in total tRNA obtained from A. ferrooxidans grown in the relevant culture conditions. Aminoacylated and deacylated forms of tRNAs were specifically detected by Northern blot after electrophoresis under denaturating conditions (Fig. 1). Contrary to our expectations, we observed that tRNAGlu1 and tRNAGln2 (equivalent to tRYACUGGlu in ref. 5), both substrates of GluRS1, were fully aminoacylated only in cells grown in Fe2+. With these tRNAs, a small fraction was deacylated in cells grown in S0 or Fe2+/ALA. Fully charged tRNAGln3 (equivalent to tRYAUUGGlu in ref. 5), aminoacylated by GluRS2, was observed in all culture conditions. These results suggest that under reduced requirements of ALA for heme synthesis lower levels of GluRS1 activity might occur.

Fig. 1.
In vivo aminoacylated or deacylated tRNAs. Total tRNA obtained from cells grown in Fe2+, S°, or Fe2+/ALA treated with periodate directly (−NaOH) or after deacylation (+NaOH) was electrophoresed on a polyacrylamide gel and transferred to ...

Growth of A. ferrooxidans in ALA Inhibits the Activity of GluRS1.

To test the hypothesis that the variations in the aminoacylation of the tRNA substrates of GluRS1 is the result of a mechanism that regulates the function of the enzyme, the activity and the steady-state level of GluRS1 were analyzed in extracts from cells grown in the relevant culture conditions.

The functionality of GluRS1 was evaluated by the aminoacylation of tRNAGlu2 transcript, an adequate substrate of the enzyme (27), using A. ferrooxidans extracts. Specific activity of GluRS1 was the same in cells grown in Fe2+ and S0 (Table 1). Strikingly, the specific activity of GluRS1 was markedly reduced when the cells were grown in either Fe2+ or S0 supplemented with ALA (200 μg/ml). When bacteria were grown in the presence of Fe2+, the effect of ALA in the activity of GluRS1 was dose-dependent (Fig. 2), suggesting that any of the products of ALA utilization, including heme, might be responsible for the reduced activity of GluRS1 in the extracts.

Fig. 2.
Effect of ALA in the endogenous activity of GluRS1. Cells were cultured in the presence of the indicated concentrations of ALA. Specific activity of GluRS1 in extracts from late exponentially growing cells was measured. The plot represents pmol of Glu-tRNA ...

The effect of ALA on the activity of five other synthetases (AspRS, TyrRS, HisRS, TrpRS, SerRS) was tested in A. ferrooxidans extracts by using total E. coli tRNA as substrate. In addition to GluRS1, only TyrRS was significantly affected (70% decrease) when the cells were cultured in the presence of the precursor of heme biosynthesis.

The steady-state level of both GluRS1 and GluRS2, measured by Western blot analysis, was slightly higher in cells grown in Fe2+ compared with S0 grown cells (Fig. 3A Top and Middle, respectively). The level of the GluRS1 and GluRS2 mRNAs was tested by RT-PCR. Whereas the mRNA encoding GluRS1 showed a slight increase in Fe2+ compared with S0, no variation in the mRNA of GluRS2 was observed (Fig. 4). These data suggest that a transcriptional mechanism might control the level of GluRS1 in cells grown in Fe2+ and/or S0. Similar results were obtained when the expression of GluTR at the level of protein or mRNA was measured (Figs. 3A Bottom and and4).4). Conversely, no significant variation of both GluRS1 and GluRS2 was observed when the cells were cultured in Fe2+ supplemented with increasing concentrations of ALA (Fig. 3B Top and Middle, respectively). Under these culture conditions, we observed a marked variability in the level of the mRNA encoding GluRS1 (from no mRNA to the same level as in cells cultured in Fe2+) measured by RT-PCR (data not shown). We believe that the presence of ALA in the culture medium induces an unknown event in the cells that led to instability of the mRNA during the isolation procedure. We concluded that the activity of GluRS1, rather than the steady-state level of the enzyme, is affected by the presence of ALA in the culture medium. Semiquantitative analysis by Western blot revealed a relatively low amount of GluRS2 (5–10% depending on the cell growth condition) compared with GluRS1. Thus, together these data indicate that GluRS1 is the major GluRS in A. ferrooxidans and suggest that it is mostly dedicated to heme biosynthesis because a reduction of the activity of GluRS1 had no effect on cell growth.

Fig. 3.
Inmunodetection of GluTR and GluRSs. Whole exponentially growing cells cultured in Fe2+ or S0 (A) or Fe2+ supplemented with various concentrations of ALA (B) were electrophoresed under denaturating conditions, transferred to a solid matrix, and probed ...
Fig. 4.
Level of mRNAs encoding GluRSs and GluTR. Total mRNA was extracted from exponentially growing cells cultured in Fe2+, Fe2+/ALA, or S0 and subjected to RT-PCR using specific primers for GluRS1, GluRS2, or GluTR. CyD is the representative amplification ...

Hemin Inhibits the Activity of GluRS1.

Because the intracellular heme content correlates with the activity of the endogenous GluRS1, it is possible that heme or an intermediate of its synthesis might affect the enzymatic activity. To test this hypothesis, the effect of hemin, an analog of heme containing Fe3+, was determined on the activity of each recombinant GluRS1 fused to GST (GST-GluRS1), recombinant GluRS1 released from GST (rGluRS1), or the endogenous GluRS1 (in extracts from cells grown in Fe2+). As shown in Fig. 5A, a dose-dependent inhibition by hemin of all tested GluRSs was observed. A recruitment of hemin by GST that is known to bind heme (28, 29) might explain the ≈10-fold higher sensitivity of GST-GluRS1 to the inhibitor. GluRS2 was also inhibited by hemin to a similar extent as GluRS1 (data not shown). Binding of hemin to GST-GluRS1, rGluRS1, and GST-GluRS2 at concentrations as low as 10 μM was observed (data not shown). These findings strongly suggest that, upon culture of A. ferrooxidans in the presence of ALA, the reduced activity of GluRS1 is the result of an interaction of the enzyme with heme or any other intermediate tetrrapyrrole (ALA added to the reaction mixture had no effect in the activity of GluRSs).

Fig. 5.
Inhibition of GluRS1 by hemin. (A) Relative activity of GST-GluRS1 ([filled triangle]), rGluRS1 (●), and endogenous GluRS1 (■) treated with variable concentrations of hemin is shown. Initial velocities without added hemin measured as Glu-tRNA ...

Heme causes oxidative damage to cellular macromolecules (23). Some damage can be reversed by general antioxidant systems that depend on NADPH or GSH (29, 30). We tested whether NADPH counteracts the inhibition of GluRS1 by hemin. GluRS1 was preincubated with hemin, then incubated with either NADPH or NADP+, and subsequently, the charging activity was measured. As shown in Fig. 5B, incubation of the enzyme in the presence of NADPH almost completely abolished the ability of hemin to inhibit GluRS1. This effect was not observed when NADP+ was added, suggesting that a role of NADPH as an antioxidant (30) is involved in the protection against the inhibition by hemin. Interestingly, the presence of GSH in the Fe2+/ALA culture medium of A. ferrooxidans led to a slight decrease in the inhibition of GluRS1 (data not shown). These data are consistent with the notion that a redox reaction is involved in the inhibition of GluRS1 by hemin. It is not clear, however, if the effect is a result of the tetrapyrrole or the Fe3+ present in hemin.

Heme Regulates the Level of GluTR.

Regulation of GluTR at the protein level or the enzymatic activity is the major mechanism to control the heme status in cells that use the C5 pathway (1517). To test whether there is also such a mechanism in A. ferrooxidans, we inspected the cellular levels of the enzyme by Western blot analysis. The relative amount of GluTR was significantly higher in the cells grown in Fe2+ compared with S0 grown cells (Fig. 3A Lower). The hemA mRNA (encoding GluTR) was higher in Fe2+ compared with S0 (Fig. 4), suggesting a transcriptional control of the expression under these culture conditions as was observed for GluRS1. However, when the culture medium contained Fe2+/ALA, the concentration of GluTR decreased to almost undetectable levels (Fig. 3B Lower). The steady state of hemA mRNA showed the same level as in cells grown in Fe2+ (Fig. 4), suggesting that with an excess of heme (Fe2+/ALA) the amount of GluTR is controlled at the posttranscriptional level by a proteolytic mechanism. As in other organisms (19), hemin binds and inhibits the activity of recombinant GluTR at concentrations as low as 5 μM (data not shown). Hence, these data suggest that heme may be an intracellular effector that controls the function of GluTR at the transcriptional level or by specific degradation depending on the heme status, as observed in other organisms (15, 18, 19). Taken together, these results reveal that in A. ferrooxidans the function of both GluRS1 and GluTR is regulated by mechanisms that involve the steady-state level of the proteins or the activity of the enzymes in response to the cellular heme status.


Regulation of Heme Biosynthesis.

Insertion of iron into protoporphyrin IX is the final step in the biosynthesis of heme. Accumulation of either iron (21) or protoporphyrin is toxic to cells and leads to oxidative stress (31). These effects imply that a fine balance between these two components is required to avoid cell damage (32). Upon induction of cytochrome c synthesis in A. ferrooxidans grown in Fe2+, an increased heme demand must be fulfilled. Cells respond to heme requirements in part by modulating the level of GluTR, the first enzyme committed to the production of ALA. Because the steady state of hemA mRNA is higher in these cells, it is likely that a transcriptional mechanism accounts for the different level of GluTR. Conversely, in cells grown in Fe2+/ALA, a higher amount of heme is produced that parallels a dramatic decrease in the steady state of GluTR. Probably the excessive heme produced triggers the proteolytic degradation of GluTR, as described in Salmonella (15). Other alternatives, including the control of translation of the mRNA by antisense RNA, cannot be ruled out.

A dual response in both the activity and the steady-state level of GluRS1 that correlates with the heme status is described here. With elevated heme requirements (cells grown in Fe2+), the level of GluRS1 increases probably through an increase in transcription. Conversely, with excess heme synthesis (cells grown in Fe2+/ALA), a dramatic reduction in the intracellular specific activity of GluRS1 occurs without any change in the protein level of this enzyme. Thus, upon increase of heme triggered by growth conditions (S < Fe2+ < Fe2+/ALA), the activity of GluRS1 correspondingly decreases. Whether this is the result of a more general cellular response, such as oxidative stress caused by elevated heme, needs to be investigated. Other synthetases (except TyrRS) did not respond in the same way. Because hemin binds and inhibits GluRS, we believe that heme or any other precursor tetrapyrrole is the intracellular effector that triggers this regulatory mechanism.

Protection against the inhibition of GluRS1 by NADPH led us to consider that the effect of heme on the activity of this enzyme might be the result of a redox process. Interestingly, E. coli GluRS is substrate of DsbA, a protein involved in the restoration of the reduced state of cysteines in proteins upon oxidation (33). Several cysteines, equivalent to the SWIM domain (34) of E. coli GluRS (35) involved in zinc binding, are found in the catalytic domain of A. ferrooxidans GluRS1. Whether these cysteines are sensitive to cellular oxidative stress caused by elevated heme content is not known.

Regulation of the GluRS activity by the interaction with other proteins has been described. Bacillus subtilis GluRS copurifies with the adenosuccinate lyase, and the complex formed increased the thermostability of GluRS and decreased the KM for glutamate and ATP (36). Also, the formation of a ternary complex between GluRS, MetRS, and the Arc1p protein has been described in Saccharomyces cerevisiae. The interaction of Arc1p with GluRS increased the binding affinity of the enzyme for tRNA by 100-fold (37). An interaction between GluRS and GluTR might occur in A. ferrooxidans like has been observed in Chlamydomonas reinhardtii (38). We hypothesize that GluRS1 and GluTR interact in the cell and that the decrease in the level of GluTR upon growing the cells in Fe2+/ALA might expose the GluRS to heme or any other effector that reduces the activity of the enzyme.

Is GluRS1 Dispensable for Protein Biosynthesis?

Because of the high demand of heme in cells grown in ferrous ions, a deviation to heme biosynthesis could result in shortening of Glu-tRNA for protein biosynthesis. In the case of A. ferrooxidans, the existence of three tRNAGlu species and two GluRS enzymes can be interpreted as an evolutionary adaptation, rather than redundancy, to the requirements of Glu-tRNA to produce high levels of heme. Under growth conditions in which cells do not require Glu-tRNA, as precursor for heme biosynthesis, up to 85% of GluRS1 is dispensable, but no major detrimental effect in the cell growth was observed. Thus, GluRS2 and the remaining 15% of the activity of GluRS1 are sufficient to provide the Glu-tRNA substrates for protein synthesis. In addition, the formation by GluRS1 of Glu-tRNAGlu3, which is not a substrate of GluTR (26), might ensure the provision of Glu-tRNA for protein biosynthesis. Thus, these data suggest that most of the activity of GluRS1 is committed to heme biosynthesis, and protein biosynthesis might be a secondary role of this enzyme.

Duplication of GluRS has been proposed as an evolutionary event to ensure the specific decoding of glutamine codons during translation of the genetic information (5, 39). The data presented here suggest that a physiological role other than protein synthesis might explain the duplication of GluRS. In the same vein, an interesting hypothesis to explain the conservation of a reduced plastid genome in nonphotosynthetic plants and algae has been proposed. These organisms also have two GluRSs (one cytoplasmic and one organellar) and the plastid-encoded tRNAGlu might be essential for heme biosynthesis (40). Thus, complex Glu-tRNA formation systems might have evolved to ensure the provision of heme.

Materials and Methods

Strains and Cell Growth Conditions.

A. ferrooxidans ATCC 23270 was cultured at 30°C in 9K medium (41) supplemented with 33.3 g/liter Fe2SO4 at pH 1.6 or 5 g/liter S0 at pH 3.5. When necessary ferrous medium was supplemented with 50–200 μg/ml of ALA. E. coli DH5, used for plasmid propagation, was grown at 37°C in LB medium. E. coli BL21 (DE3) was used for overexpression and purification of proteins. When necessary, the LB medium was supplemented with 100 μg/ml ampicillin.

DNA Manipulations.

General DNA manipulations were carried out according to standard procedures (42). Plasmid DNAs were obtained by using the Qiagen (Valencia, CA) extraction kit. Cloned DNA was sequenced at the Centro de Biotecnología, Facultad de Ciencias, Universidad de Chile. The preliminary sequence of A. ferrooxidans ATCC23270 genome was obtained from The Institute for Genomic Research (www.tigr.org).

Expression and Purification of Recombinant Proteins.

Overproduction and purification of the recombinant proteins GST-GluRS1 and His(6)-GluTR was carried out as described (27). When required GST-GluRS1 purified from E. coli was digested with thrombin (Sigma, St. Louis, MO) according to the manufacturer's instructions to remove the GST. GST was overexpressed from pGEX2T vector (Amersham, Piscataway, NJ) and purified according to the manufacturer's instructions. His(6)-fGluRS2 was obtained by cloning of the nucleotides encoding the 190 carboxyl-terminal residues of GluRS2 into pET15b. Overexpression and purification from inclusion bodies was performed as described (27).

Production of Specific Antisera.

His(6)-GluTR, GST-GluRS1, and His(6)-fGluRS2 were electrophoresed in SDS/PAGE, detected by staining with Coomassie blue, and excised from the gel. Gel strips were crushed and used to immunize rabbits (BiosChile, Santiago, Chile). Polyclonal antibodies obtained against GluTR were further purified by affinity chromatography on immobilized GluTR-agarose as described (43).

Western Blot Analysis.

Samples were heated to 100°C for 10 min before gel loading. Protein samples were separated on 12% SDS/PAGE Tris-glycine gels and transferred to nitrocellulose membranes. Membranes were blocked with TBS (Tris-borate-NaCl) containing 5% nonfat dry milk and subjected to Western blot analysis by standard procedure. Polyclonal antibodies against GluRS1, fGluRS2, or GluTR were used at 1:10,000 dilutions. HRP-conjugated secondary antibodies (Pierce, Rockford, IL) were used at 1:30,000 dilution. Reacting protein were visualized by using the SuperSignal West Pico Chemiluminescent detection kit (Pierce).

Analysis of mRNA Expression.

Total RNA from A. ferrooxidans was prepared according to Hangen and Young (44) and reverse-transcribed with the primers pGI-INR (5′-aagtttcctgatcgccatgg-3′), pGII (5′-gaaccgcagggtaggcaataac-3′), or pGluTRAf6 (5′-tgcgcaacaccttccacctc-3′) for mRNAs encoding gltX1 (GluRS1), gltX2 (GluRS2), or hemA (GluTR), respectively and reverse-transcribed with RevertAid M-MuLV (Fermentas, Vilnius, Lithuania). PCR amplifications of gltX1, gltX2, and hemA cDNAs were carried out with primers pGI-INR and pGI-INF (5′-attgctattgcagtcgggaag-3′), pGII-INR and pGII-INF (5′-gaaccgcagggtaggcaataac-3′), or pGluTRAf6 and pGluTRAF8 (5′-aacatcaggcagcataggcg-3′), respectively. PCRs were carried out by using conventional procedures. The primers cydA1 (5′-aagctggcggccatggaagc-3′) and cydA4 (5′-actgcatggccaggaaaccg-3′) were used to amplify the mRNA encoding cydA, a constitutively transcribed cytochrome, as loading control.

Preparation of tRNA.

In vivo-expressed tRNAs [A. ferrooxidans's tRNAGln3 (UUG) and E. coli total tRNA] were obtained as described (5). tRNAGlu2 (CUC) transcript was synthesized by in vitro transcription (27).

Detection of in Vivo Aminoacylated/Deacylated tRNA.

To determine the level of aminoacylated/deacylated tRNAs in vivo total tRNA was extracted from A. ferrooxidans as described (5). Briefly, periodate oxidation followed by β-elimination, electrophoresis on denaturing gels, and subsequent Northern blot were carried out as described (6). Northern blot probes were designed complementary to nucleotides 15–39 for tRNAGlu1, nucleotides 14–37 for tRNAGln2, and nucleotides 13–37 for tRNAGln3.

Preparation of Cellular Extracts from A. ferrooxidans.

Cells were harvested, washed with 10 mM H2SO4 for Fe3+ removal, and subsequently washed with 30 mM Hepes, pH 8.0. The bacterial pellet was resuspended in lysis buffer (30 mM Hepes, pH 8.0/150 mM NaCl/1 mM DTT), and cells were disrupted by cycles of freezing at −78°C and thawing by sonication. Extracts were centrifuged at 20,000 × g for 30 min. The supernatant was ultracentrifuged at 150,000 × g for 90 min. When necessary the cellular extract was dialyzed against lysis buffer containing 50% glycerol. Protein concentration was measured with Bradford reagent, and aliquots were stored at −80°C.

Determination of GluRS Activity.

GluRS activity was determined according to Salazar et al. (5). Enzymatic activity was measured in 100 μl of reaction mixture, and the results are expressed as pmol of Glu-tRNA formed in 15-min incubation (initial velocity) relative to protein amount. Inhibition of activity by hemin was carried out by 15-min preincubation at 37°C. For the reversion assay, inhibited GluRS1 was further incubated with the corresponding reagent for 30 min at 37°C. The reaction was started by the addition of 0.2 μg/μl transcript tRNAGlu2 or 1 μg/μl total E. coli tRNA. Transcript tRNAGlu2 was previously denatured at 90°C for 3 min and cooled down slowly to facilitate RNA folding. The determination of the activity of GluRS1 from cellular extracts was carried out as described (5) with 0.1 μg of the cellular extract.

Heme Quantification.

Cellular heme was quantified as described (45) with minor modifications. Harvested A. ferrooxidans (washed with 10 mM H2SO4) and E. coli cells were washed with 30 mM Hepes, pH 8.0. Thirty milligrams of the bacterial pellet was resuspended in 90 μl of a solution containing 0.5 M NaOH and 2.5% Triton X-100. Disrupted cells were sonicated to reduce the viscosity of the medium and centrifuged at 20,000 × g for 30 min. Formed haemetine was measured spectrophotometrically by the absorbance at 575 nm. Bovine hemin (Sigma) was used as standard for heme measurements (39).


We thank Dr. Dieter Söll (Yale University, New Haven, CT) in whose laboratory part of this work was carried out and Dr. C. C. Allende, E. Jedlicki, and O. León for critical review of the manuscript. This work was supported by Fondo Nacional de Desarrollo Científico y Tecnológico Grant 1020087 (to O.O.), the Direccion de Investigacion and Instituto de Ciencias Biomédicas, Universidad de Chile (O.O.), and a grant from the National Institute of General Medical Sciences (to Dieter Söll). A.K. and M.d.A. are recipients of graduate studies fellowships from Comision Nacional de Investigacion Cientifica y Tecnologica, Chile and Deustchen Akademische Austauschdienst, respectively.


Glu-tRNA synthetase
Glu-tRNA reductase
amino levulinic acid.


The authors declare no conflict of interest.


1. Ibba M, Söll D. Genes Dev. 2004;8:731–738. [PubMed]
2. Park SG, Ewalt KL, Kim S. Trends Biochem Sci. 2005;30:569–574. [PubMed]
3. Ryckelynck M, Giege R, Frugier M. Biochimie. 2005;87:835–845. [PubMed]
4. Feng L, Tumbula-Hansen D, Min B, Namgoong S, Salazar J, Orellana O, Söll D. In: Aminoacyl-tRNA Synthetases. Ibba M, Cusack S, editors. Georgetown, TX: Landes; 2005. pp. 314–319.
5. Salazar JC, Ahel I, Orellana O, Tumbula-Hansen D, Krieger R, Daniels L, Söll D. Proc Natl Acad Sci USA. 2003;100:13863–13868. [PMC free article] [PubMed]
6. Salazar JC, Ambrogelly A, Crain PF, McCloskey JA, Söll D. Proc Natl Acad Sci USA. 2004;101:7536–7541. [PMC free article] [PubMed]
7. Blaise M, Becker HD, Lapointe J, Cambillau C, Giege R, Kern D. Biochimie. 2005;87:847–861. [PubMed]
8. Beale SI, Gough SP, Granick S. Proc Natl Acad Sci USA. 1975;72:2719–2723. [PMC free article] [PubMed]
9. Schön A, Krupp G, Gough S, Berry-Lowe S, Kannangara CG, Söll D. Nature. 1986;322:281–284. [PubMed]
10. Jahn D, Verkamp E, Söll D. Trends Biochem Sci. 1992;17:215–218. [PubMed]
11. Schauer S, Chaturvedi S, Randau L, Moser J, Kitabatake M, Lorenz S, Verkamp E, Schubert WD, Nakayashiki T, Murai M, et al. J Biol Chem. 2002;277:48657–48663. [PubMed]
12. Smith MA, Kannangara CG, Grimm B. Biochemistry. 1992;31:11249–11254. [PubMed]
13. Ilag LL, Jahn D. Biochemistry. 1992;31:7143–7151. [PubMed]
14. Beale S. In: Escherichia coli and Salmonella typhimurium. Neilhardt FC, editor. Washington, DC: Am Soc Microbiol; 1996. pp. 731–748.
15. Wang L, Elliott M, Elliott T. J Bacteriol. 1999;181:1211–1219. [PMC free article] [PubMed]
16. Goslings D, Meskauskiene R, Kim C, Lee KP, Nater M, Apel K. Plant J. 2004;40:957–967. [PubMed]
17. Falciatore A, Merendino L, Barneche F, Ceol M, Meskauskiene R, Apel K, Rochaix JD. Genes Dev. 2005;19:176–187. [PMC free article] [PubMed]
18. Krieger R, Rompf A, Schobert M, Jahn D. Mol Genet Genomics. 2002;267:409–417. [PubMed]
19. Vothknecht U, Kannangara CG, von Wettstein D. Phytochemistry. 1998;47:513–519. [PubMed]
20. Thöny-Meyer L. Microbiol Mol Biol Rev. 1997;61:337–376. [PMC free article] [PubMed]
21. Andrews SC, Robinson AK, Rodriguez-Quinones F. FEMS Microbiol Rev. 2003;27:215–237. [PubMed]
22. Galperin M. Environ Microbiol. 2004;6:552–567. [PMC free article] [PubMed]
23. Graca-Souza AV, Maya-Monteiro C, Paiva-Silva GO, Braz GR, Paes MC, Sorgine MH, Oliveira MF, Oliveira PL. Insect Biochem Mol Biol. 2006;36:322–335. [PubMed]
24. Ingledew WJ. Biochim Biophys Acta. 1982;683:89–117. [PubMed]
25. Yarzabal A, Brasseur G, Bonnefoy V. FEMS Microbiol Lett. 2002;209:189–195. [PubMed]
26. Núñez H, Lefimil C, Min B, Söll D, Orellana O. FEBS Lett. 2004;557:133–135. [PubMed]
27. Levicán G, Katz A, Valezuela P, Orellana O. FEBS Lett. 2005;579:6383–6387. [PubMed]
28. Arias IM. Med Biol. 1979;57:328–334. [PubMed]
29. Vincent SH, Grady RW, Shaklai N, Zinder JM, Muller-Eberhard U. Arch Biochem Biophys. 1988;265:539–550. [PubMed]
30. Kirsch M, De Groot H. FASEB J. 2001;15:1569–1574. [PubMed]
31. Winyard PG, Moody CJ, Jacob C. Trends Biochem Sci. 2005;30:453–461. [PubMed]
32. Stojiljkovic I, Perkins-Balding D. DNA Cell Biol. 2002;21:281–295. [PubMed]
33. Leichert LI, Jacob U. PLoS Biol. 2004;2:e333. [PMC free article] [PubMed]
34. Makarova KS, Aravind L, Koonin EV. Trends Biochem Sci. 2002;27:384–386. [PubMed]
35. Banerjee R, Dubois DY, Gauthier J, Lin SX, Roy S, Lapointe J. Eur J Biochem. 2004;271:724–733. [PubMed]
36. Gendron N, Breton R, Champagne N, Lapointe J. Proc Natl Acad Sci USA. 1992;89:5389–5392. [PMC free article] [PubMed]
37. Graindorge JS, Senger B, Tritch D, Simos G, Fasiolo F. Biochemistry. 2005;44:1344–1352. [PubMed]
38. Jahn D. FEBS Lett. 1992;314:77–80. [PubMed]
39. Skouloubris S, Ribas de Pouplana L, De Reuse H, Hendrickson TL. Proc Natl Acad Sci USA. 2003;100:11297–11302. [PMC free article] [PubMed]
40. Barbrook AC, Howe CJ, Purton S. Trends Plant Sci. 2006;11:101–108. [PubMed]
41. Silverman MP, Lundgren DG. J Bacteriol. 1959;77:642–647. [PMC free article] [PubMed]
42. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor, NY: Cold Spring Harbor Lab Press; 1989.
43. Salazar JC, Zúñiga R, Lefimil C, Söll D, Orellana O. FEBS Lett. 2001;491:257–260. [PubMed]
44. Hangen GS, Young E. J Virol. 1978;26:793–804. [PMC free article] [PubMed]
45. Zander R, Lang W, Wolf U. Clin Chem Acta. 1984;136:83–93. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...