|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
J. Biol. Chem., Vol. 281, Issue 35, 25270-25277, September 1, 2006
An RNA-binding Respiratory Component Mediates Import of Type II tRNAs into Leishmania Mitochondria*
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
The precise manner in which polyanionic tRNA molecules cross the double mitochondrial membrane, is largely unresolved. In yeast, the mitochondrial protein import pore, as well as cytosolic carrier proteins, appears to be involved (5, 6). On the other hand, biochemical studies in kinetoplastid protozoa have revealed that these organisms have specialized mechanisms for import of tRNA that are distinct from those for protein import (7) but involve direct interaction of tRNAs with membrane-bound proteins (8). Leishmania and Trypanosoma mitochondria recognize sequence/structure motifs (import signals) in distinct domains of individual tRNAs (9, 10), and the former rapidly select oligoribonucleotide-containing motifs matching those in importable tRNAs from a random sequence pool of high complexity (11).
However, there are differences in the intrinsic efficiencies of transfer of individual RNAs through the inner, as opposed to the outer membrane, in vitro; some, designated as type I RNAs, are imported efficiently into the matrix, whereas others (type II) are not. The potential problem of an imbalanced matrix tRNA pool is solved by the unique phenomenon of allosteric regulation; type I RNAs stimulate the inner membrane transfer of type II RNAs, whereas type II RNAs inhibit transfer of type I (11, 12). On the basis of these interactions, tRNATyr, tRNAArg, and tRNATrp have been identified as type I, and tRNAIle, tRNAVal, and tRNAMet-e have been identified as type II, respectively (1113); in vitro evolution experiments suggest that many more tRNA species belong to these categories (11). The "ping-pong" model (14) postulates that the two types of tRNA bind to different receptors; binding of type I tRNA to its receptor induces a conformational change that is transmitted to the type II receptor, opening up its tRNA-binding site. Type II tRNA loading in turn induces an allosteric transition resulting in the destabilization of the type I complex.
Recently, a multi-protein complex (the RNA import complex (RIC))2 that induces translocation of tRNAs through artificial or biological membranes was isolated from Leishmania inner membrane (1416). In a reconstituted liposome system, RIC retains the characteristic type I-type II interactions (14). Two tRNA-binding proteins with the properties of type I and type II receptors are associated with this complex (14). We have recently identified the type I receptor and showed that, in vivo, it is required for import of type I as well as type II tRNAs (13). We now report the identification of a subunit of this complex that binds type II tRNAs and is essential for the import of these tRNAs in vivo.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Cloning of RIC8A GeneSense and antisense primers (supplemental Table S1) corresponding to the open reading frame of RIC8A were used to amplify the intact or truncated gene from L. tropica strain UR6 genomic DNA. The sense primer was designed to have a BamHI restriction site at the 5' end in frame with the coding sequence of the insert; the antisense primer had an inframe stop codon followed by a SalI site. The amplified fragment was cloned into TA vector pTZ57R (MBI Fermentas). Other genes were cloned and expressed similarly (13).
Preparation of Recombinant Proteins and AntibodiesThe RIC8A open reading frame was transferred as a BamHI-EcoRI fragment from pTZ57R to the pGEX4T-1 vector (Amersham Biosciences) downstream of, and in-frame with, the glutathione S-transferase gene. Recombinant plasmids were expressed in Escherichia coli BL21, and fusion protein was extracted from inclusion bodies and digested with thrombin, and the recombinant protein was recovered by gel electrophoresis, as described (13). To refold the protein, the soluble SDS extract was diluted 5-fold into TETN250 buffer (14) containing 0.1% Triton X-100 and incubated for 2 h at 4°C before assay. Electrophoretically pure recombinant protein was used to raise polyclonal antibody in BALB/c mice.
Blue Native (BN) PAGEInner membrane mitochondrial complexes of L. tropica or the human HepG2 cell line were resolved as previously described (13). Briefly, mitochondria were extracted with BAM buffer (50 mM BisTris-HCl, pH 7.0, 0.75 M
-aminocaproic acid, 2% dodecyl maltoside) for 45 min at 4 °C, and the extract was concentrated to
10 µl/200 µgof mitochondria by centrifugal ultrafiltration in a Microcon 30 unit (Amicon). Coomassie Blue G-250 (0.5%) was added to the extract before electrophoresis on 6% Blue Native gels (21). For two-dimensional analysis, protein bands of the first dimension were denatured with 0.125 M Tris-HCl, pH 6.8, 1% SDS, 1%
-mercaptoethanol for 40 min at 37 °C and then subjected to SDS-PAGE.
Preparation of Radiolabeled tRNA32P-Labeled tRNAs of high and low specific activity were prepared by T7 polymerase-mediated run-off transcription as described (11, 13).
Import AssaysPurified RIC (100 ng) was incorporated into phosphatidylcholine vesicles (50 µg lipid) and incubated with 32P-labeled tRNA (5 nM) and 4 mM ATP, and uptake was analyzed by RNase protection, as previously described (14). For immuno-inhibition experiments, proteoliposomes were preincubated with antiserum (1:50) for 30 min on ice. Where indicated, low specific activity effector tRNAs were present at one-tenth the concentration of the high specific activity substrate tRNA.
Western and Northwestern BlotsNative complexes resolved by BN PAGE were denatured in situ before blot transfer, as described (13). The blots were probed with 1:100 dilution of antiserum and developed by the alkaline phosphatase colorimetric method. For Northwestern blots, the membrane was probed with 32P-labeled tRNA (17).
Binding AssaysIndicated amounts of purified, recombinant refolded RIC8A were incubated with 32P-labeled tRNAIle (10100 fmol) in 10-µl reactions containing binding buffer (10 mM Tris-HCl, pH 7.5, 10 mM MgCl2, 2 mM dithiothreitol, 0.1 M KCl) for 30 min on ice and then electrophoresed on native gradient polyacrylamide gels (14) to resolve the ribonucleoprotein from free tRNA. Dried gel bands were quantified by liquid scintillation counting. Liposome binding assays were carried out under similar conditions with 250 ng of purified RIC incorporated into 80 µg of lipid and 1250 fmol of tRNAIle, the vesicles were washed, and bound RNA was recovered for electrophoresis. Scatchard analysis was carried out by titrating a fixed amount of RIC8A (50 fmol) or liposome-bound RIC with tRNAIle. The plot of bound/free tRNA against the bound tRNA concentration yields a best fit curve with a slope of 1/Kd, where Kd is the dissociation constant for the complex. The total receptor concentration, [R]0, is the intercept on the x axis.
Photochemical Cross-linkingT7 RNA polymerase transcripts were doubly labeled with [
-32P]UTP and 5-bromo-UTP and cross-linked to protein as previously described (14). Briefly, RNA was incubated with affinity-purified RIC or mitochondrial complexes and then UV-irradiated. Cross-linked RNA·protein complexes were immunoprecipitated after dissociation of the subunits with SDS and resolved by urea-PAGE (14).
Northern BlottingTotal promastigote RNA was electrophoresed on a 5% acrylamide, 8 M urea gel, electroblotted on to a Hybond N+ membrane, and probed with radiolabeled RIC8A coding region.
Conditional KnockdownDetails of the construction of L. tropica 1390, a host containing constitutively expressed T7 RNA polymerase and Tet repressor genes, and of the targeting vector pGET, have been described (13). Knockdown vector pGET(AS)RIC8A was constructed by inserting the RIC8A gene between the HindIII and BamHI sites of pGET (i.e. in the reverse orientation with respect to the inducible T7 promoter). L. tropica 1390 was transfected with pGET(AS)RIC8A and transformants selected on semi-solid agar containing G418, hygromycin, and 2.5 µg/ml phleomycin. The clones were grown in medium 199 containing the same antibiotics. The cultures were induced with 1 µg/ml tetracycline, and cell growth was monitored. Intracellular parameters were measured at 48 h, the time point at which cessation of growth was first observed.
RT-PCRUninduced or induced cells were harvested, lysed, and separated into soluble (cytosolic) and particulate (mitochondrial) fractions (9). The crude mitochondria were treated with DNase and RNase before RNA isolation. RNA from 102105 cells was denatured at 95 °C and reverse transcribed with Superscript II (Invitrogen) and the appropriate primer, as follows: (1) for antisense RNA, the sense primer from the 5' end of the coding region of RIC8A; (2) for mRNA or tRNA, the antisense primer complementary to the 3' end of the gene (supplemental Table S1). The second primer was then added, and the cDNA was amplified with Taq DNA polymerase. To obtain proportional PCR signals, RT-PCR was performed with different amounts of input RNA.
Import ReconstitutionLiposomes were reconstituted with mitochondrial extracts from knockdown cells and recombinant import factor as described (13). Briefly, refolded RIC8A (8 ng in 2 µl) and concentrated mitochondrial extract (5 x 106 cell equivalent in 8 µl) were incubated with liposomes (50 µg of lipid in 10 µl) for 1 h at 4°C and then assayed for import as above.
RNA End LabelingTotal mitochondrial tRNA was dephosphorylated with shrimp alkaline phosphatase, ethanol-precipitated, and 5'-labeled with [
-32P] ATP in the presence of T4 polynucleotide kinase.
Mitochondrial AssaysMitochondrial translation assays by [35S]methionine labeling of promastogotes in presence of cycloheximide, oxygen uptake measurements, and cytochrome oxidase cytochemical assays were performed as described (13).
| RESULTS |
|---|
|
|
|---|
|
-helices, one of which includes a periodic repeat of basic amino acids (between residues 42 and 61); and 2) a short cleaved mitochondrial targeting sequence, with the mature N terminus at Met-10 (as indicated by peptide sequencing).
|
Type II tRNA Binding by RIC8A/UCR6bTo assess the tRNA binding activity of RIC8A, Western blots of bacterially expressed, gel-purified RIC8A, after suitable renaturation treatments in situ, were probed with radiolabeled tRNAs. Under these conditions, RIC8A interacted with tRNAIle, tRNAVal, and tRNAMet-e (all type II tRNAs), but with none of the type I tRNAs (Fig. 2A). The tRNA·protein complex was detected by gel shift assays using purified RIC8A, which confirmed the specificity for type II tRNA (Fig. 2B). Scatchard analysis (Fig. 2C) yielded a dissociation constant (Kd) of 0.42 nM at 4 °C for the tRNAIle·RIC8A complex (Table 1). Thus, RIC8A has an intrinsic affinity for type II tRNAs.
|
Presence of RIC8A/UCR6b in Two Mitochondrial ComplexesIn view of the facts that RIC8A/UCR6b is the product of a single gene, has structural similarity with UCR6b, binds tRNA, and is associated with the import complex, it is possible that RIC8A/UCR6b is a bi-functional protein with roles in both tRNA import and electron transport. To address this question, we resolved the mitochondrial inner membrane respiratory complexes of Leishmania by Blue Native gel electrophoresis (21) (Fig. 3A, left panel). As observed by us and others in the kinetoplastid protozoa Leishmania, Crithidia, and Phytomonas, (13, 2226), four complexes were discernible, corresponding to Complexes III, IV, and V and the largest complex, specific to kinetoplastid protozoa, that has been identified as the import complex (13). Immunoblotting of the resolved complexes revealed the presence of RIC8A/UCR6b in RIC and Complex III (Fig. 3A, right panel). The subunits of complex III and RIC were separated by denaturing electrophoresis in the second dimension. Most of the subunits in the two complexes are distinct, but Western blot analysis showed that the same 21-kDa RIC8A/UCR6b is shared by both (Fig. 3B), as expected of a bi-functional protein.
Because RIC8A/UCR6b is shared by RIC and Complex III, we inquired whether both complexes can bind tRNA. When the mixture of native respiratory complexes obtained by detergent extraction of mitochondria was exposed to tRNAIle (type II) doubly labeled with 32P and 5-bromouridine (BU), a photoactivable nucleoside analogue, and then UV-irradiated, only RIC among the mitochondrial complexes was tagged; cross-linking required the presence of type I tRNA (Fig. 3C). In contrast, none of the human respiratory complexes interacted with tRNA under any condition (Fig. 3C). After dissociation of the complex formed with RIC, the single RNA-protein adduct could be resolved by denaturing PAGE and was immunoprecipitated with anti-RIC8A/UCR6b antibody; the antibody, if present during the tRNA binding reaction, prevented formation of the adduct (Fig. 3D). Identical results were obtained with BU-labeled tRNAVal and tRNAMet, two other type II tRNAs; nonimmune serum, or antibody against other RIC subunits, failed to immunoprecipitate this adduct.3 Thus, RIC8A/UCR6b is the major, if not the sole, component of the import complex that directly interacts with type II tRNAs.
The allosteric activation of binding of the type II tRNAIle by type I tRNATyr was quantified by binding assays on the liposome-bound import complex. This showed that the type I tRNATyr increases the affinity, i.e. lowers the Kd, of tRNAIle for the complex by a factor of 5 (Table 1). Additionally, there is a 10-fold increase in the effective RIC8A concentration in the presence of the effector (Table 1), indicating enhanced exposure of the tRNA-binding site.
|
(Ref. 13 and Fig. 4B), attesting to the specificity of the targeting procedure. Moreover, the mitochondrial levels of proteins other than RIC8A/UCR6b were unaffected, and inner membrane proteins such as Complex II iron-sulfur protein remained resistant to trypsin in knockdown mitochondria until detergent disruption (Fig. 4B), indicating that the knockdown protocol did not affect the intactness of the organelle or the mitochondrial targeting of nucleus-encoded proteins.
In the absence of tetracycline, the anti-RIC8A/UCR6b transformants grew normally with a doubling time of
24 h at 22 °C, which is close to the doubling time for cells transfected with the empty vector (Fig. 4C). The addition of tetracycline resulted in the cessation of growth from
48 h (Fig. 4C); promastigotes became immotile, rounded or flask-shaped, and nonviable. Withdrawal of tetracycline after 72 h resulted in resumption of growth after a lag of 34 days, suggesting that the knockdown is reversible or that a few cells escape knockdown.
To check the effect of the knockdown on import, cytoplasmic and mitochondrial tRNAs of the induced cells were examined. There was an overall decline in the amount of mitochondrial tRNAs, although some tRNAs persisted (Fig. 4D). At the level of individual tRNA species, cytosolic levels were not appreciably altered, but mitochondrial tRNAIle(UAU), tRNAVal (CAC), and tRNAMet-e(CAU) (all type II tRNAs) were reduced to less than 10% of normal by 48 h (Fig. 4E). In contrast, there was an increase of 5070% in the amount of mitochondrial tRNATyr(GUA), tRNAArg(ACG), or tRNATrp(CCA) (all type I tRNAs) (Fig. 4E). This slight increase in signal intensity was maintained at different levels of input RNA in the RT-PCR assay (Fig. 4F). Thus, in vivo knockdown of RIC8A/UCR6b results in depletion of the mitochondrial type II tRNAs and a slight elevation of mitochondrial type I tRNAs.
In addition to the loss of mitochondrial tRNAs, severe pleiotropic effects on mitochondrial protein synthesis and the structure and function of respiratory complexes were observed upon knockdown of RIC8A/UCR6b. Although cytosolic protein synthesis was normal,3 chloramphenicol-sensitive mitochondrial translation was reduced to barely detectable levels (Fig. 4G). The number of cells positive for cytochrome oxidase (or Complex IV) was reduced to
16% of the uninduced value. There was simultaneously a reduction in the rate of O2 uptake from 1.6 to 0.7 fmol min1 cell1 at 30 °C. The decline in mitochondrial tRNA import in vivo upon RIC8A/UCR6b knockdown could be due to this respiration defect, a defect in the import apparatus, or a combination of both factors. Although it is difficult to formally distinguish between these possibilities, it would be important to investigate whether the import function of RIC was affected by depletion of RIC8A/UCR6b.
|
The interaction of RIC' with tRNAs was examined by photoaffinity cross-linking experiments. Although the wild-type complex was cross-linked to BU-labeled tRNAIle (type II) in the presence of type I tRNATyr effector, the knockdown complex, lacking RIC8A/UCR6b, was not (Fig. 3F). Conversely, the wild-type complex interacted with BU-labeled tRNATyr, but this type I interaction was inhibited by tRNAIle (Fig. 3F), as previously observed with the affinity-purified complex (14). In contrast, RIC' interacted with tRNATyr (Fig. 3F), but this interaction was not inhibited by type II tRNAIle (Fig. 3F). Thus, knockdown of RIC8A/UCR6b resulted in the formation of a subcomplex in which both the binding of, and regulation by, type II tRNA were affected.
To determine whether the deficiency of RIC8A/UCR6b is sufficient to account for these defects, we tested mitochondrial extracts from normal or tetracycline-induced cells for functional reconstitution of import activity. When incorporated into liposomes, an extract from RIC8A knockdown cells was active in the import of type I, but not of type II tRNAs even in the presence of type I effector (Fig. 5A). Importantly, type II activity of the knockdown extract could be restored by adding back purified recombinant RIC8A/UCR6b and a type I tRNA to the reconstitution reaction (Fig. 5B); unrelated proteins such as bovine serum albumin were unable to substitute for RIC8A, and RIC8A/UCR6b alone was unable to protect the RNA from degradation by ribonuclease (Fig. 5B). Thus, the observed effects of RIC8A/UCR6b knockdown were directly attributable to a role of this protein in type II tRNA import.
|
| DISCUSSION |
|---|
|
|
|---|
How does RIC8A/UCR6b recognize multiple type II tRNAs? We showed previously that the variable arm-T loop region of tRNAIle contains a type II signal for transfer through the inner mitochondrial membrane (11, 12). Subsequently, the T stems of T. brucei tRNAIle and tRNAMet-e (which are identical in sequence to the corresponding Leishmania species) have been found to contain import determinants (10). The T stems of all three type II tRNAs used in this study have the sequence YGRGU, of which the second G (paired to a C in the stem) is apparently critical (10); moreover, two major sequence classes of type II import aptamers selected by in vitro evolution contain the motif UG34U (11); finally, none of the type I tRNAs studied here contains this exact sequence in the T arm. Thus, it is conceivable this or a related motif directly interacts with RIC8A/UCR6b.
The intrinsic affinity of free RIC8A/UCR6b for type II tRNA is high (Table 1) and comparable with that of the type I receptor RIC1 for its cognate tRNA (13). However, within the native complex, in the absence of type I tRNA, this affinity is much lower, presumably because of the association of RIC8A with other subunits. Thus, the consequence of allosteric activation is to increase both the affinity as well as the number of binding sites exposed to tRNA (Table 1).
Knockdown of RIC8A/UCR6b led to a small but reproducible enhancement in the import of all three type I tRNAs (Fig. 4). The inhibitory effect of individual type II tRNAs, tested at a fixed concentration, is generally high but somewhat variable, e.g. tRNAIle and tRNAVal are more effective than tRNAMet-e (supplemental Fig. S2), probably as a result of the different affinities of these tRNAs for RIC8A/UCR6b. Variable concentrations and affinities of the multitude of type II tRNAs encountered in vivo may account for the limited effect (5070% increase) of RIC8A/UCR6b knockdown on import of type I tRNAs.
The RNA import complex is an assemblage of known and unknown proteins, some of which, such as RIC8A, RIC1 (13), and others3 are also present in other respiratory complexes. The possibility of RIC being a nonspecific aggregate is ruled out by the following observations. 1) It has a definite subunit composition that is reproducible in different preparations. 2) Knockdown of a particular subunit has a subunit-specific effect on this composition; for example, the subunit profile of the RIC8A knockdown complex (Fig. 3E) is different from that of the complex in RIC1 knockdown cells3; such effects can be reasonably explained by a defined assembly pathway but not by nonspecific aggregation. 3) Specific import-related interactions occur between different subunits of the complex, e.g. between RIC1 and RIC8A, and respond in a predictable way to knockdown of the interacting partners (Fig. 3F).
Subunit UCR6b of respiratory Complex III is absent from prokaryotes. The eukaryotic protein is highly conserved and has no enzymatic activity but is essential for cell viability, apparently because of its role in assembly of the functional complex (27). The Leishmania homologue, RIC8A/UCR6b, the product of a single gene, is shared between Complex III and the RNA import complex (Fig. 3), suggesting it to be a bi-functional protein. RIC8A/UCR6b (and probably, the corresponding protein from other kinetoplastid protozoa with almost identical sequence) has a tRNA-binding site that is exposed in RIC but not in Complex III (Fig. 3), i.e. the assembly of the protein in the two complexes is different. This suggests that one event in the evolution of import function may have been the acquisition of a tRNA binding N-terminal domain by Leishmania UCR6b as well as additional protein-protein interaction domains/motifs for alternate assembly. The emerging view of the RNA import complex is that of an assemblage of subunits including several bi-functional respiratory proteins. This differs from the yeast tRNA import system, which involves translation and mitochondrial protein import components (5, 6). Thus, different ancient metabolic pathways may have been the origin of the independent evolution of tRNA import in different species.
| FOOTNOTES |
|---|
The on-line version of this article (available at http://www.jbc.org) contains supplemental Table S1 and supplemental Figs. S1 and S2. ![]()
1 To whom correspondence should be addressed: Genetic Engineering Laboratory, Indian Institute of Chemical Biology, 4 Raja S.C. Mullick Road, Calcutta 700032, India. Tel.: 91-33-2473-3491 (ext. 136); Fax: 91-33-2473-5197; E-mail: sadhya{at}iicb.res.in.
2 The abbreviations used are: RIC, RNA import complex; Tet, tetracycline; BN, Blue Native; RT, reverse transcription; ORF, open reading frame; UCR, ubiquinol cytochrome c reductase; BU, 5-bromouridine. ![]()
3 S. Chatterjee and S. Adhya, unpublished data. ![]()
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Zikova, A. K. Panigrahi, A. D. Uboldi, R. A. Dalley, E. Handman, and K. Stuart Structural and Functional Association of Trypanosoma brucei MIX Protein with Cytochrome c Oxidase Complex Eukaryot. Cell, November 1, 2008; 7(11): 1994 - 2003. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Home, S. Mukherjee, and S. Adhya A mosaic of RNA binding and protein interaction motifs in a bifunctional mitochondrial tRNA import factor from Leishmania tropica Nucleic Acids Res., October 1, 2008; 36(17): 5552 - 5561. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Basu, S. Mukherjee, and S. Adhya Proton-guided movements of tRNA within the Leishmania mitochondrial RNA import complex Nucleic Acids Res., March 1, 2008; 36(5): 1599 - 1609. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Salinas, A.-M. Duchene, L. Delage, S. Nilsson, E. Glaser, M. Zaepfel, and L. Marechal-Drouard The voltage-dependent anion channel, a major component of the tRNA import machinery in plant mitochondria PNAS, November 28, 2006; 103(48): 18362 - 18367. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |