Elongator function in tRNA wobble uridine modification is conserved between yeast and plants

Abstract

Based on studies in yeast and mammalian cells the Elongator complex has been implicated in functions as diverse as histone acetylation, polarized protein trafficking and tRNA modification. Here we show that Arabidopsis mutants lacking the Elongator subunit AtELP3/ELO3 have a defect in tRNA wobble uridine modification. Moreover, we demonstrate that yeast elp3 and elp1 mutants expressing the respective Arabidopsis Elongator homologues AtELP3/ELO3 and AtELP1/ELO2 assemble integer Elongator complexes indicating a high degree of structural conservation. Surprisingly, in vivo complementation studies based on Elongator-dependent tRNA nonsense suppression and zymocin tRNase toxin assays indicated that while AtELP1 rescued defects of a yeast elp1 mutant, the most conserved Elongator gene AtELP3, failed to complement an elp3 mutant. This lack of complementation is due to incompatibility with yeast ELP1 as coexpression of both plant genes in an elp1 elp3 yeast mutant restored Elongator's tRNA modification function in vivo. Similarly, AtELP1, not ScELP1 also supported partial complementation by yeast-plant Elp3 hybrids suggesting that AtElp1 has less stringent sequence requirements for Elp3 than ScElp1. We conclude that yeast and plant Elongator share tRNA modification roles and propose that this function might be conserved in Elongator from all eukaryotic kingdoms of life.

Fig. 1

Failure of AtELP3 to complement the elp3Δ mutant despite interaction with yeast Elp1 protein. A. To test for functional complementation the elp3Δ mutant (CMY135) was transformed with plasmids containing ELP3 (pFF9), AtELP3 (YEpA4) and vector control (YEplac195) and subsequently with the GAL1 promoter driven γ-toxin expression plasmid pHMS14. Transformants were spotted in replica onto glucose-repressing (glc) or galactose-inducing (gal) media and grown for 4 days at 30°C. Growth on galactose indicates γ-toxin resistance (Tox^R) and no growth corresponds to γ-toxin sensitivity (Tox^S). B. To test for thermosensitivity and hypersensitivity to caffeine strains were serially diluted and replica spotted on YPD plates lacking (control) or containing 7.5 mM caffeine (right) and incubated for 4 days at 30°C and 37°C (middle). C. Anti-c-myc immunoprecipitates (IP) of strains containing chromosomally tagged ELP3-c-myc (FFY3t), AtELP3-c-myc on a plasmid in an elp3Δ background (CMY307 + yatELP3M) and wild-type parent without epitope (FY1679-8A) were analysed on Western blots probed with anti-c-myc or anti-Elp1 antibodies. Positions of the respective proteins are indicated by arrows. Protein extracts from cells without epitope-tag served as negative control. The asterisk denotes an N-terminally truncated Elp1 form (Fichtner et al., 2003). The relative abundance of this truncated form varies between experiments but is probably not dependent on which Elp3 peptide is present.

Fig. 2

Reconstitution of Elongator subunit interactions by AtELP3 demonstrating AtELP3 integration in the yeast Elongator complex. A. Protein extracts from cells expressing HA-tagged Elp5 together with c-myc-tagged Elp2 in an elp3Δ background (strains CMY304 and CMY301, respectively) and transformed with plasmids containing ELP3 (pFF9), AtELP3 (YEpA4), AtELP3-c-myc (yatELP3M) or vector control (YEplac195) were immunoprecipitated using anti-HA antibody and analysed by Western blotting. B. Same as (A) but with strain CMY301, which expresses HA-tagged Kti12 instead of Elp5-HA. The anti-HA antibody was used to detect Elp5-HA or Kti12-HA and anti-c-myc antibodies recognized Elp2 and AtELP3. Protein extracts from cells without epitope tag and cells expressing only AtELP3-c-myc served as negative controls. The pre-IPs served as loading control (bottom panels).

Fig. 3

Restoration of Elp3 stability and Elongator subunit interactions in an elp1Δ strain expressing AtElp1. A. AtElp1 is able to replace yeast Elp1 and to support Elp2–Elp3 interaction and Elp3 protein stability. Protein extracts from cells expressing chromosomally c-myc-tagged Elp2 together with HA-tagged Elp3 in an elp1Δ background (FFY2/3-dt-1d) and transformed with plasmids containing ELP1 (pFF13), AtELP1 (pDJ98) or vector control (YEplac181) were immunoprecipitated using anti-c-myc antibodies. Western blots of the precipitates were probed with anti-c-myc or anti-HA antibodies to detect Elp2 and Elp3 respectively (arrows). Instability of Elp3 in the elp1Δ strain is revealed in pre-IPs (bottom panel, lane 3). B. AtElp1 is able to restore Kti12-Elp2 interaction. Protein extracts from cells expressing c-myc-tagged Elp2 together with HA-tagged Kti12 in an elp1Δ background (CMY300), and transformed with plasmids containing ELP1 (pFF13), AtELP1 (pDJ98) or vector control (YEplac181) were immunoprecipitated and analysed as in (A).

Fig. 5

Restoration of the tRNA modification function of Elongator by complementation of a yeast elp1Δelp3Δ double mutant with AtELP1 and AtELP3. A. Restoration of toxin sensitivity. The elp1Δ elp3Δ (CMY134) strain was transformed with plasmids containing ELP1 (pFF13), ELP3 (pFF9), AtELP1 (pCM26), AtELP3 (YEpA4) and vector controls (YEplac181 and YEplac195) in the indicated combinations. The γ-toxin was expressed from pHMS14 (Frohloff et al., 2001) on galactose medium. Growth on galactose after 4 days at 30°C indicates γ-toxin resistance (Tox^R) and no growth equals γ-toxin sensitivity (Tox^S). B. Restoration of SUP4 suppressor tRNA function. SUP4 elp1Δ elp3Δ (CMY160) cells were transformed with plasmids containing wild-type ELP1 and ELP3 (pFF13 and p424TDH-ScElp3myc), the respective empty vectors (YEplac181 and p424TDH), the plant homologues AtELP1 and AtELP3 (pCM26 and p424TDH-AtElp3myc) or AtELP1 and the swapped HAT domain variants pElp3-AtHAT and pAtElp3-ScHAT. To check for suppression of the ade2-1 (UAA) ochre mutation (Huang et al., 2005) serial dilutions of the transformants were spotted on SC and SC-Ade plates and cultivated for 4 days at 30°C. SUP4 suppressor tRNA function results in growth on SC-Ade plates (Ade⁺ phenotype). C. Schematic representation of the Elp3 domain structure and domain swapped hybrids generated between AtELP3 and ScElp3 used as in (B).

Fig. 4

Formation of a chimeric Elongator complexes composed of yeast and plant subunits. A. AtELP1 and AtELP3 interact with each other in yeast. Anti-c-myc immunoprecipitates of elp1Δ elp3Δ strains (CMY134) containing FLAG-AtELP1 (pJET13) and AtELP3-c-myc were probed with anti-c-myc antibodies to detect AtELP3-c-myc and with anti-FLAG-antibodies to detect FLAG-AtELP1. Immunoprecipitates with anti-c-myc antibodies from elp1Δelp3Δ cells containing pFLAG-AtELP1 (lane 2) or no epitope-tag (lane1) served as negative controls. B. AtELP1 and AtELP3 restore Elp2–Elp5 interaction in the elp1Δ elp3Δ mutant. Protein extracts from cells expressing HA-tagged Elp5 together with c-myc-tagged Elp2 in an elp1Δ elp3Δ double mutant background (CMY305) and transformed with plasmids containing ELP1 (pFF13), ELP3 (pFF9), AtELP1 (pDJ98), AtELP3 (YEpA4), FLAG-AtELP1 (pJET13), AtELP3-c-myc (pELO3-myc) and vector controls (YEplac181, YEplac195) were immunoprecipitated using anti-HA antibodies. The antibodies used to detect the indicated proteins by Western blotting are marked on the right. Protein extracts from cells without epitope-tag and cells expressing only AtELP3-c-myc served as negative controls. Pre-IPs served as loading control (bottom panel). C. AtELP1 and AtELP3 restore Elp2–Kti12 interaction. Protein extracts from strain CMY302 expressing HA-tagged Kti12 together with c-myc-tagged Elp2 in an elp1Δ elp3Δ background and transformed (see B) were analysed as described in (B).

Fig. 6

AtELP3/ELO3 is required for formation of mcm⁵s²U and ncm⁵U modified nucleosides in tRNA. A.–D. Total tRNA isolated from wild-type and elo3 plants was analysed by HPLC (see Experimental procedures). Wild-type profiles are shown on the left, elo3 profiles in the right panels. A. and B. The parts of the chromatograms between retention times 15.5 and 19.6 min are displayed. The arrow in (B) indicates the expected retention time of ncm⁵U. Chromatograms were monitored at 254 nm. C. and D. The parts of the chromatograms between retention times 40.0 and 52.5 min are displayed. The arrow in (D) indicates the expected retention time of mcm⁵s²U. Chromatograms were monitored at 314 nm. E. Uridine modifications found at the wobble position of eukaryotic tRNAs. Elongator-dependent side groups are circled (Huang et al., 2005).