The N-end rule pathway controls multiple functions during Arabidopsis shoot and leaf development

Abstract

The ubiquitin-dependent N-end rule pathway relates the in vivo half-life of a protein to the identity of its N-terminal residue. This proteolytic system is present in all organisms examined and has been shown to have a multitude of functions in animals and fungi. In plants, however, the functional understanding of the N-end rule pathway is only beginning. The N-end rule has a hierarchic structure. Destabilizing activity of N-terminal Asp, Glu, and (oxidized) Cys requires their conjugation to Arg by an arginyl-tRNA-protein transferase (R-transferase). The resulting N-terminal Arg is recognized by the pathway's E3 ubiquitin ligases, called "N-recognins." Here, we show that the Arabidopsis R-transferases AtATE1 and AtATE2 regulate various aspects of leaf and shoot development. We also show that the previously identified N-recognin PROTEOLYSIS6 (PRT6) mediates these R-transferase-dependent activities. We further demonstrate that the arginylation branch of the N-end rule pathway plays a role in repressing the meristem-promoting BREVIPEDICELLUS (BP) gene in developing leaves. BP expression is known to be excluded from Arabidopsis leaves by the activities of the ASYMMETRIC LEAVES1 (AS1) transcription factor complex and the phytohormone auxin. Our results suggest that AtATE1 and AtATE2 act redundantly with AS1, but independently of auxin, in the control of leaf development.

Fig. 1.

The N-end rule pathway in mammals and plants. N-terminal amino acid residues are indicated by single-letter abbreviations. Yellow ovals denote the rest of a protein substrate. Primary, secondary, and tertiary denote distinct subsets of destabilizing N-terminal residues. C* represents oxidized Cys. Primary destabilizing residues are recognized in mammals by N-recognins of the UBR family (11, 12). In plants, aromatic hydrophobic type-2 residues are recognized by PRT1 (16), whereas basic type-1 residues are recognized by PRT6 (10).

Fig. 2.

AtATE1 and AtATE2 act redundantly in the control of plant development. (A) Loss of R-transferase activity in ate1 ate2 mutant seedlings. R-transferase activities in different mutant backgrounds were examined in vitro. The assay measures the conjugation of [³H]Arg to bovine α-lactalbumin, which bears N-terminal Glu, a substrate of R-transferases. Wild-type R-transferase activity was set to 100%. Activities are represented as a percentage of wild-type activity. Error bars represent standard errors calculated based on 6 independent measurements obtained with 2 different protein extracts. (B and C) Cleared wild-type (B) and ate1 ate2 double-mutant (C) leaves from plants grown in short-day conditions. Note the lobes and wavy leaf margins in C. (D and E) Wild-type (D) and ate1 ate2 double-mutant (E) plants grown for 3 months in short-day conditions. The ate1 ate2 mutants show early outgrowth of axillary meristems, as indicated by the formation of leaves in the axils of rosette leaves (arrowheads). (F) Phyllotaxis (red arrows) and internode elongation defects (yellow arrow) in ate1 ate2 double mutants. (G and H) Scanning electron micrograph of part of a stem from a wild-type (G) and an ate1 ate2 plant (H), respectively. Note the presence of patches of small cells in the double mutant. (Scale bars: 500 μm.) (I and J) Wild-type and ate1 ate2 mutant plants were grown in short-day conditions for 2 months and transferred to continuous light for a synchronous induction of flowering (see Fig. S1A). After transfer to inducing conditions, stems of ate1 ate2 double-mutant plants (J) exhibited reduced elongation compared with those of the wild type (I). The arrow in J points to an inflorescence with mature flowers. Pictures were taken 19 days after transfer to continuous light.

Fig. 3.

Expression patterns of AtATE1 and AtATE2. AtATE1 and AtATE2 GUS translational fusions were introduced into wild-type plants. T2 and T3 plants were stained at different stages of development to detect GUS expression. (A–D) AtATE1 reporter activity was detected in 5-day-old seedlings (A), especially in root (B) and shoot (C) apices, as well as in the vasculature and in hydathodes (arrow) of more mature leaves (D). (E and F) In flowers, AtATE1 reporter activity was found mainly in carpels and the connective tissue of anthers (E), whereas AtATE2 reporter activity was also detected in pollen grains (F).

Fig. 4.

Phenotypes of ate1 ate2 plants result from a disruption of the N-end rule pathway. Pictures of 70-day-old plants grown in short-day conditions. (A–D) In prt6-5 (B), ate1 ate2 (C), and ate1 ate2 prt6-5 plants (D), but not in the wild type (A), leaves formed in the axils of rosette leaves (arrows), indicating loss of apical dominance. (E–L) Contrary to leaves from the wild type (E and I), the leaf margins of prt6-5 (F and J), ate1 ate2 (G and K), and ate1 ate2 prt6-5 plants (H and L) were lobed and wavy (arrowheads). The leaves shown in I–L were cleared.

Fig. 5.

BP is expressed in leaves of ate1 ate2 plants but is not required for the leaf morphology defects. A BP::GUS reporter (26) was crossed into the ate1 ate2 mutant background, and BP expression was monitored in plants grown in continuous light. (A–C) Whereas no BP::GUS reporter activity was detected in wild-type leaves (A), GUS staining was observed in the serration tips of mature ate1 ate2 leaves (B and C). (C) Close-up on the leaf margin of the leaf shown in B (area indicated by a blue rectangle). (D–F) In contrast to bp-1 (D), leaf margins of ate1 ate2 bp-1 triple-mutant plants (F) are lobed and wavy, similar to those of ate1 ate2 mutants (E). The leaves shown in D–F were cleared. Plants were grown in short-day conditions for 2 months.

Fig. 6.

R-transferases act together with AS1, but independently of auxin, to regulate leaf development. (A–E) Synergistic genetic interaction between ate1 ate2 and as1-1. Plants were grown in short-day conditions for 2.5 months. (A and B) Leaf series of as1-1 and ate1 ate2 as1-1 plants, respectively, starting from leaf 15 upward. Arrowheads mark leaves with leaflets. (C–E) Close-up on leaves from the leaf series presented in A and B (indicated by asterisks), showing an as1-1 leaf (C) and ate1 ate2 as1-1 leaves with leaflets (D and E). (E) A late-arising leaf. (F and G) Mature leaf of an axr1-3 single mutant (F) and of an ate1 ate2 axr1-3 triple mutant (G), showing a more strongly lobed and wavy margin in the triple mutant. Pictures of 4-month-old plants grown in short days. (H and I) DR5:GUS reporter activity in the wild-type (H) and ate1 ate2 mutant (I) plants. The activity of the DR5:GUS reporter was monitored by staining 5-day-old seedlings grown in long-day conditions. (J) A working model for the functions of R-transferases in the regulation of leaf development. AS1 and auxin act together in the control of leaf development (indicated by the green area) and repress BP expression (37). Misexpression of BP in leaves of ate1 ate2 double-mutant plants implies that AtATE1/AtATE2 also act as negative regulators of BP. Moreover, the synergistic interaction between ate1 ate2 and as1-1 suggests that R-transferases and AS1 regulate common processes during leaf development (indicated by the blue sector). The absence of synergism between ate1 ate2 and axr1-3 further implies that R-transferases and auxin regulate leaf development independently.