Feedback control and diurnal regulation of gibberellin 20-oxidase transcript levels in potato

Abstract

Tuber formation in potato (Solanum tuberosum) is promoted by short photoperiods and is inhibited by gibberellins (GAs). Endogenous levels of GA1 were shown to decrease in stolons and leaves of potato plants induced to tuberize, which suggests that photoperiodic regulation of GA biosynthesis may play a role in tuber induction. We report the isolation of three potato cDNA clones (StGA20ox1-3) encoding GA 20-oxidase, a key regulatory enzyme in the GA-biosynthetic pathway. Using northern analysis, we detected a differential pattern of tissue-specific expression of the mRNAs corresponding to these clones. StGA20ox mRNAs were also very abundant in leaves of the potato ga1 mutant, which is blocked in the 13-hydroxylation step, and were strongly down-regulated by gibberellic acid, suggesting a feedback regulation of these genes. In plants grown in short-day (inductive) conditions, levels of the StGA20ox transcripts in leaves fluctuated during a 24-h period, with a peak of accumulation observed about 4 h after the lights were turned off. Interruption of the night with a 30-min "night break" of light (noninductive conditions) did not have a marked effect on the levels of accumulation of the three GA 20-oxidase mRNAs during the day, but it induced a second peak of expression of StGA20ox1 and StGA20ox3 transcripts late in the night. This observation, together with the finding that StGA20ox1 mRNA is expressed at high levels in leaves, suggests that night-break induction of this gene might play a role in the control of tuberization by regulating endogenous levels of GAs in response to daylength conditions.

Figure 1

Alignments of the deduced amino acid sequences of the potato GA 20-oxidase clones StGA20ox1, StGA20ox2, and StGA20ox3 with those of pea (accession no. PsU58830), spinach (accession no. SoU33330), pumpkin (Cm20ox; accession no. X73314), and Arabidopsis (At2301; accession no. X83379). Identical residues are boxed in black; similar residues are shaded in gray. The conserved LPWKET and NYYPXCQKP regions thought to be involved in binding the GA substrate and the 2-oxoglutarate cosubstrate are underlined. Conserved H and D residues involved in the binding of Fe²⁺ are also indicated (*). Alignments were made using the PileUp and PrettyBox programs of the Genetics Computer Group.

Figure 2

Hybridization of clones StGA20ox1, StGA20ox2, and StGA20ox3 to a Southern blot of S. tuberosum subsp. andigena genomic DNA digested with EcoRI and HindIII. Ten micrograms of genomic DNA was loaded per lane. The HindII-XhoI fragments corresponding to the 3′-ends of the clones were used as probes. Molecular-mass markers are shown on the left (in kb).

Figure 3

Expression of the StGA20ox clones in different potato tissues. Northern analysis was carried out using 30 μg of total RNA from apex, young leaves, internodes (stem), roots, stolons, and tubers of S. tuberosum subsp. andigena plants grown under short days. Flower, fruit, and seed samples were obtained from S. demissum plants grown under long days. Ethidium-bromide staining of the gel is included to assess an equal loading of the lanes. RNA blots hybridized to probes StGA20ox1 and StGA20ox2 were exposed overnight, whereas the blot hybridized to probe StGA20ox3 was exposed for 4 d.

Figure 4

Feedback regulation of StGA20ox gene expression in potato shoots. Total RNA was isolated from control and ga₁ dwarf potato plants, ga₁ plants treated for 2 d with GA₃ (10 μm), and control plants treated with the GA-biosynthesis inhibitors ancymidol (5 mg L⁻¹) and prohexadione (50 mg L⁻¹). Thirty micrograms of total RNA was loaded per lane.

Figure 5

Diurnal variation in the levels of accumulation of transcript StGA20ox1 in leaves. Plants were grown in short days (SD) or short days with a night break (SD + NB) for 3 weeks. Leaf samples were harvested 30 min after the lights were turned on and taken at intervals of approximately 3 h and immediately before and after the night break. A, Light regimes used to grow the plants. The light period and night break are indicated by white bars; the dark period by a black bar. B, RNA blots hybridized with the StGA20ox1 probe. After hybridization to the StGA20ox1 probe, blots were probed with a fragment corresponding to the ribosomal S4 protein (Braun et al., 1994). StGA20ox1 mRNA levels were quantified by densitometric scanning of the film and normalized to S4 mRNA levels. The graph represents the levels of StGA20ox1 mRNA estimated for each sample compared with the highest hybridization signal (short days, 23.5 h), to which we assigned an arbitrary value of 100. Thirty micrograms of total RNA was loaded per lane. The experiment was repeated four times.

Figure 6

Diurnal variation of levels of StGA20ox2 and StGA20ox3 transcripts in leaves. Plants were grown in short days (SD) or short days with a night break (SD + NB) for 3 weeks. RNA blots were hybridized to probes StGA20ox2 and StGA20ox3, and subsequently to the ribosomal S4 protein probe for normalization. Densitometric scanning of the film quantified the levels of StGA20ox2 and StGA20ox3 mRNAs. The graphs represent the levels estimated for both transcripts compared with the strongest hybridization signal (short days, 23.5 h, in RNA blots hybridized with probe StGA200ox1), to which we assigned an arbitrary value of 100. Values correspond to approximately the levels of the respective mRNAs, as probes of about the same size and specific activities were used for hybridization. The experiment was repeated twice.