Arachidonic acid alters tomato HMG expression and fruit growth and induces 3-hydroxy-3-methylglutaryl coenzyme A reductase-independent lycopene accumulation

Abstract

Regulation of isoprenoid end-product synthesis required for normal growth and development in plants is not well understood. To investigate the extent to which specific genes for the enzyme 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGR) are involved in end-product regulation, we manipulated expression of the HMG1 and HMG2 genes in tomato (Lycopersicon esculentum) fruit using arachidonic acid (AA). In developing young fruit AA blocked fruit growth, inhibited HMG1, and activated HMG2 expression. These results are consistent with other reports indicating that HMG1 expression is closely correlated with growth processes requiring phytosterol production. In mature-green fruit AA strongly induced the expression of HMG2, PSY1 (the gene for phytoene synthase), and lycopene accumulation before the normal onset of carotenoid synthesis and ripening. The induction of lycopene synthesis was not blocked by inhibition of HMGR activity using mevinolin, suggesting that cytoplasmic HMGR is not required for carotenoid synthesis. Our results are consistent with the function of an alternative plastid isoprenoid pathway (the Rohmer pathway) that appears to direct the production of carotenoids during tomato fruit ripening.

Figure 1

Isoprenoid pathway in plants. Plant growth regulators and some enzymes of the pathway are circled. HMGR and PSY are highlighted in gray because of their possible rate-limiting roles. The steps catalyzed by PDS, FTase, and GGTase are also indicated. FPP and GGPP are the main branch points of the pathway. The alternative Rohmer pathway for the production of IPP in plastids is also indicated.

Figure 2

Effect of AA on young fruit development. Young tomato fruits attached to the plant were injected with a control or AA solution. Groups of three to five fruits were collected after 3 d or 1 month and used for experiments. A, Fruit weight 1 month after treatment with different concentrations of AA. B, Fruit samples from the experiment shown in C. C, Fruit weight 3 d after injection of 4 μL of a control solution or 5 μg/μL AA solution. Columns represent means and bars represent sd values. D, RNA blot of total RNA samples from the fruit samples shown in C. The same blot was hybridized with gene-specific HMG1, HMG2, and 18S rRNA probes.

Figure 3

Effect of AA treatment on mature tomato fruit. Mature-green fruits were injected with either 10 μL of a control solution or 5 μg/μL AA solution while attached to the plant, and collected 3 d later. A, Proportion of control and AA-treated fruits at different stages of pigmentation (Gillaspy et al., 1993). B, Lycopene concentration in control and AA-treated fruits from three different experiments. Columns represent means and bars represent sd values from three independent experiments.

Figure 4

RNA-blot analysis of gene expression in control and AA-treated mature fruit. Mature-green fruits were injected with 10 μL of a control solution or 5 μg/μL AA solution while attached to the plant. Groups of three to five fruits were collected and ground together after 6, 12, 24, 48, or 72 h. A, RNA-blot analysis of total RNA extracted from the different fruit pools. The blot was hybridized with a gene-specific HMG2 probe, and other probes from tomato PSY1, PDS, and FTA. An 18S rRNA probe was used to compare the RNA amounts loaded in each lane. B, Quantification of the steady-state levels of HMG2 and PSY1 mRNA. Open symbols, Control fruit; closed symbols, AA-treated fruit. The values shown are normalized with the 18S rRNA amounts and are expressed relative to the level detected in red, firm fruit.

Figure 5

HMGR activity and lycopene concentration in control and AA-treated fruit. Fruits were treated as described in Figure 4 with 10 μL of a control solution (C), with an AA solution (AA), or with an AA solution containing 9 mm mevinolin (AA+MEV), collected after 24, 48, and 72 h, and used for determination of microsomal HMGR activity (A) and measurement of lycopene concentration (B). Columns represent means and bars represent sd values from three independent fruit samples.