Colin Roger Bird

Carotenoid Biosynthesis during Tomato Fruit Development (Evidence for Tissue-Specific Gene Expression)

Tomato (Lycopersicon esculentum Mill. cv Ailsa Craig) fruit, at five stages of development, have been analyzed for their carotenoid and chlorophyll (Chl) contents, in vitro activities of phytoene synthase, phytoene desaturase, and lycopene cyclase, as well as expression of the phytoene synthase (Psy) and phytoene desaturase (Pds) genes. During ripening, the total carotenoids increased with a concomitant decrease in Chl. Although the highest carotenoid content (consisting mainly of lycopene and [beta]-carotene) was found in ripe fruit, the greatest carotenogenic enzymic activities were found in green fruit. Phytoene synthase was located in the plastid stroma, whereas the metabolism of phytoene was associated with plastid membranes during all stages of fruit development. The in vitro products of phytoene desaturation altered from being predominantly phytofluence and [zeta]-carotene in chloroplasts to becoming mainly lycopene in chromoplasts. The expression of Psy was detected in breaker and ripe fruit, as well as flowers, but was not detectable by northern blot analysis in leaves or green fruits. The Pds gene transcript was barely detectable in green fruit and leaves but was expressed in flowers and breaker fruit. These results suggest that transcription of Psy and Pds is regulated developmentally, with expression being considerably elevated in chromoplast-containing tissues. Antiserum to the Synechococcus phytoene synthase cross-reacted with phytoene synthase of green fruit only on western blots and not with the enzyme from ripe fruit. In contrast, a monoclonal antibody to the Psy gene product only cross-reacted with phytoene synthase from ripe fruit. The enzymes from green and ripe fruit had different molecular masses of 42 and 38 kD, respectively. The absence of detectable Psy and Pds mRNA in green tissues using northern blot analyses, despite high levels of phytoene synthase and desaturase activity, lends support to the hypothesis of divergent genes encoding these enzymes.

Inheritance and effect on ripening of antisense polygalacturonase genes in transgenic tomatoes

The role of the cell wall hydrolase polygalacturonase (PG) during fruit ripening was investigated using novel mutant tomato lines in which expression of the PG gene has been down regulated by antisense RNA. Tomato plants were transformed with chimaeric genes designed to express anti-PG RNA constitutively. Thirteen transformed lines were obtained of which five were analysed in detail. All contained a single PG antisense gene, the expression of which led to a reduction in PG enzyme activity in ripe fruit to between 5% and 50% that of normal. One line, GR16, showed a reduction to 10% of normal PG activity. The reduction in activity segregated with the PG antisense gene in selfed progeny of GR16. Plants homozygous for the antisense gene showed a reduction of PG enzyme expression of greater than 99%. The PG antisense gene was inherited stably through two generations. In tomato fruit with a residual 1% PG enzyme activity pectin depolymerisation was inhibited, indicating that PG is involved in pectin degradation in vivo. Other ripening parameters, such as ethylene production, lycopene accumulation, polyuronide solubilisation, and invertase activity, together with pectinesterase activity were not affected by the expression of the antisense gene.

Tomato exo-(1-->4)-beta-D-galactanase. Isolation, changes during ripening in normal and mutant tomato fruit, and characterization of a related cDNA clone.

An exo-(1->4)-[beta]-D-galactanase was isolated from ripe tomato fruit (Lycopersicon esculentum Mill. cv Ailsa Craig and cv Better Boy) using anion-exchange, gel filtration, and cation-exchange chromatography. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the most active fraction revealed a predominant protein band at 75 kD and several minor bands. A 30-amino acid N-terminal sequence from this 75-kD protein showed a high degree of homology with other recently identified [beta]-galactosidase/galactanase proteins from persimmon and apple fruits (I.-K. Kang, S.-G. Suh, K.C. Gross, J.-K. Byun [1994] Plant Physiol 105: 975–979; G.S. Ross, T. Wegrzyn, E.A. MacRae, R.J. Redgwell [1994] Plant Physiol 106: 521–528) and with the predicted polypeptide sequence encoded by the ethylene-regulated SR12 gene in carnation (K.G. Raghothama, K.A. Lawton, P.B. Goldsbrough, W.R. Woodson [1991] Plant Mol Biol 17: 61–71). The enzyme focused to a single band of [beta]-galactosidase activity on an isoelectrofocusing gel at pH 9.8. The enzyme was specific for (1->4)-[beta]-D-galactan substrates with a pH optimum of 4.5. The only reaction product detected was monomeric galactose, indicating that the enzyme was an exo-(1->4)-[beta]-D-galactanase. [beta]-Galactanase activity increased at the onset of ripening in normal fruit, but no similar increase was detected in the nonripening mutants nor and rin. A tomato homolog (pTom[beta]gal 1) was isolated using the SR12 cDNA clone from carnation as a probe. This clone showed 73% identity at the amino acid level with [beta]-galactosidase-related sequences from apple and asparagus and 66% identity with SR12. pTom[beta]gal 1 is a member of a gene family. Northern analysis demonstrated that pTom[beta]gal 1 expression was ripening related in normal fruits, with lower levels apparent in the nonsoftening mutants.

Gene Expression in the Pulp of Ripening Bananas (Two-Dimensional Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis of in Vitro Translation Products and cDNA Cloning of 25 Different Ripening-Related mRNAs)

mRNA was extracted from the pulp and peel of preclimacteric (d 0) bananas (Musa AAA group, cv Grand Nain) and those exposed to ethylene gas for 24 h and stored in air alone for a further 1 (d 2) and 4 d (d 5). Two-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis of in vitro translation products from the pulp and peel of these fruits revealed significant up-regulation of numerous transcripts during ripening. The majority of the changes were initiated by d 2, with the level of these messages increasing during the remainder of the ripening period. Pulp tissue from d 2 was used for the construction of a cDNA library. This library was differentially screened for ripening-related clones using cDNA from d-0 and d-2 pulp by a novel microtiter plate method. In the primary screen 250 up- and down-regulated clones were isolated. Of these, 59 differentially expressed clones were obtained from the secondary screen. All of these cDNAs were partially sequenced and grouped into families after database searches. Twenty-five nonredundant groups of pulp clones were identified. These encoded enzymes were involved in ethylene biosynthesis, respiration, starch metabolism, cell wall degradation, and several other key metabolic events. We describe the analysis of these clones and their possible involvement in ripening.

HIGH LEVELS OF RIPENING-SPECIFIC REPORTER GENE EXPRESSION DIRECTED BY TOMATO FRUIT POLYGALACTURONASE GENE-FLANKING REGIONS

The 1.4 kb 5′ polygalacturonase (PG) gene-flanking region has previously been demonstrated to direct ripening-specific chloramphenicol acetyl transferase (CAT) expression in transgenic tomato plants. The steady state level of CAT mRNA in these plants was estimated to be less than 1% of the endogenous PG mRNA. Further constructs containing larger PG gene-flanking regions were generated and tested for their ability to direct higher levels of reporter gene expression. A 4.8 kb 5′-flanking region greatly increased levels of ripening-specific reporter gene activity, while a 1.8 kb 3′ region was only shown to have a positive regulatory role in the presence of the extended 5′ region. Transgenic plants containing the CAT gene flanked by both of these regions showed the same temporal pattern of accumulation of CAT and PG mRNA, and steady-state levels of the transgene mRNA were equivalent to 60% of the endogenous PG mRNA on a per gene basis. The proximal 150 bp of the PG promoter gave no detectable CAT activity. However, the distal 3.4 kb of the 4.8 kb 5′ PG promoter was shown to confer high levels of ripening-specific gene expression when placed in either orientation upstream of the 150 bp minimal promoter. The DNA sequence of the 3.4 kb region revealed a 400 bp imperfect reverse repeat, and sequences which showed similarity to functionally significant sequences from the ripening-related, ethylene-regulated tomato E8 and E4 gene promoters. The possible roles of the flanking regions in regulating PG gene expression are discussed.

An endo-1,4-beta-glucanase expressed at high levels in rapidly expanding tissues.

Plant developmental processes involving modifications to cell wall structure, such as cell expansion, organ abscission and fruit ripening, are accompanied by increased enzyme activity and mRNA abundance of endo-1,4-β-glucanases (EGases). An EGase cDNA clone, Ce14, isolated from tomato (Lycopersicon esculentum) has been shown to be identical to a tomato pistil-predominant EGase cDNA, TPP18. In addition to its previously reported expression during certain stages of early pistil development, Ce14 mRNA was also detected at high levels in the growing zones of etiolated hypocotyls (about 2.5-fold less than in pistils) and in young expanding leaves (about 3.5-fold less than in pistils). The abundance of Ce14 mRNA declined precipitously in older tissues as cells became fully expanded, and was barely detectable in mature vegetative tissues. Ce14 mRNA abundance was also low in abscission zones, and did not increase as abscission progressed. In fruit, Ce14 mRNA was present at low levels during fruit expansion, but was essentially absent during subsequent fruit development and ripening. Treatment of etiolated hypocotyls with ethylene or high concentrations of auxin sufficient to induce rapid lateral cell expansion and hypocotyl swelling also brought about an approximate doubling of Ce14 mRNA abundance, suggesting that Ce14 mRNA accumulation may be promoted directly or indirectly by ethylene. Thus, accumulation of Ce14 mRNA was found to be correlated with rapid cell expansion in pistils, hypocotyls and leaves.

CONTROL AND MANIPULATION OF GENE-EXPRESSION DURING TOMATO FRUIT RIPENING

Ripening is a complex developmental process involving changes in the biochemistry, physiology and gene expression of the fruit. It is an active process characterised by changes in all cellular compartments. cDNA cloning has been used as an approach to analyse changes in gene expression during fruit ripening. This has revealed that several genes are switched on specifically during fruit ripening, including one encoding polygalacturonase (PG), a major cell wall protein. These cDNA clones have been used to study the expression of the genes in normal and ripening mutant fruits, and under environmental stress conditions.The PG gene has been isolated and it has been demonstrated that 1450 bases 5′ of the coding region are sufficient for the tissue- and development-specific expression of a bacterial marker gene in transgenic tomatoes. Antisense RNA techniques have been developed to generate novel mutant tomatoes in which the biochemical function of this enzyme and its involvement in fruit softening has been tested.

MOLECULAR CHARACTERISATION OF CDNA CLONES REPRESENTING PECTIN ESTERASE ISOZYMES FROM TOMATO

Two pectin esterase cDNA clones representing different isozymes with ca. 95% homology were isolated from an early ripening tomato fruit cDNA library. Both clones were longer than previously published sequences, and the encoded proteins possessed extended (229–233 amino acid) putative N-terminal extensions. In addition, the mRNA species corresponding to the two clones showed differential levels of expression in fruit.

Polymer mobility in cell walls of transgenic tomatoes with reduced polygalacturonase activity

Abstract Cell walls were prepared from unripe and red-ripe tomato fruit, cv. Ailsa Craig, and from ripe transgenic fruit carrying antisense genes downregulating the ripening-related polygalacturonase activity. The cell walls were examined by 13 C NMR using a cross-polarization/magic-angle spinning experiment with variable contact time to estimate the proton magnetic relaxation parameter T 1 p , which is sensitive to molecular motions on the kHz timescale and hence to the rigidity of the polymer network of the cell walls; and the cross-polarization time constant T CH which is sensitive to similar, and more local, motional effects. The proton T 1 p values for the 13 C resonances from each cell wall preparation were separated into two groups. One of these showed slow relaxation, implying low mobility, and included the 13 C resonances characteristic of cellulose. The other, corresponding to resonances characteristic of pectins only, showed two-component relaxation behaviour with one component relaxing much faster than the previous group, and hence corresponding to the pectic polymers in the middle lamella and perhaps also between the microfibrils. The T 1 p values for the first group shortened on ripening as polymer mobility increased within the cell wall. This was still evident, but less so, in the PG-antisense material, showing that the antisense genes reduced the softening of the cell wall, but did not do so as much as might be expected from the almost complete loss of polygalacturonase activity. The T CH was separated into very fast and slower components corresponding to the times required for local and larger-scale magnetization transfer.

The effect of reduced activity of phytoene synthase on isoprenoid levels in tomato pericarp during fruit development and ripening

Carotenoids, gibberellins (GAs), sterols, abscisic acid and β-amyrins were analysed in tomato (Lycopersicon esculentum Mill.) pericarp during fruit development and ripening. The contents of these isoprenoids in wild-type (cv. Ailsa Craig) fruit were compared with those in fruit of the carotenoid-deficient R-mutant and a transgenic plant containing antisense RNA to a phytoene synthase gene. In both carotenoid-deficient genotypes, a 14-fold reduction in carotene and twofold decrease in xanthophyll content, compared to the wild type, was found in ripe fruit. Immature green fruit from wild type and R-mutant plants contained similar amounts of the C19-GAs, GA1, and GA20, and their C20 precursor, GA19. Immature fruit from the transgenic plants contained three- to fivefold higher contents of these GAs. In wild-type fruit at the mature green stage the contents of these GAs had decreased to < 10% of the levels in immature fruit. A similar decrease in GA19 content occurred in the other genotypes. However, the contents of GA1 and GA20 in fruit from phytoene synthase antisense plants decreased only to 30% between the immature and mature green stages and did not decrease at all in R-mutant fruit. At the breaker and ripe stages, the contents of each GA were much reduced for all genotypes. The amount of abscisic acid was the same in immature fruit from all three genotypes, but, on ripening, the levels of this hormone in antisense and R-mutant fruit were ca. 50% of those in the wild type. Quantitative differences in the amounts of the triterpenoid β-amyrins, total sterols, as well as individual sterols, such as campesterol, stigmasterol and sitosterol, were apparent between all three genotypes during development. Amounts of free sterols of wild type and antisense fruit were greatest during development and decreased during ripening, whereas the opposite was found in the R-mutant. This genotype also possessed less free sterol and more bound sterol in comparison to the other varieties. These data provide experimental evidence to support the concept of an integrated metabolic relationship amongst the isoprenoids.