A DEFICIENS homolog from the dioecious tree black cottonwood is expressed in female and male floral meristems of the two-whorled, unisexual flowers.

We isolated PTD, a member of the DEFICIENS (DEF) family of MADS box transcription factors, from the dioecious tree, black cottonwood (Populus trichocarpa). In females, in situ hybridization experiments showed that PTD mRNA was first detectable in cells on the flanks of the inflorescence meristem, before differentiation of individual flowers was visually detectable. In males, the onset of PTD expression was delayed until after individual flower differentiation had begun and floral meristems were developing. Although PTD was initially expressed throughout the inner whorl meristem in female and male flowers, its spatial expression pattern became sex-specific as reproductive primordia began to form. PTD expression was maintained in stamen primordia, but excluded from carpel primordia, as well as vegetative tissues. Although PTD is phylogenetically most closely related to the largely uncharacterized TM6 subfamily of the DEF/APETELA3(AP3)/TM6 group, its spatio-temporal expression patterns are more similar to that of DEF and AP3 than to other members of the TM6 subfamily.

Diverse effects of overexpression of LEAFY and PTLF, a poplar (Populus) homolog of LEAFY/FLORICAULA, in transgenic poplar and Arabidopsis.

PTLF, the Populus trichocarpa homolog of LEAFY (LFY) and FLORICAULA, was cloned to assess its function in a dioecious tree species. In situ hybridization studies showed that the gene was expressed most strongly in developing inflorescences. Expression was also seen in leaf primordia and very young leaves, most notably in apical vegetative buds near inflorescences, but also in seedlings. Although ectopic expression of the PTLF cDNA in Arabidopsis accelerated flowering, only one of the many tested transgenic lines of Populus flowered precociously. The majority of trees within a population of 3-year-old transgenic hybrid Populus lines with PTLF constitutively expressed showed few differences when compared to controls. However, phenotypic effects on growth rate and crown development, but not flowering, were seen in some trees with strong PTLF expression and became manifest only as the trees aged. Competence to respond to overexpression of LFY varied widely among Populus genotypes, giving consistent early flowering in only a single male P. tremula x P. tremuloides hybrid and causing gender change in another hybrid genotype. PTLF activity appears to be subject to regulation that does not affect heterologously expressed LFY, and is dependent upon tree maturation. Both genes provide tools for probing the mechanisms of delayed competence to flower in woody plants.

Structure and expression of duplicate AGAMOUS orthologues in poplar.

To investigate the homeotic systems underlying floral development in a dioecious tree, and to provide tools for the manipulation of floral development, we have isolated two Populus trichocarpa genes, PTAG1 and PTAG2, homologous to the Arabidopsis floral homeotic gene AGAMOUS (AG). PTAG1 and PTAG2 are located on separate linkage groups, but their non-coding regions are highly similar, consistent with a phylogenetically recent duplication. Intron/exon structure is conserved in relation to AG and the Antirrhinum AG orthologue, PLENA (PLE), and low-stringency Southern analysis demonstrated the absence of additional genes in the poplar genome with significant PTAG1/2 homology. PTAG1 and PTAG2 exhibit an AG-like floral expression pattern, and phylogenetic analysis of the AG subfamily strongly supports evolutionary orthology to C-class organ identity genes. The high degree of similarity shared by PTAG1 and PTAG2 in both sequence (89% amino acid identity) and expression indicates that they are unlikely to be functionally associated with specification of tree gender. Unexpectedly, PTAG transcripts were consistently detected in vegetative tissues.

A mutation hotspot in the chloroplast genome of a conifer (Douglas-fir: Pseudotsuga) is caused by variability in the number of direct repeats derived from a partially duplicated tRNA gene.

We determined the DNA sequence of a 2.7-kb cpDNA XbaI fragment from douglas-fir [Pseudotsuga menziesii (Mirb.) Franco]. RFLPs revealed by the 2.7-kb XbaI clone were observed to vary up to 1 kb among species within the genus Pseudotsuga and up to 200 bp among trees of P. menziesii. The polymerase chain reaction (PCR) allowed the locus of polymorphism to be identified, and the variable region was then sequenced in a second Douglas-fir tree, a single tree of a related species, Japanese Douglas-fir (P. japonica), and in a species lacking a mutation hotspot in the region, Pinus radiata (Monterey pine). The locus of polymorphism is characterized by hundreds of base pairs of imperfect, tandem direct repeats flanked by a partially duplicated and an intact trn Y-GUA gene. The duplication is direct in orientation and consists of 43 bp of the 3' end of trnY and 25 bp of its 3' flanking sequence. Tandem repeats show high sequence similarity to a 27-bp region of the trnY gene that overlaps one end of the duplication. The two trees of Douglas-fir sequenced differed by a single tandem repeat unit, whereas these trees differed from the Japanese Douglas-fir sequenced by approximately 34 repeat units. Repetitive DNA in the Pseudotsuga cpDNA hotspot was most likely generated at the time of the partial trnY gene duplication and these sequences expanded by slipped-strand mispairing and unequal crossing-over.

LE-ACS4, a fruit ripening and wound-induced 1-aminocyclopropane-1-carboxylate synthase gene of tomato (Lycopersicon esculentum). Expression in Escherichia coli, structural characterization, expression characteristics, and phylogenetic analysis.

ACC (1-aminocyclopropane-1-carboxylic acid) synthase is the key regulatory enzyme in the biosynthetic pathway of the plant hormone ethylene and is encoded by a highly divergent multigene family in tomato (Rottmann, W. H., Peter, G. F., Oeller, P. W., Keller, J. A., Shen, N. F., Nagy, B. P., Taylor, L. P., Campbell, A. D., and Theologis, A. (1991) J. Mol. Biol. 222, 937-961). Two members of the family, LE-ACS2 and LE-ACS4, are induced during fruit ripening and upon treatment of mature green fruits with exogenous ethylene (C2H4) in a dose-dependent manner. Both genes are superinduced by wounding of pericarp tissue during various stages of ripening. The wound-induced accumulation of LE-ACS2 mRNA is more rapid and greater than that of LE-ACS4. Both mRNAs accumulate in the absence of protein synthesis, suggesting that their induction is a primary response to the inducer. The LE-ACS4 gene was isolated and structurally characterized. The function of the LE-ACS4 protein (53,509 Da, pI 5.4) was verified by expression experiments in Escherichia coli. The promoters of LE-ACS2 and LE-ACS4 contain potential cis-acting regulatory elements responsible for induction by ethylene, wounding, and anaerobiosis. In addition, elements for binding the transcriptional factors EmBP1, GBF-1, and OCSBF-1 are also present. Phylogenetic analysis of 20 ACC synthases from dicots and monocots indicate that the LE-ACS2 and LE-ACS4 proteins belong to an unique sublineage that includes an additional member of the tobacco family, NT-ACS1. The divergence of this sublineage is a relatively recent event in the evolution of ACC synthase protein.

Use of a tomato mutant constructed with reverse genetics to study fruit ripening, a complex developmental process.

Fruit ripening is one of the most dramatic developmental transitions associated with extensive alteration in gene expression. The plant hormone ethylene is considered to be the causative ripening agent. Transgenic tomato plants were constructed expressing antisense or sense RNA to the key enzyme in the ethylene (C2H4) biosynthetic pathway, 1-aminocyclopropane-1-carboxylate (ACC) synthase using the constitutive CaMV 35S and fruit specific E8 promoters. Fruits expressing antisense LE-ACS2 RNA produce less ethylene and fail to ripen only when ethylene production is suppressed by more than 99% (> 0.1 nl/g fresh weight). Ethylene production is considerably inhibited (50%) in fruits expressing sense LE-ACS2 RNA. Antisense fruits accumulate normal levels of polygalacturonase (PG), ACC oxidase (pTOM13), E8, E17, J49, and phytoene desaturase (D2) mRNAs which were previously thought to be ethylene-inducible. E4 gene expression is inhibited in antisense fruits and its expression is not restored by treatment with exogenous propylene (C3H6). Antisense fruits accumulate PG mRNA, but it is not translated. Immunoblotting experiments indicate that the PG protein is not expressed in antisense fruits but its accumulation is restored by propylene (C3H6) treatment. The results suggest that at least two signal-transduction pathways are operating during tomato fruit ripening. The independent (developmental) pathway is responsible for the transcriptional activation of genes such as PG, ACC oxidase, E8, E17, D2, and J49. The ethylene-dependent pathway is responsible for the transcriptional and posttranscriptional regulation of genes involved in lycopene, aroma biosynthesis, and the translatability of developmentally regulated genes such as PG.

1-aminocyclopropane-1-carboxylate synthase in tomato is encoded by a multigene family whose transcription is induced during fruit and floral senescence.

The key regulatory enzyme in the biosynthetic pathway of the plant hormone ethylene is 1-aminocyclopropane-1-carboxylic acid (ACC) synthase (EC 4.1.1.14). It catalyzes the conversion of S-adenosylmethionine to ACC, the precursor of ethylene. We isolated complementary DNA sequences, ptACC2 and ptACC4, for two distinct and differentially regulated ACC synthase mRNAs expressed in ripe tomato fruit. The authenticity of the clones has been confirmed by expression experiments in E. coli. The predicted size of the encoded polypeptides (54,690 and 53,519 Da) is similar to that of the primary in vitro translation products and to the proteins found in vivo. The sequence of the gene encoding one mRNA, LE-ACC2, has been determined and its transcription initiation site defined. Four additional genes, LE-ACC1A, LE-ACC1B, LE-ACC3 and LE-ACC4, have also been identified and the sequence of their coding regions determined. The LE-ACC1A and LE-ACC1B genes are adjacent to each other and are convergently transcribed. Their encoded polypeptides are 96% identical; the identity of the other polypeptides to each other varies between 50 and 70%. The proteins predicted to be encoded by the ACC synthase genes so far cloned from tomato and zucchini contain 11 of the 12 conserved amino acid residues found in various aminotransferases involved in the binding of the substrate and the cofactor pyridoxal-5'-phosphate. The data indicate that ACC synthase is encoded by a divergent multigene family in tomato that encodes proteins related to aminotransferases.

Transfer of methomyl and HmT-toxin sensitivity from T-cytoplasm maize to tobacco.

The mitochondrial gene, T-urf13, which is unique to the T-cytoplasm of maize, has been expressed in tobacco plants using the Cauliflower Mosaic Virus 35S promoter. Tobacco plants expressing T-urf13 exhibit a variety of responses to methomyl. Leaf discs and petiole sections bleach when exposed to methomyl or HmT-toxin; this effect increases with the age of the tissue. The bleaching effect is not however observed when light is excluded. Plants homozygous for T-urf13 exhibit extreme sensitivity when sprayed with methomyl. The growth of seedling which are either homozygous or heterozygous for T-urf13 is inhibited by methomyl and by kanamycin, whereas seedlings from untransformed tobacco or tobacco which has lost the T-urf13 gene through segregation are sensitive to kanamycin but develop normally when exposed to methomyl. The results demonstrate that T-URF13 need not be specifically targeted to the mitochondrion for it to induce methomyl or HmT-toxin sensitivity in tobacco.

Two genes encoding 1-aminocyclopropane-1-carboxylate synthase in zucchini (Cucurbita pepo) are clustered and similar but differentially regulated.

A 17-kilobase (kb) region of the zucchini (Cucurbita pepo) genome has been sequenced and contains two genes, CP-ACC1A and CP-ACC1B, encoding 1-aminocyclopropane-1-carboxylate synthase (ACC synthase; S-adenosyl-L-methionine methylthioadenosine-lyase, EC 4.4.1.14). The genes are transcribed convergently and are separated by a 5.7-kb intergenic region. Their coding regions are interrupted by four introns located in identical positions. While the DNA identity in their coding regions is 97%, their 5' and 3' flanking regions are highly divergent. Transcription of CP-ACCIA is rapidly induced by wounding in fruit and etiolated hypocotyls and by indoleacetic acid (IAA)/benzyladenine/LiCl only in fruit tissue. Conditions that induce CP-ACC1B expression have not been found. Protein synthesis inhibition derepresses the expression of CP-ACC1A and other unidentified ACC synthase genes, suggesting that they may be under negative control. The amino acid sequences deduced from the nucleotide sequences of the genes are 493 and 494 residues long with 95% identity. The most notable feature of the amino acid sequence is the presence of 11 of the 12 invariant amino acid residues involved in the binding of the substrate and pyridoxal-5'-phosphate in various aminotransferases. We conclude that ACC synthase is encoded by a multigene family of which certain members are differentially induced by auxin in a tissue-specific manner. Furthermore, ACC synthase, a pyridoxal-containing enzyme, may have an evolutionary relationship with the superfamily of aminotransferases.

A mitochondrial gene is lost via homologous recombination during reversion of CMS T maize to fertility.

The Texas (T) male sterile cytoplasm of maize is distinguished by a mitochondrially synthesized 13-kd polypeptide and a high susceptibility to the toxin produced by the fungal pathogen Helminthosporium maydis. Fertile, toxin-resistant revertants show an altered restriction profile for mitochondrial DNA and do not produce the 13-kd polypeptide. Characterization of cosmid clones from CMS T maize and a revertant shows that a heavily transcribed open reading frame named T-URF13, potentially coding a 13-kd product, is deleted in the revertant mitochondria. Six transcripts present in CMS T mitochondria, 4000, 3000, 2000, 1800, 1500 and 1200 nucleotides in length, are lacking in revertant mitochondria. T-URF25, an open reading frame coding for a 25-kd product, lies to the 3' end of T-URF13 but is retained in the revertants. T-URF13 and T-URF25 are co-transcribed in CMS T mitochondria; in the revertant T-URF25 is present on a 3100-nucleotide species. The recombination that caused these changes involved a 127-bp repeated sequence. Homologous recombination took place within the central 55 bp of this imperfect repeat. Hybridization analysis of DNA and RNA from other revertants demonstrates that a similar or identical event has taken place independently in these revertants.

The complete amino acid sequence of the human aldolase C isozyme derived from genomic clones.

The complete protein sequence of the human aldolase C isozyme has been determined from recombinant genomic clones. A genomic fragment of 6673 base pairs was isolated and the DNA sequence determined. Aldolase protein sequences, being highly conserved, allowed the derivation of the sequence of this isozyme by comparison of open reading frames in the genomic DNA to the protein sequence of other human aldolase enzymes. The protein sequence of the third aldolase isozyme found in vertebrates, aldolase C, completes the primary structural determination for this family of isozymes. Overall, the aldolase C isozyme shared 81% amino acid homology with aldolase A and 70% homology with aldolase B. The comparisons with other aldolase isozymes revealed several aldolase C-specific residues which could be involved in its function in the brain. The data indicated that the gene structure of aldolase C is the same as other aldolase genes in birds and mammals, having nine exons separated by eight introns, all in precisely the same positions, only the intron sizes being different. Eight of these exons contain the protein coding region comprised of 363 amino acids. The entire gene is approximately 4 kilobases.

Complete amino acid sequence for human aldolase B derived from cDNA and genomic clones.

Several aldolase B clones from a human liver cDNA library have been identified by using a rabbit aldolase A cDNA as a hybridization probe. The most complete of these, pHL413, is 1389 base pairs long and covers approximately equal to 80% of the length of the mRNA, including 90% of the translated region. The cDNA, pHL413, was used to identify a genomic clone, lambda HG313, which encoded the remaining amino acids of human aldolase B. We demonstrate that the amino acid and nucleotide sequences of aldolase are strongly conserved even between different isozymes. Furthermore, in the 3'-untranslated regions of the mRNAs for the B isozyme of human and rat there is an extensive stretch of homology. Aldolase B lacks a cysteine at positions 72 and 338 and lacks a histidine at position 361. These residues, which are present in rabbit aldolase A, have previously been proposed to take part in catalysis. Our findings suggest that this may not be the case.