Engineering of Tomato Glandular Trichomes for the Production of Specialized Metabolites.

Glandular trichomes are specialized tissues on the epidermis of many plant species. On tomato they synthesize, store, and emit a variety of metabolites such as terpenoids, which play a role in the interaction with insects. Glandular trichomes are excellent tissues for studying the biosynthesis of specialized plant metabolites and are especially suitable targets for metabolic engineering. Here we describe the strategy for engineering tomato glandular trichomes, first with a transient expression system to provide proof of trichome specificity of selected promoters. Using microparticle bombardment, the trichome specificity of a terpene-synthase promoter could be validated in a relatively fast way. Second, we describe a method for stable expression of genes of interest in trichomes. Trichome-specific expression of another terpene-synthase promoter driving the yellow-fluorescence protein-gene is presented. Finally, we describe a case of the overexpression of farnesyl diphosphate synthase (FPS), specifically in tomato glandular trichomes, providing an important precursor in the biosynthetic pathway of sesquiterpenoids. FPS was targeted to the plastid aiming to engineer sesquiterpenoid production, but interestingly leading to a loss of monoterpenoid production in the transgenic tomato trichomes. With this example we show that trichomes are amenable to engineering though, even with knowledge of a biochemical pathway, the result of such engineering can be unexpected.

The tomato Mi-1 gene confers resistance to both root-knot nematodes and potato aphids.

Mi-1, a Lycopersicon peruvianum gene conferring resistance to the agricultural pests, root-knot nematodes, and introgressed into tomato, has been cloned using a selective restriction fragment amplification based strategy. Complementation analysis of a susceptible tomato line with a 100 kb cosmid array yielded a single cosmid clone capable of conferring resistance both to the root-knot nematode Meloidogyne incognita and to an unrelated pathogen, the potato aphid Macrosiphum euphorbiae. This resistance was stable. The Mi-1 gene encodes a protein sharing structural features with the nucleotide-binding site leucine-rich repeat-containing type of plant resistance genes.

Dissection of the fusarium I2 gene cluster in tomato reveals six homologs and one active gene copy.

The I2 locus in tomato confers resistance to race 2 of the soil-borne fungus Fusarium oxysporum f sp lycopersici. The selective restriction fragment amplification (AFLP) positional cloning strategy was used to identify I2 in the tomato genome. A yeast artificial chromosome (YAC) clone covering approximately 750 kb encompassing the I2 locus was isolated, and the AFLP technique was used to derive tightly linked AFLP markers from this YAC clone. Genetic complementation analysis in transgenic R1 plants using a set of overlapping cosmids covering the I2 locus revealed three cosmids giving full resistance to F. o. lycopersici race 2. These cosmids shared a 7-kb DNA fragment containing an open reading frame encoding a protein with similarity to the nucleotide binding site leucine-rich repeat family of resistance genes. At the I2 locus, we identified six additional homologs that included the recently identified I2C-1 and I2C-2 genes. However, cosmids containing the I2C-1 or I2C-2 gene could not confer resistance to plants, indicating that these members are not the functional resistance genes. Alignments between the various members of the I2 gene family revealed two significant variable regions within the leucine-rich repeat region. They consisted of deletions or duplications of one or more leucine-rich repeats. We propose that one or both of these leucine-rich repeats are involved in Fusarium wilt resistance with I2 specificity.

AFLP-based fine mapping of the Mlo gene to a 30-kb DNA segment of the barley genome.

Resistance of barley (Hordeum vulgare) to the powdery mildew fungus Erysiphe graminis f.sp. hordei is conferred by several dominant genes, but also by recessive alleles of the Mlo locus mapping on the long arm of chromosome 4. In addition, this single-factor-mediated resistance is active against all known physiological races of the parasite. Thus the mechanism underlying mlo-mediated resistance should differ substantially from that mediated by the dominant genes. A positional cloning strategy to isolate the Mlo gene from the barley genome, the size of which is almost double the size of the human genome, has been designed. The AFLP technique was employed to identify markers tightly linked to the Mlo locus and to produce a local high-resolution genetic map. The use of this high-volume marker technology allowed the rapid screening of approximately 250,000 loci for linkage to Mlo. A large number of Mlo-linked AFLP markers were identified, one of which cosegregated with Mlo on the basis of more than 4000 meiotic events. A four-genome-equivalent barley YAC library (average insert size 480 kb) was constructed and screened with this cosegregating marker. Four YACs containing this marker were isolated and subsequent characterization by AFLP-based physical mapping allowed the physical delimitation of the Mlo locus to a DNA segment of 30 kb.

The barley Mlo gene: a novel control element of plant pathogen resistance.

Mutation-induced recessive alleles (mlo) of the barley Mlo locus confer a leaf lesion phenotype and broad spectrum resistance to the fungal pathogen, Erysiphe graminis f. sp. hordei. The gene has been isolated using a positional cloning approach. Analysis of 11 mutagen-induced mlo alleles revealed mutations leading in each case to alterations of the deduced Mlo wild-type amino acid sequence. Susceptible intragenic recombinants, isolated from mlo heteroallelic crosses, show restored Mlo wild-type sequences. The deduced 60 kDa protein is predicted to be membrane-anchored by at least six membrane-spanning helices. The findings are compatible with a dual negative control function of the Mlo protein in leaf cell death and in the onset of pathogen defense; absence of Mlo primes the responsiveness for the onset of multiple defense functions.

Detection of a new submicroscopic Norrie disease deletion interval with a novel DNA probe isolated by differential Alu PCR fingerprint cloning.

Differential Alu PCR fingerprint cloning was used to isolate a DNA probe from the Xp11.4-->p11.21 region of the human X chromosome. This novel sequence, cpXr318 (DXS742), detects a new submicroscopic deletion interval at the Norrie disease locus (NDP). Combining our data with the consensus genetic map of the proximal short arm of the X chromosome, we propose the physical order Xcen-DXS14-DXS255-(DXS426, TIMP)-(DXS742-([MAOB-MAOA-DXS7], NDP)-DXS77-DXS228)-DXS209-DXS148-DXS196-++ +Xpter. The cpXr318 probe and a subclone from a cosmid corresponding to the DXS7 locus were converted into sequence-tagged sites. Finally, DXS742, DSX7, DXS77, and MAOA were integrated into a physical map spanning the Norrie disease locus.

Physical fine-mapping of a deletion spanning the Norrie gene.

Norrie disease (ND), atrophia bulborum hereditaria, is caused by a gene defect on the proximal short arm of the X-chromosome. As shown by us and others, microdeletions spanning the DXS7 locus are not uncommon in this disorder, and there is recent evidence that, at least in some of the Norrie deletion patients, the monoamine oxidase (MAO) A and B genes are deleted as well. Molecular hybridization experiments with 19 cloned DNA fragments have enabled us to construct a preliminary long-range restriction map around DXS77, DXS7, MAO-A and MAO-B, and to localize the distal end point of an ND deletion between DXS77 and DXS7.

Microdeletion in the X-chromosome and prenatal diagnosis in a family with Norrie disease.

We have studied a three-generation family in which Norrie disease is segregating and have performed prenatal diagnosis on the fetus of an obligatory carrier. Deletions at loci DXS7 and DXS77 defined by probes L1.28, L1.28-p59, and pX59 were detected in the affected male. DNA studies of chorionic villus biopsy material indicated that the male fetus had inherited the normal allele from the carrier mother. This prediction was confirmed on eye examination at age 5 months.

Physical fine mapping of the choroideremia locus using Xq21 deletions associated with complex syndromes.

Characterization of several male-viable deletions and duplications with 20 random DNA probes has enabled us to subdivide the Xq21 region into seven discernible intervals. Almost all of the deletions spanning part of Xq21 are associated with choroideremia and mental retardation, with deafness being another common feature. The gene locus for choroideremia was assigned to interval 3 spanning the loci DXS95, DXS165, and DXS233. Genes for X-linked deafness and mental retardation were tentatively assigned to interval 2. Deletions of intervals 4 through 7 were not associated with any clinical abnormality. We have constructed a preliminary long-range restriction map of intervals 2 and 3 using field-inversion gel electrophoresis. The DXS232, DXS121, and DXS233 loci are located on the same SfiI fragment, whereas the DXS165 and DXS95 loci could not be linked to this cluster using SfiI and SalI.

Endothelial plasminogen activator inhibitor (PAI): a new member of the Serpin gene family.

A human endothelial cDNA expression library, based on the Escherichia coli plasmid pUC9, was screened with a heterologous antibody raised against purified bovine aortic endothelial plasminogen activator inhibitor (PAI). A synthetic oligonucleotide, derived from a partial PAI cDNA expression clone, was used to select a full-length PAI cDNA, the size of which coincides with the length of PAI mRNA (approximately 2350 nucleotides) as determined by Northern blot analysis. The authenticity of full-length PAI cDNA is demonstrated by the expression of biologically active PAI both in lysates of transformed E. coli cells and in conditioned media of mouse Ltk- cells, transfected with PAI cDNA inserted into vector pSV2. Analysis of the de novo synthesized anti-plasminogen activator activity, employing reverse fibrin autography, shows that transfected mouse Ltk- cells synthesize a polypeptide with a mol. wt identical to that of the native PAI glycoprotein (Mr 52,000), whereas in E. coli an unglycosylated, active product with a mol. wt of 43,000 is made. The amino acid sequence, derived from the determined nucleotide sequence, shows that pre-PAI consists of 402 amino acids. It is proposed that the mature PAI is preceded by a signal peptide of 23 amino acid residues. The amino acid sequence of mature PAI includes three potential asparagine-linked glycosylation sites and lacks cysteine residues. The predicted amino acid sequence reveals significant homology with members of the serine protease inhibitor (Serpin) family, e.g. alpha 1-proteinase inhibitor and antithrombin III.(ABSTRACT TRUNCATED AT 250 WORDS)

Full-length von Willebrand factor (vWF) cDNA encodes a highly repetitive protein considerably larger than the mature vWF subunit.

Full-length human von Willebrand factor (vWF) cDNA was assembled from partial, overlapping vWF cDNAs. This cDNA construct includes a coding sequence of 8439 nucleotides which encode a single-chain precursor of 2813 amino-acid residues, representing a putative signal peptide, a prosequence and mature vWF of 22, 741 and 2050 amino acids, respectively. This represents the longest coding sequence determined to date. In-vitro expression of full-length vWF cDNA revealed the synthesis of a polypeptide with a mol. wt corresponding with that of the unglycosylated precursor. The precursor is a highly repetitive protein which consists of two duplicated (B, C), a triplicated (A), a quadruplicated (D) and a partly duplicated domain (D'), in the following order: H-D1-D2-D'-D3-A1-A2-A3-D4-B1-B2-C1-C2-OH. Both the prosequence, composed of two D domains (D1, D2), and mature vWF harbor an arg-gly-asp ('R-G-D') sequence which has been implicated in cell-attachment functions. It is argued that the pro-sequence is equivalent to von Willebrand Antigen II (vW AgII).