Genetic control of the conversion of dihydroflavonols into flavonols and anthocyanins in flowers of Petunia hybrida

Petunia hybrida mutants, homozygous recessive for one of the genes An1, An2, An6, or An9 do not show anthocyanin synthesis in in vitro complementation experiments per se (see also Kho et al. 1977). Extracts of flowers of these mutants all provoke anthocyanin synthesis in isolated petals of an an3an3 mutant. Mutants homozygous recessive for one of the genes An1, An2, An6, or An9 and homozygous recessive for F1 accumulate dihydroflavonols in comparable amounts. The synthesis of dihydromyricetin is blocked in an1an1 mutants, which indicates a regulating effect of the gene An1 on the gene Hfl. Similar mutants, but dominant for F1, accumulate flavonols (kaempferol and quercetin) instead of dihydroflavonols. Myricetin is accumulated in minor amounts and not at all in an1an1 mutant. Furthermore, the synthesis of this flavonol is not controlled by the gene F1. The synthesis of cyanidin (derivatives) is greatly reduced when flavonols are synthesized (F1 dominant). In mutants dominant for Ht1 and Hf1 and thus able to synthesize cyanidin (derivatives) and delphinidin (derivatives), predominantly delphinidin (derivatives) are synthesized. The results indicate that kaempferol (derivatives), quercetin (derivatives), and delphinidin (derivatives) are the main endproducts of flavonoid biosynthesis in Petunia hybrida.     

Recombination behaviour and gene transfer in Petunia hybrida after pollen irradiation

Summary The genes An2, Rt and An1 are located in chromosome VI and closely linked. Pollination of the triple recessive line W127 (an2an2rtrtan1) with irradiated pollen of the triple dominant line M1 (An2An2RtRtAn1An1) led to the recovery of at least 3.3% induced an2 recessives. Karyotype analysis and genetic data showed that these mutants all contained a deletion on the short arm of chromosome VI, ranging from non-detectable (a non-transmissable mutant, showing no visible deletion) to the complete short arm. It is concluded that An2 is located distally in the short arm, Rt and An1 in the long arm of chromosome VI. Deleted chromosomes are not transmitted to the next generation, neither through the male nor through the female; transmission of the dominant markers in the long arm of chromosome VI is possible after completion of the chromosome by crossing-over. There is a relationship between the length of the deletion in the short arm and the recombination frequency between the markers (Rt and An1) in the long arm: recombination increases with increasing length of the deletion. After completion of the chromosme by crossing-over, the normal recombination frequency is restored.     

Determination of copy number and linkage relationships among five actin gene subfamilies in Petunia hybrida

The actin gene superfamily of Petunia hybrida cv. Mitchell contains greater than 100 gene members which have been divided into several highly divergent subfamilies [1]. Five subfamily-specific probes have been used to compare the actin genes among the Mitchell, Violet 23 (V23) and Red 51 (R51) cultivars of P. hybrida. The sum total of actin genes in these five subfamilies was estimated to be between 10 and 34 members in both V23 and R51. Restriction fragment length polymorphisms (RFLPs) between V23 and R51 were examined with these five probes and eleven different restriction endonucleases. Among the 55 comparisons, 87% exhibited RFLPs. These data indicate extreme divergence between V23 and R51 in DNA sequence and/or the presence of small insertions and deletions surrounding these actin gene subfamilies. This divergence suggests that V23 and R51, which have contrasting phenotypic marker loci on every chromosome, may be useful for the development of a complete RFLP linkage map of the Petunia genome. The segregation of Hind III RFLPs among the progeny of two backcrosses demonstrated that representatives of the five subfamilies of Petunia actin genes exist at four distinct genetic locations and suggested that two of these loci are tightly linked. Apparently, amplification of the numerous members of the Petunia actin gene superfamily occurred via gene dispersal of the original subfamily progenitors and not primarily as a result of amplification of a single chromosomal region.     

Molecular characterization of a nonautonomous transposable element (dTph1) of petunia.

An insertion sequence of 283 base pairs has been isolated from the DFR-C gene (dihydroflavonol-4-reductase) of petunia. This insert was found only in a line unstable for the An1 locus (anthocyanin 1, located on chromosome VI) and not in fully pigmented progenitor and revertant lines or in stable white derivative lines. This implies that the An1 locus encodes the DFR-C gene. The unstable An1 system in the line W138 is known to be a two-element system, the autonomous element being located on chromosome I. In the presence of the autonomous element, W138 flowers exhibit a characteristic pattern of red revertant spots and sectors on a white background. In the absence of the autonomous element, the W138 allele gives rise to a stable recessive (white) phenotype. Sequence analysis of progenitor, unstable, and revertant alleles revealed dTph1 to contain perfect terminal inverted repeats of 12 base pairs. In DFR-C, it is flanked by an 8-base pair target site duplication. Sequences homologous to dTph1 are present in at least 50 copies in the line W138. Sequence analysis of An1 revertant alleles indicated that excision, including removal of the target site duplication, is required for reversion to the wild-type phenotype. Derivative stable recessive alleles showed excision of dTph1 and a rearrangement of the target site duplication. dTph1 is the smallest transposable element described to date that is still capable of transposition. The use of dTph1 in tagging experiments and subsequent gene isolation is discussed.     

Endothelial cell junctions

In the course of a freeze-cleave study on intercellular junctions in the regenerating rat liver, we observed an unusual array of intramembranous particles located in regions of contact between endothelial cells lining the hepatic sinusoids. These arrays were characterized by an accumulation of particles which resembled a zonula occludens in their linear deployment but differed in that the contact regions were composed of individual particles which remained separated from each other by regular particle-free intervals.     

Can Nep1-like proteins form oligomers?

Nep1-like proteins (NLPs) are a novel family of microbial elicitors of plant necrosis that induce a hypersensitive-like response in dicot plants. The spatial structure and role of these proteins are yet unknown. In a paper published in BMC Plant Biology (2008; 8:50) we have proposed that the core region of Nep1-like proteins (NLPs) belong to the Cupin superfamily. Based on what is known about the Cupin superfamily, in this addendum to the paper we discuss how NLPs could form oligomers.Key words: quaternary structure, necrosis and ethylene inducing proteins, NLPs, MpNEP1, MpNEP2, NPP1, Moniliophthora perniciosa, Phytophthora parasiticaCupins may be organized as monomers, dimers, hexamers and octamers of β-barrel domains.¹ To the best of our knowledge trimers have not been detected yet. The interaction of two monomers building up a dimeric structure is basically performed by three types of interactions: hydrophobic interactions between β-strands in different subunits, salt bridges and hydrogen bonds between β-strands. In cupin dimers, the hydrophobic interactions occur between two βI strands in different subunits (Fig. 1A and B). This strand represents the central axis of rotation of the dimer as one residue in βI interacts with the corresponding residue in the other subunit (Fig. 1B). Therefore, all residues in βI must be hydrophobic, as one residue interacts with the other subunit and the next one in the sequence interacts with the interior of the protein. Charged residues in βI would disrupt such interactions. Most cupin dimers have strong hydrophobic residues such as tryptophan (W), phenylalanine (F) and methionine (M) pointing towards the own subunit (↓), while small hydrophobic residues such as leucine (L), isoleucine (I), and valine (V) point to the other subunit (↑). A particular case is leucine that interacts with other subunits, for instance, βI = liaW (positions 217–220 in Fig. 1B) and βI = LVsw of type I and II NLP consensuses, respectively. Therefore, the pattern of hydropathicity suggests that the side chain orientation is βI = l217 ↑ i218 ↓ a219 ↑ W220 ↓ d221 ↑. However we observe that just after βI there is a charged residue (aspartate D221) which would point outwards disrupting the dimer or at least making it less stable. It is interesting to observe that the requirement for a negatively charged residue at this last position is very high: 96% of all type I NLPs contains an aspartate (D) or glutamate (E) indicating an important role for it, maybe in avoiding dimerization of the NLPs. A second interesting hypothesis is as follows: several cupins are oxygenases, decarboxylases, etc. and use a negatively charged residue, such as aspartate or glutamate as proton donor.¹ Now, if the alternate pattern of side chains of the residues is βI = l217 ↓ i218 ↑ a219 ↓ W220 ↑ d221 ↓, instead of the previous one, then the aspartate or glutamate residue would point to the hydrophobic pocket and would be positioned to interact with the metal ion, as in cupins with enzymatic activity. However, there are no experimental evidences that the NLPs have enzymatic activity.Open in a separate windowFigure 1(A) Three-dimensional structure prediction for type I NLP consensus, (B) Interface between two βI strands in type I NLP consensus. From the left to the right: EF-coil with the conserved residue H162, βC and βH strands (superposed) with the conserved histidines H133 and H135 in βC, H193 and leucine L195 in βH, W220 in βI and W118 in βB. The strands in the right subunit follow the same pattern but rotated.The second type of interaction is salt bridges between charged residues in different subunits. Analyzing all interacting side chains in the 1VJ2 protein (dimer), we verify that the charged side chains of N35 and E57 (numbers in original structure) are only 2.72 Å apart. In the NLPs, this corresponds to N108^36% (Q108^60%) at the border of βB and E138^98%. The negatively charged residue D125 helps to correct the orientation of the subunits in relation to each other avoiding any disorientation. The high conservation level of these residues suggests that NLPs are dimeric structures. However, as we will see next, only hydrophobic and charged interactions are not enough to build a dimer.Garcia et al. (2007)² have used small angle X-ray scattering (SAXS) to show that, in solution, at low concentrations (<2 mg/ml) the two copies of the NLPs of Moniliophthora perniciosa, MpNEP1 and MpNEP2, exist as dimers and monomers, respectively. The same technique showed that at higher concentrations, >5 mg/ml, both proteins exist as dimers, as is the case for PpNPP1.² They also reported, based on electrophoresis analysis, that PpNPP1 and MpNEP1 exist as oligomers and MpNEP2 as monomers.² However, experiments with the PpNPP1 in size exclusion chromatography using myoglobin as size standard suggest that PpNPP1 is a monomer.³ Figure 2 compares MpNEP1, MpNEP2 and PpNPP1, where the most relevant differences in sequence are marked with asterisks (*) and are possibly related to the differences in oligomeric properties between MpNEP1 and PpNPP1 with MpNEP2. These positions are methionine M27 and leucine L35, which occur only in MpNEP2, glycine G250, which occurs only in MpNEP2 and NEP1 (Fusarium oxysporum) and lysine K31, which occurs only MpNEP2, {"type":"entrez-protein","attrs":{"text":"BAB04114","term_id":"10173008"}}BAB04114 (Bacillus halodurans) and {"type":"entrez-protein","attrs":{"text":"AAU23136","term_id":"52003194"}}AAU23136 (Bacillus licheniformis). The other residues are aspartate D28, which occurs 9 times and alanine A37 which occurs 7 times of all investigated NLPs. Thus, the sequence mdHDkiakl at the start of the NLPs seems to explain the monomeric state of MpNEP2, although at higher concentrations they form dimers. Besides the weak hydrophobic interactions, dimeric cupins and bicupins (two β barrels in the same sequence building up a dimeric-like 4d-structure) are stable structures (see Fig. 1 in ref. ⁴). By aggregating the first β-strand in the start domain of one β-barrel to the ABIDG β-sheet of the other β-barrel, composing a big ABIDGY β-sheet (Y is the first β-strand). For instance, using the bicupin 1L3J (oxalate decarboxylase) as template, the low confidence level β-strand at position 26–33 (v in H29D30 avv) in type I NLPs corresponds to the first β-strand. Since this proceeds from both barrels they can build a stable structure (see Fig. 1 in ref. ⁴). The quaternary structure is related to the presence of interaction residues in the BID β-sheet of the cupin structure. These are present in the NLPs and would enable them to form dimers.Open in a separate windowFigure 2Alignment of type I NLP consensus, PpNPP1, MpNEP1 and MpNEP2. Solid line boxes are β-strands, double line boxes are α-helices. The sequence positions marked with asterisks (*) are possibly related to the differences in oligomeric properties between MpNEP1 and PpNPP1 with MpNEP2.     

Analysis of the Petunia TM6 MADS box gene reveals functional divergence within the DEF/AP3 lineage

Antirrhinum majus DEFICIENS (DEF) and Arabidopsis thaliana APETALA3 (AP3) MADS box proteins are required to specify petal and stamen identity. Sampling of DEF/AP3 homologs revealed two types of DEF/AP3 proteins, euAP3 and TOMATO MADS BOX GENE6 (TM6), within core eudicots, and we show functional divergence in Petunia hybrida euAP3 and TM6 proteins. Petunia DEF (also known as GREEN PETALS [GP]) is expressed mainly in whorls 2 and 3, and its expression pattern remains unchanged in a blind (bl) mutant background, in which the cadastral C-repression function in the perianth is impaired. Petunia TM6 functions as a B-class organ identity protein only in the determination of stamen identity. Atypically, Petunia TM6 is regulated like a C-class rather than a B-class gene, is expressed mainly in whorls 3 and 4, and is repressed by BL in the perianth, thereby preventing involvement in petal development. A promoter comparison between DEF and TM6 indicates an important change in regulatory elements during or after the duplication that resulted in euAP3- and TM6-type genes. Surprisingly, although TM6 normally is not involved in petal development, 35S-driven TM6 expression can restore petal development in a def (gp) mutant background. Finally, we isolated both euAP3 and TM6 genes from seven solanaceous species, suggesting that a dual euAP3/TM6 B-function system might be the rule in the Solanaceae.