The effects of the plant signal molecule acetosyringone (AS) and the osmoprotectant betaine phosphate (BP) have been examined for their ability to increase the transformation efficiency of Agrobacterium tumefaciens (At), C58C1::pGV3850 harboring the binary vector pKIWI105. This binary plasmid encodes the β-glucuronidase (GUS) gene and was previously shown to be expressed exclusively in plant tissues. Bacteria were grown in one of two previously reported virulence induction media (MS20 and SIM) for 5h and GUS activity was measured fluorimetrically in individual 6 week old leaf discs as a quantitative measure of stable transformation events. Bacteria induced in MS20 supplemented with AS (0.1 mM) and BP (1 mM) showed a significant increase in GUS activity as compared to media containing AS or BP added singly or control media lacking the supplements. The effects of another osmoprotectant proline (1 mM) could replace the beneficial effects of betaine. No significant difference was observed among treatments with respect to the two induction media.

Sharka is a severe apricot viral disease caused by the plum pox virus (PPV) and is responsible for large crop losses in many countries. Among the known PPV strains, both PPV-D (Dideron) and PPV-M (Marcus) are virulent in apricot, the latter being the most threatening. An F1 apricot progeny derived from Lito, described in the literature as resistant, crossed to the susceptible selection BO81604311 (San Castrese 9 Reale di Imola) was used to study the genetic control of resistance to PPV. A population of 118 individuals was phenotyped by inoculating both PPV-D and PPV-M strains in replicated seedlings and scored for 3 years. An additional set of 231 seedlings from the same cross was also phenotyped for 2 years. SSRbased linkage maps were used for quantitative trait locus (QTL) analysis. A major QTL of resistance to both PPV-M and PPV-D strains was found in the top half of the Lito linkage group 1, where a QTL was previously described in Stark Earli-Orange, the donor of Lito resistance. The LOD score was considerably enhanced when the recovery of plants from infection was taken into account. The results obtained in Lito were compared with those observed in a second apricot cross progeny (Harcot 9 Reale di Imola) in which QTL of resistance to sharka were also mapped in the same linkage group 1 for both PPV strains. Several models of resistance to sharka disease are discussed considering the segregation frequencies, the QTL alignment in the two maps and the information gathered from the literature.

Walnuts (Juglans spp.) are difficult-to-root woody plants. The rolABC genes (rolA + rolB + rolC), derived from the bacteria Agrobacterium rhizogenes, have been shown to increase the rooting potential of other difficult-to-root woody plants. We inserted the rolABC genes into somatic embryos of a `Paradox' hybrid (J. hindsii × J. regia) clone PX1 using the A. tumefaciens gene transfer system. A transgenic sub-clone, designated PX1 rolABC 2-2 was selected and compared to the untransformed clone for a variety of phenotypic characteristics, including rooting potential. Transformed and untransformed shoots were budded onto seedling J. regia rootstock in the greenhouse and established in the field. Transformed trees displayed reduced internode length, an increase in lateral branching, and wrinkled leaves. In another test, a commercial persian walnut cultivar J. regia `Chandler' was grafted onto rooted cuttings of both the untransformed and transformed plants. The presence of the rolABC genes in the rootstock had no visible effects on the grafted scion. Several of these trees were excavated from the field and the root systems of each genotype were examined for root number, diameter, and biomass. Trees with the rolABC rootstock had significantly more small diameter roots compared to the controls and less recovered biomass. Tests of the rooting potential of leafy semi-hardwood cuttings for two years resulted in 14% to 59% rooting of the transformed cuttings compared to 51% to 81% rooting of the control. Both transformed hardwood cuttings and microshoots in tissue culture also rooted significantly less (52% and 29% respectively) than untransformed hardwood cuttings and tissue cultured shoots (82% and 54% respectively). Thus, although the rolABC genes induced a shorter internode length and a more fibrous root system (typical of rol-tranformed plants), they were not useful for increasing the rooting potential, and as rootstock they did not affect the phenotype of the scion.

No abstract

No abstract

Seed bitterness, due to cyanogenic glucosides, has been reported in apricot as a recessive trait, being determined by a single gene. In this study, 21 F1 and 10 F2 populations from parents with either bitter or non-bitter ('sweet') phenotype were tested by seed tasting. Both the 'bitter' and the 'sweet' phenotypes were represented in populations from 'bitter×bitter' and 'sweet×sweet' crosses, as well as from self-pollination of either bitter-or sweet-seeded trees, providing evidence that more than one gene is involved in this trait. Ten populations showed segregation ratios inconsistent with a monofactorial inheritance of seed taste with the 'sweet' trait dominant over the 'bitter'. On the other hand, data from spectrophotometric assays indicate that seed cyanoglucoside content cannot be regarded as a quantitative trait. All the observed segregation ratios can be explained by an inheritance mechanism based on five, non-linked genes, involved in two distinct biochemical pathways. Three genes would control different steps in an 'additive' pathway (either the biosynthesis of cyanoglucosides, or their transport, or both) leading to accumulation of these metabolites in seeds:homozygosis of recessive alleles of at least one of them would result in the sweet phenotype. Two more genes would provide a cleaving activity, participating to cyanoglucoside catabolism; heterozygosis or homozygosis of dominant alleles at these loci would produce the 'sweet' phenotype, while homozygosis for recessive alleles of at least one of them would interrupt the catabolic pathway, leading to the 'bitter' trait, if associated with the anabolic function.

No abstract

Data collection and processing in fruit-tree breeding programs have been performed mostly manually, which is time-consuming and also can be the source of transcription errors. Breeders recently have had opportunities to implement a wide range of computer applications. These have included estimating the structure and genetic variation of populations (Dowling and Moore, 1984) and linkage and mutation frequency analysis (St. John et al., 1983). The availability of directly applicable programs, such as genotype-environment interaction characterization (Kang, 1985) and germplasm bank management (Styles et al., 1985), allows for increased effectiveness. Fruit breeding programs have used data processing procedures to optimize operational management (V. Beres and R. Scorza, personal communication). Except for the use of punch cards (Fogle, 1974), computerization of procedures is scarcely mentioned in the literature, except, recently, in forestry research for direct field data collection