Proceedings
of the XLVI Italian Society of Agricultural Genetics - SIGA Annual Congress
Giardini Naxos, Italy - 18/21
September, 2002
ISBN 88-900622-3-1
Poster Abstract -
4.55
CROSSED
MOLECULAR TOOLS FOR THE IDENTIFICATION OF PETAL COLOUR GENES IN ROSA GENUS
NUNZIATA
A.*, FOGLIANO V.**, MONTI L.***
*) Dept. of Soil, Plant and Environmental Sciences,
University of Naples "Federico II", Via Università 100, 80055
Portici, Italy
**) Dept of Food Sciences, University of Naples
"Federico II", Via Università 100, 80055 Portici, Italy
***) Res. Inst. for Vegetable and Ornamental Plant
Breeding, CNR, Via Università 133, 80055 Portici, Italy
Rosa, mass spectrometry, cDNA AFLP, molecular
markers
The
genus Rosa consists of species, sub-species and varieties whose taxonomy is
still under investigation. Cultivated roses are generally indicated as Rosa x
hybrida and they derive from a complex scheme of crosses, which involved only
10 of the 150 existing species of roses.
The high number of different existing
species of roses offers a continuous source of genetic variability, which
requires a wary planning and management of new strategies for genetic
improvement. In particular, total overturning of varieties at each new cross
should be avoided, but little variants of the same cultivar should be sought
for. As an example, an interesting objective for genetic improvement can be
getting lots of colour variants for the same cultivar.
In
Rosa, petal colour is a very complex trait. Rose colour depends on the pattern
of pigmented substances in vacuoles (among which flavonoids), on their spatial
distribution, and even on basal metabolic traits, such as pH. As shown in
Petunia, also copigmentation phenomenon occurring between anthocianins,
flavonols and metallic ions is likely to determine petal colour phenotype.
Subepitelial cell shape can determine velvet or metallic effects or flatten the
tonalities.
The
complexity of this trait has to be necessarily factorised to provide an
important tool for genetic improvement of the genus. In order to find molecular
markers associated to petal colour trait and to understand its metabolic and
genetic bases, two advanced molecular tools are proposed for quickly monitoring
the trait at different levels: cDNA AFLP analysis at the transcription level,
Maldi TOF mass spectrometry at end product level.
Flavonoid
extracts from mature petals of 64 genotypes have undergone mass spectrometry.
This analysis provided evidence of the presence of delphinidin and its
complexes with rhamnose and glucose. This anthocyanin has never been found
before in rose and it is thought to be responsible of blue colour in several
ornamental species. The elaboration of mass spectrometry results has given back
a binary matrix, obtained attributing 1 value to present substances, 0 value to
absent ones.
Among
these genotypes, 13 individuals with different petal colour have undergone RNA
extraction, cDNA construction and its AFLP analysys. Experimented techniques
for nucleic acid extraction have brought to the getting of good quality mRNA,
whose cDNA AFLP analysis has led to easily readable autoradiographic panels.
The analysis performed with two primer/enzyme combinations led to the
individuation of 181 polymorphic fragments. A binary matrix was build up
attributing 1 value to present bands, 0 value to absent ones.
Once
a consistent number of genotypes has undergone both mass spectrometry and cDNA
AFLP analysis, the two matrices will be compared each other in orther to find
out associations between markers and substances, and this can be done even
thought the chemical nature of the substance and the nucleic sequence of the
marker are unknown. This is due to the fact that both mass spectrometry and
AFLP analysis data allow the segregation of both known and unknown factors
determining a trait to be monitored. When a little number of substances will be
definitively associated to a little number of markers, further analysis will be
able to reveal more details about them. In particular, the markers that will be
associated to known pigmented substances will be rapidly isolated and
sequenced.