The fungus Colletotrichum gloeosporioides is a plant pathogen that causes wilting
and mortality of annual and perennial Limonium plants from the Plumbaginaceae
family. In the Arava desert, Limonium is an important export flower crop, produced
by tissue culture. The growers suffer extensive economic damage that also includes
the need to replace expensive seedlings.
the winter, and mortality occurs when the temperatures increase in the summer and
autumn (June-October), a phenomenon that began in 1995. Plants may be infected at
the nursery stage, and in addition, the pathogen can be transferred from infected to
healthy material in the field – during pruning and picking, or via the irrigation system,
that spreads the fungal conidia. Under suitable humidity conditions, the pathogen may
affect all plant parts including crown, foliage, stems and the flower pedicles.
The disease is usually diagnosed according to classical methods based on
morphological criteria such as conidia size and shape, and colony growth rate, which
are not considered accurate parameters for detection. Therefore, molecular techniques,
using specific PCR amplification, which are more accurate for identification, have
is even more )been recently adopted. The quantitative real-time PCR (QRT-PCR
sensitive than regular PCR and is based on identification via reaction and
measurement of it's kinetics. This method enables identification of the pathogen in planta,
and can also measure quantitative and qualitative levels of infection.
In this study we used two isolates of the fungus Colletotrichum gloeosporioides,
isolates no. 28 and 48A, which originated from diseased Limonium plants. They were
chosen due to their high percentage of conidial germination and high disease severity
in inoculated plants, relative to other isolates that were examined. In preliminary
experiments two plant inoculation methods were examined, spray and dipping in
conidial suspensions. The dipping method was chosen due to a more rapid disease
response and higher disease severity.
Two local Limonium cultivars, 'Supreme' and 'Safora', were evaluated for their
response to the pathogen by: a. visual estimation of symptom severity on leaves and
seedling mortality; b. isolation from different plant parts on a semi-selective media. In
addition, DNA was isolated from infected plants and examined for presence of the
pathogen by QRT-PCR with specific primers for the internal transcribed spacer region
(ITS) and β-tubulin (TUB I) fungal genes.
Plants from both cultivars were incubated in the greenhouse by dipping them in
different concentrations of conidial suspensions (103, 104, 105, 106, 107 conidia per
ml). Percentage of pathogen colonization was higher in above-ground parts compared
to that in the below-ground parts, indicating that the source of the fungus is from
foliar inoculation. Disease severity increased with higher concentrations of conidial
suspensions. In addition, the 'Safora' cultivar was more tolerant to the pathogen
compared to 'Supreme'. Symptoms that appeared on leaves of inoculated plants in the
greenhouse disappeared in time in plants following field transplantation and isolation
of the pathogen from the inoculated plants became more difficult. Mortality of
inoculated 'Supreme' plants reached 80% a few months after inoculation, whereas
'Safora' plants remained healthy. 'Supreme' plants from the non-inoculated controls
also died, indicating that infection spread via roots and/or the irrigation system, after
they were transplanted to the field.
Regular PCR reaction enabled identification of fungal DNA in the presence and
absence of plant DNA at a threshold concentration of 30 pg/reaction. However, there
was difficulty in identification of the fungal DNA in inoculated plants.
Calibrations of QRT-PCR reactions were performed successfully with primers for the
C. gloeosporioides fungal ITS and TUB I genes, and for the Limonium 18S gene. In
order to identify the fungus in infected plants by QRT-PCR, primers that identified
the TUB I gene were chosen, since their reaction values were closer to the ideal
reaction values when fungal DNA was examined in the presence of plant DNA. This
was done even though the identification of the TUB I gene was performed in a later
cycle in comparison with the identification of the ITS gene, for the same fungal DNA
QRT-PCR reactions enabled qualitative but not quantitative identification of the
infection stages – as early as one day post-inoculation with higher inoculum densities.
However, as time elapsed, more plant samples inoculated with lower inoculum
densities were identified as colonized by the pathogen. The identification of the