Study area

The study was carried out in a forest core of 3 000 ha of which 265.21 ha have been established with species of the Eucalyptus genus. The plantations are located on the plateau of San Pedro in the municipality of Villanueva, Department of Casanare, Colombia, at the geographical coordinates 4°57 "N and 73°94" W with an average altitude of 420 m.

The study area is located in the tropical moist forest, with a monomodal precipitation regime which favors the development and dispersal of eucalyptus cancer throughout the year (excepting December, January, February and March), since C. cubensis presents optimum growth at elevated temperatures and precipitation of more than 2 000 mm annually (^{Gryzenhout et al., 2004}).

Obtaining of isolates and identification of the pathogen

Samples of plant material with symptoms and signs of Eucalyptus cancer were obtained from bark, wood and apical meristems of two-year old E. pellita and E. urophylla S. T. Blake four-year-old trees. All samples were placed in properly labeled plastic bags and transported in a portable refrigerator at a temperature of about 4 °C (^{Bernal et al., 2009}), prior to their analysis in the Health Laboratory of the Distrito Francisco José de Caldas, Vivero headquarters, in Bogotá.

The thirty-one field-collected samples were individually disinfected in a 2 % sodium hypochlorite (NaCl) solution for two minutes, and rinsed twice with sterile distilled water (^{Bernal et al., 2009}). Two replicates were obtained from each sample, one of which was kept in refrigeration at 4 °C and another was used to obtain reproductive structures of the fungus and the first isolates. Each sample was examined in detail with the help of a Discovery V8 Carl Zeiss stereomicroscope with which the ascostroma, perithecial neck and mycelium were identified on the plant tissue. Isolates were obtained from these samples by direct transfer of conidia masses, monosporic culture from suspension of conidia in sterile distilled water (ADE) and seeding sections of infected plant tissue in culture medium. In all cases, Merck-branded 2 % Agar Maltose (Merck) culture medium was used (^{Gryzenhout et al., 2004}; ^{Nakabonge 2006}; ^{Bernal et al., 2009}, ^{Juárez et al., 2013}). The isolates were preserved by the cryopreservation, lyophilization and sterile mineral oil methods.

Corroboration of the causal agent was based on macroscopic characteristics such as color and texture of the mycelium and microscopic morphologies by the observation and identification of conidiophores and conidia. The microscopic characterization of C. cubensis and C. doradensis Gryzenh. & M.J. Wingf. was performed by direct staining and staining in lactophenol blue. In addition, sections of the 25 μm thick infected material were obtained with a LEICA RM 2255 microtome and added with a non-coagulating fixative with chromic acid and acetic acid (adapted from ^{Reig et al., 2002}). Transverse sections of tissues were made with Lactophenol Blue and basic Fushina (^{González et al., 2011}). The macroscopic and microscopic characteristics were compared with that described for C. cubensis by ^{Gryzenhout et al. (2004}) and ^{Nakabonge (2006}) and C. doradensis were compared with those described by ^{Gryzenhout et al. (2005}). In addition, sequences of the molecular markers ITS4 and ITS5 were obtained using the sequencing and molecular analysis service of the Instituto de Genética de la Universidad Nacional de Colombia. (Genetics Institute of the National University of Colombia).

In order to select two virulent isolates, four of the isolates obtained from monosporic cultures were compared and inoculation was performed in 100 branches from ten clones of seven - year - old E. pellita following the methodology developed by ^{Juárez et al. (2013}). The length of the lesion developed from the point of inoculation with C. cubensis was used as the response variable and the treatments were determined by the clone and the inoculation strain (strain 17, 26, 29, 31 and control). Additionally, the daily mycelial growth of each isolate in culture medium in the Petri dishes was compared as another parameter equivalent to the invasion capacity that was used as an indicator of virulence (^{Díaz and Lecuona, 1995}; ^{Ochoa, 2004}). Based on these tests, isolates 17 and 26 (obtained in E. pellita) were selected to perform the assay.

Susceptibility of Eucayptus pellita and Eucalyptus urophylla seedlings to the inoculation by C. cubensis under greenhouse conditions

Seedlings from ten E. pellita clones and one from E. urophylla (Table 1) were displayed in 17 cm diameter plastic pots and kept in an isolated greenhouse built for this study. The second species was also selected for the inoculation in the greenhouse, from its resistance to the attack by C. cubensis (^{Alfenas et al., 1983}; ^{Van Heerden and Wingfeld, 2001}).

The assay was arranged in three completely randomized blocks determined by the isolation (17 and 26) and the other block for the control treatment (inoculation of seedlings with 2 % AM culture medium). Each block in turn was divided into sub-blocks, which corresponded to the number of replicates, so that in each sub-block all the clones evaluated were represented.

From the subcultures of each isolate of the pathogen, 4 mm diameter mycelial discs were cut with the aid of a pressure cuff and sterile needles. Each disc was stored inside pieces of sterile, non-adherent gauze within sterile Petri dishes, moistened with ADE.

To perform the inoculation, a wound was initially made at the base of the stem of each seedling, using a 4 mm diameter punch. Subsequently, the wet gauze section was placed with the mycelial disk, in direct contact with the wound made in the stem of the seedling and covered with Parafilm.

The development of the lesion was measured weekly for one month with a calibrator (Mitutoyo, CD 6 "CS of 0.01 mm accuracy) for a total of four measurements. At each measurement the precaution was taken to use nitrile gloves and seal again with Parafilm to prevent the entry of any type of contaminant.

Estimation of prevalence of the Eucalyptus cancer in a seven-year old E. pellita plantation

The prevalence of the disease (incidence and severity) was assessed in a stand made up by 570 seven-year old E. pellita individuals. Incidence was considered as the affected trees per cent, while severity was graded with a nominal scale for estimation of this condition of C. cubensis in E. pellita plantations, with a total number of six categories (adapted from ^{Paredes et al., 2010}). (Table 2).

Table 2 Scale for estimating Chrysoporthe cubensis Brunes Gryzenhout & M. J. Wingf.severity in Eucayptus pellita F.Muell. plantations. 

Source: ^{Paredes et al., 2010}.

The calculation of the average damage degree, incidence and severity was made as follows (^{Parra et al., 1999}):

Average damage degree (GMD)

GMD=∑((Ni *CDi)/(N))

Where:

i = 1 to 6

Ni = Number of trees in the damage category i

CDi = Damage category i

N = Total of measured trees

Incidence (INC)

INC = (Number of affected trees * 100)/total number of plants

Intensity or damage severity

I = (∑((Ni * CDi)/(K * N))*100

Where:

i = 1 to 6

Ni = Number of trees in the damage category i

CDi = Damage category i

N = Total of measured trees

K = Greatest category mayor

Data processing and analysis

Data obtained from daily mycelial growth of the isolates, lesion size of inoculated branches and inoculum seedling lesion size were organized and processed with descriptive statistics, and the assumptions of normality, homoscedasticity and independence were evaluated by means of the tests of Kolmogorov-Smirnov and Levene, prior to comparison using a variance analysis (ANOVA) and Duncan's post-hoc test. When the data did not present normal distribution, transformations were tested by the Box Cox method, using the SPSS 15.0 statistical program Otherwise, the analysis was performed using non-parametric statistics. The response variables evaluated were diameter of mycelial growth for the daily evaluation of isolates in laboratory as virulence factor and length of the lesion developed from the point of inoculation for the inoculation in branches and seedlings.

Pathogen agent

The affected trees in the field and from which samples of diseased material were taken showed symptoms and signs characteristic of the cancer caused by C. cubensis, such as swelling and fissures in the stem, gum redness associated with cracks on the cortex, presence of pycnidia, long pycnidial neck and globose base with reddish, coffee and black hue. In addition, the release of conidia was observed in mucilaginous masses of yellow tones accumulating on the ostium, composed of conidia (Figure 1). Conidiophors with bifurcations, hyaline with rectangular basal cells and globular appearance (Figure 2B-D). The conidia are unicellular, oval, non-septate hyaline, with bright exudates (Figure 2C), the size of conidia, conidiophores and pycnidia agree with those described in literature (Table 3). Isolates obtained from the field samples produced pycnidia and conidia after approximately 30 days of incubation (27 °C).

A. Grouping of pycnidia on wood. B - C. Pycnidia on bark with mass of conidia in the ostium. D-E. Pycnidia on wood with mass of conidia in the ostium.

Figure 1 Pycnidia of Chrysoporthe cubensis Brunes Gryzenhout & M. J. Wingf. with mass of conidia ejected through the ostium. 

A. Transverse section of pycnidia with conidia (blue-stained by Lactophenol Blue staining (10 X); B, D, Conidiophores (40 X, 100 X); C. Oval-shaped conidia (40 X)

Figure 2 Photomicrographs of Chrysoporthe cubensis Brunes Gryzenhout & M. J. Wingf. 

Table 3 Size of conidia, conidiophores and pycnidia of Chrysoporthe cubensis Brunes Gryzenhout & M. J. Wingf. 

Identification by DNA extraction and sequencing with the ITS markers are consistent in 99.9 % with sequences published in GenBank as C. cubensis for the isolation 17 obtained in E. urophylla (Access number JN942342.1) and C. doradensis for the strain 26 (Access number JN942330.1) from E. pellita.

The colonies of the four assessed isolates (codes 17, 26, 29 and 31) did not differentiate each other in terms of average mycelial growth (Kolmogorov-Smirnov, for K-S of a sample: 0.365). Strain 17 had the highest average of daily growth (0.75 mm day^-1), followed by strain 31 (0.74 mm day^-1). Strain 17 exhibited colonies that reached maximum diameter of mycelium (9 cm) at 5 days of measurement, and for strain 31 (9 cm) at 8 days of measurement (Table 4).

The inoculation of eucalyptus branches with the isolates with Chrysoporthe induced the development of typical lesions (Figure 3) and were significantly higher in comparison with the control (inoculated with AM), (df = 4, p> 0.05, p = 0.000). The average length of lesions was 63.4 mm while in the control, 26.3 mm. No significant difference was observed between the strains in the lesion size nor in the lesion induced in the different clones (Table 5).

Figure 3 Injury caused by Chrysoporthe cubensis Brunes Gryzenhout & M. J. Wingf. in Eucayptus pellita F.Muell. branches in the second week of inoculation. Necrosis in branches of E. pellita associated to the inoculation point of C. cubensis. 

Table 5 Sub-sets (1, 2) statistically different from mean length of the lesion caused by four strains of Chrysoporthe sp. in eucalyptus branches (Duncan's test). 

Duncan. Average

The means for the groups in the homogeneous subsets are shown in Table 5. The sample size of the harmonic mean was 20.00.

Susceptibility of greenhouse seedlings

All clones evaluated showed susceptibility to inoculation with strains 17 and 26 of Chrysoporthe (df = 222, p <0.05); however, no statistically significant differences were found between the 11 clones nor between the two assessed Chrysoporthe isolates (Table 6).

Table 6 Length (mm) of injury after 53 days of inoculation with Chrysoporthe cubensis Brunes Gryzenhout & M. J. Wingf. 

Seedlings developed progressive and rapidly advanced symptoms of cancer. After 53 days of inoculation with Chrysoporthe, they presented increased wound length, forming larger sores. In the inoculated seedlings in turn, exudation of reddish shade of viscous and bright texture was evidenced, which is also characteristic in Eucalyptus trees affected by the fungus; also, these seedlings formed additional sores at the inoculation site and which were displayed along the stem (Figure 4).

A, B, C, D. Seedlings with presence of exudates of reddish hue at the inoculation site. E. Primary bifurcated or dead meristems. F. Galls along seedling stem.

Figure 4 Development of wounds caused by Chrysoporthe infection in Eucayptus pellita F.Muell. seedlings, 53 days after inoculation. 

The disease progressed gradually and caused the death of several of the evaluated seedlings (Table 7). Treatment with strain 26 brought together the largest number of dead individuals (25). Clones 5 026 (20 %) and 5 002 (16 %) achieved the highest mortality rates attributable to treatment.

Table 7 Absolute and relative Eucalyptus seedling mortality inoculated with Chrysoporthe per treatment. 

With strain 17, a 22.2 % mortality rate was verified, corresponding to 20 of the total assessed (90) in this treatment; the clone with the highest number of deaths was 5 013 with 30 % of the seedlings in the treatment described, followed by 5 632 (20 %). Finally, in the control, the death of five seedlings was recorded in the last measurement, which constitutes 5.68 %, of which three were of clone 1 110 (Table 7).

Prevalence of Chrysoporthe sp. in plantations

Of all clones tested (20), 54 % had an incidence of 90 to 100 %. Despite the high values, 51 % had a severity between 50 and 60 %, so the disease was not in advanced stages of development.

The 570 trees evaluated in the plantation of E. pellita revealed an average degree of damage (GMD), that between 0.6 and 3.5, that is, with mild to intermediate conditions. However, 51.3 % of the clones had mean damage levels of 2.5 to 3.0, corresponding to deep fissures with exudate of reddish tonality, reproductive structures (pycnidia) and, in some cases, swelling at the base of the tree.

At the plantation, the ten clones of E. pellita evaluated in greenhouse had an average degree of damage between 2.47 and 3.33. They exhibited bark detachment and reddish exudations mainly in clones 1 110 and 5 026. On the other hand, the incidence of C. cubensis of these clones in particular, is damaging between 80 and 100 % of the total of individuals, with severity of 49 % and 67 %.