Collection of parental material

Plant material of lulo cv. La Selva was collected in the town of Caucayá, municipality of Belén de Umbría, Risaralda (Colombia), with the following geographic coordinates: 05°13' N and 75°48' W, at an altitude of 1,320 m a.s.l., with an average temperature of 26 °C and an annual rainfall of 2,150 mm.

Cuttings of basal suckers from selected plants were obtained due to their excellent production and good phytosanitary condition. Cuttings were planted in plastic bags with a substrate composed of three parts of soil plus one of organic matter. Once planted, a drench application with Trichoderma spp. was made to prevent the attack of the damping-off complex (Fusarium oxysporum, Phytophthora infestans, Sclerotinia sclerotiorum) (ICA, 2011).

For the case of natural selection of lulo cv. Castilla Larga Vida, fruits were harvested from the best plants of a commercial cultivation that were isolated from any other cultivar in the town of Andica de Belén de Umbría, Risaralda. The location has an altitude of 1,370 m a.sl, an average temperature of 25 °C, an average annual precipitation of 2,030 mm, and have the following geographic coordinates: 05°14' N and 75°52' W.

Seeds were extracted from fruits weighing more than 150 g and a diameter greater than 8 centimeters, mixed as a mass selection and planted in a germinator. Twenty-eight days after planting, seedlings that emerged were registered, achieving a germination of 70.3 %. When seedlings emitted the third pair of leaves, they were transplanted to plastic bags, where they remained for four months until they had three true leaves.

Parent plant management

Seedlings from both parents were planted in a plot of Universidad Nacional Abierta y a Distancia (UNAD) in the municipality of Dosquebradas, Risaralda, located at an altitude of 1,460 m a.s.l., with an average temperature of 21 °C, an average annual rainfall of 2,200 mm, and an average relative humidity of 85 %.

Plants were sown in holes of 30 χ 30 χ 30 cm in a square arrangement with distances of 2 m between plants and 3 m between rows. We added 1.5 kg of composted organic matter to each hole as well as 150 g of commercial mycorrhizae to favor a better phosphorus absorption.

All the traditional cultural management practices obtained due to their excellent production and such as weed control, soil fertilization with NPK and secondary elements, as well as foliar fertilization with minor elements were carried out. Shape pruning was also done to eliminate shoots or suckers from the stem that were below 30 cm in height.

Crossings

Four months after transplantation flowering occurred. At that time, flowers with long pistils were selected in both materials (lulo cv. La Selva and lulo cv. Castilla Larga Vida), and reciprocal crosses were made, that is, the two materials were used as parent, i.e. as mother and as father plants. Flowers were emasculated removing the stamens, and the chosen flowers were pollinated as females. In figure 1, we can see the fruits resulting of these crosses during their development process.

Figure 1 Fruits in development obtained from the crosses of lulo cv. Castilla Larga Vida × cv. La Selva. 

From these crosses, fruits were obtained from which the Fi seed was extracted (figure 2). The seeds of all the F1 fruits of each reciprocal cross were kept separate as a mass selection. Three hundred and forty-four seeds of each of these crosses were sown and the percentage of germination was 73 %. One hundred F1 plants were selected from each crossing, which were transferred to plastic bags, and then a part of these was taken to the field to perform the morphoagronomic character evaluation.

Figure 2 F1 lulo fruits. a. cv. Castilla Larga Vida χ cv. La Selva; b. cv. La Selva χ cv. Castilla Larga Vida. 

The F1 plants of the crosses of the lulo cv. Castilla χ cv. La Selva, cv. La Selva χ cv. Castilla, and cv. La Selva χ cv. Larga Vida were sown in a plot located in the installations of UNAD, in the municipality of Dosquebradas, Risaralda. The study plots were located at an altitude of1,460 m a.s.l. with an average temperature of 21 °C, an annual rainfall of 2,200 mm and a relative humidity of 80 %. The region corresponds to a very humid premontane forest (bmh-PM) according to the Holdridge life zone classification.

Planting was carried out in a completely random design with 25 plants per treatment, for 100 plants in total. Every seven days the following quantitative morphoagronomic descriptors were obtained: plant height, stem diameter, fruit volume, number of leaves per week, number of buttons per cushion, number of cushions, number of fruits per bunch, number of bunches, and leaf area. From each of the treatments 10 rooted plants were plucked six months after sowing to evaluate level of root infestation by nematodes.

Nematode isolation

Pure cultures of M. incognita were obtained from infected roots of lulo cv. Castilla and from the soil surrounding these, using the Baermann funnel method.

Experiment set up for evaluation of resistance to Meloidogyne incognita

To evaluate the resistance of F1 plants to nematodes, seedlings with four or five well-formed leaves were planted in plastic jars with a capacity of 2 kg, with a substrate composed of three parts of soil and one part of organic matter without disinfecting.

A completely random design was used. The expe rimental unit corresponds to 10 plants, and the sampling unit was a complete plant by treatment and by repetition. Treatments correspond to the interaction of crosses by inoculation with or without nematodes, as follows.

T1: Fi plants of cv. Castilla Larga Vida χ cv. La Selva, without nematode inoculation

T2: Fi plants ofcv. Castilla Larga Vida χ cv. La Selva, inoculated with nematodes

T3: Plants of cv. Castilla Larga Vida, inoculated with nematodes (control)

T4: Fi plants of cv. La Selva χ cv. Castilla Larga Vida, inoculated with nematodes

T5: Fi plants of cv. La Selva χ cv. Castilla Larga Vida, without nematode inoculation

For plant inoculation, four holes were made in the soil inside each plot and around the neck of the root, where 50 cm³ of distilled water containing approximately 10,000 nematodes, between adults and eggs, were inoculated. These plants were kept inside a greenhouse for 60 days, at an average temperature of 23 °C and 85 % relative humidity.

At the end of this period, plants were extracted from their pots, and the presence or absence of nodes in the root system was evaluated, according to a scale of 0 to 10, where 0 corresponds to a healthy plant without root nodes, and 10 corresponds to plants with a root system completely covered in nodes. Plants were considered resistant when the mean root-node index was less than 2 and susceptible when that value was higher. Response variables used were level of infestation, root length and fresh root weight.

Multiplication of Neoleucinodes elegantalis

Adults of N. elegantalis were obtained from lulo fruits (from an undetermined cultivar) collected from plants in the area, which had entry holes but no exit holes. They were taken to the laboratory where four or five fruits were placed inside transparent plastic jars with wet sand and pieces of cloth paper in the bottom (figure 3). Once the larvae left the fruits, they started the pupation process on paper fabrics and adults emerged later. Adults obtained were sexed by wing color and by their genital opening.

Figure 3 Multiplication process of Neoleucinodes elegantalis in the laboratory using infested lulo fruits inside plastic jars. 

Experiment set up for the evaluation of tolerance to Neoleucinodes elegantalis

To evaluate the tolerance of the F1 generations to N. elegantalis, entomological sleeves (40 χ 15 cm) were used, selecting a branch of the middle third of the plant, with fruits in the first formation phase, in which two male and two female N. elegantalis moths were introduced.

Daily observations were made in order to record oviposition. For the evaluation of tolerance to N. elegantalis, a completely randomized design of the progenies of the reciprocal crosses of lulo cv. Castilla Larga Vida χ cv. La Selva, and as control, plants from cv. Castilla Larga Vida were used. An experi mental unit corresponded to a plant with its respective entomological sleeve, and a sampling unit with 10 plants per treatment.

Four quantitative descriptors were used: percentage of affected fruits, average number of exit holes per fruit, percentage of fruits with exit holes, and average number of larvae per fruit. Data obtained from the experiments in relation to the evaluation of resistance to nematodes and tolerance to N. elegantalis were subjected to an analysis of variance (Anova) and mean comparison tests (Duncan) using the statistical program SPSS V15.0.

Resistance evaluation of the F1 materials to the nematode Meloidogyne incognita

An analysis of variance for the variables level of infestation, root length and fresh weight of the root in seedlings 60 days after inoculation were carried out in function of the resistance of F1 plants of the reciprocal crosses of lulo cv. La Selva χ cv. Castilla Larga Vida to the nematode M. incognita; results showed highly significant differences between treatments.

Infestation level of Meloidogyne incognita in seedlings

When analyzing the level of infestation of the root by the nematodes in seedlings 60 days after inoculation, treatment 1 (F1 plants cv. Castilla Larga Vida χ cv. La Selva, without nematode inoculation) showed a level of nematode infestation of zero (0). Moreover, treatments 2 (F1 plants cv. Castilla Larga Vida χ cv. La Selva, inoculated with nematodes) and 5 (F1 plants cv. La Selva χ cv. Castilla Larga Vida, without nematode inoculation) showed a low percentage (2.5 %).

On the other hand, treatments 3 (cv. Castilla plants, inoculated with nematodes [control]) and 4 (F1 plants cv. La Selva χ cv. Castilla Larga Vida, inoculated with nematodes), were the most susceptible, with a level of infestation of 4.2 % and 4.7 %, respectively. Furthermore, 50 % of the plants of treatment 2 showed levels of infestation between 0 and 2 and can be suggested as resistant lines (figure 4).

Source: Prepared by the authors

Figure 4 Evaluation of the level of infestation of the nematode Meloidogyne incognita in roots of seedlings of F1 lulo plants. Same letters on the bars mean that they do not differ statistically (p > 0.05). 

Root length

There were significant differences among treatments. Roots of treatment 5 (F1 plants cv. La Selva χ cv. Castilla Larga Vida, without nematode inoculation) were the longest, with an average lenght of 40.5 cm, followed by those of this same material, but inoculated with nematodes (figure 5).

Figure 5 Root elongation affectation of F1 lulo plants by infestation of the nematode Meloidogyne incognita in seedlings. Same letters on the bars mean that they do not differ statistically (p > 0.05). 

Source: Prepared by the authors Although in this case we can say that nematodes affected root lengthening, this did not occur in the F1 plants cv. Castilla Larga Vida χ cv. La Selva material. This might be due to an oligogenic effect provided by the parent cv. Castilla Larga Vida when it is used as a male (figure 5).

Fresh weight of roots

Fresh weight of roots alone did not prove to be a good susceptibility or tolerance indicator to the nematode M. incognita, although in treatment 4 (hibrid plants: cv. La Selva χ cv. Castilla Larga Vida, inoculated with nematodes) there was a highly significant difference of 39.3 g compared to all other treatments (figure 6).

Source: Prepared by the authors

Figure 6 Root fresh weight in the seedling phase of F1 lulo plants evaluated for resistence to the nematode Meloidogyne incognita. Same letters on the bars mean that they do not differ statistically (p > 0.05). 

A higher root weight found in this treatment may correspond to the hyperplastic effect of an infection by M. incognita, which increased root volume compared to those that were not infected, as reported in the literature (^{Hafez & Sundararaj, 2000}; ^{Hashmi, Huettel, Hammerschlag, & Krusberg, 1993}, ^{Perrotta, Magnano, & Bassi, 1980}).

In addition, figure 7 shows the development of plants 60 days after the treatment was applied. In the first place, there is a lower plant vigor obtained from the crosses when the parent cv. Castilla Larga Vida was used as the mother; meanwhile the progenies of the crossing that used cv. La Selva as a mother showed a more vigorous development.

Figure 7 Lulo plants of five treatments cultivated in plastic containers 60 days after inoculation. 

Control plants of cv. Castilla Larga Vida inoculated with M. incognita, in addition to showing a poor development, showed yellowing, i.e. symptom of the affectation by the nematode. Apparently, the F1 materials inoculated with M. incognita showed no symptoms of such affectation when plant infestation was evaluated.

From the previous results and subsequent from the comparison of treatments Ti and T5, we can conclude that there is no cytoplasmic effect on the resistance or tolerance to the nematode; however, this can be a result of an oligogenic effect, since a greater resistance is observed when the lulo cv. La Selva is used as a male parent.

Nonetheless, as stated by ^{Dahms (1972)} , it is necessary to bear in mind that this type of experiment, in which high inoculum pressures are exerted through the artificial inoculation of treatments, may mask the results, so it may be possible that we cannot detect if these F1 materials are really resistant.

Nevertheless, it should be considered that, because they are normally cross-pollinated plants, and since cv. La Selva is the result of a backcrossing, both parents are heterozygous, and segregation in the progeny obtained from crosses may be present. This causes that each plant shows differential levels of resistance to these crop pests.

When performing a correlation analysis between variables, we observed a non-significant inverse relationship between levels of nematode infestation and root weight, and a direct, non-significant relation ship between that level of infestation and root length. On the other hand, the correlation between root fresh weight and root length shows a direct and highly significant relationship, as is expected accor ding to the biological logic in plant development. However, this is a fact that does not account for the tolerance or susceptibility to nematodes (table 1).

Table 1 Pearson's correlation analysis among response variables used in the evaluation of the susceptibility of lulo materials to the nematode Meloidogyne incognita 

* Significance at 1 %.

Source: Prepared by the authors

Infestation level of Meloidogyne incognita under normal field conditions

The F1 plants were evaluated under normal field conditions during the fruiting phase (six months after field transplantation). Because it was a destructive test, and since other phenological variables of the plants continued to be evaluated, only ten plants of each reciprocal crossing and of the evaluated cultivars were randomly removed; results show highly significant statistical differences between the materials related to the infestation ofM. incognita.

Based on the comparisons between the means of the evaluated materials regarding root infestation with M. incognita, the degree of infestation of both reciprocal crosses was less than 1 %, which indicates a high level of resistance of these materials to nematodes, e.g. cv. La Selva (0). On the other hand, cv. Castilla Larga Vida was highly susceptible (3.5 %) (figure 8).

Source: Prepared by the authors

Figure 8 Evaluation of the level of infestation of Meloidogyne incognita in the roots of two cultivars and two F1 materials of lulo under field conditions. Same letters on the bars mean that the values do not differ statistically (p > 0.05). 

Evaluation of the resistance to Neoleucinodes elegantalis

The F1 materials of both reciprocal crosses of lulo cv. La Selva x cv. Castilla Larga Vida did not show N. elegantalis larvae inside the fruits, nor exit holes in these. On the contrary, plants of the cv. Castilla Larga Vida used as a control were largely affected (60 %) and showed, on average, two exit holes and two larvae per fruit (table 2).

Table 2 Average values of the evaluated descriptors to establish the resistance of the Fi plants of the crosses of lulo cv. La Selva x cv. Castilla Larga Vida to the moth Neoleucinodes elegantalis 

Averages with different letters in the same column are significantly different.

Source: Prepared by the authors

The high percentage of fruits affected and the high average of larvae that attacked each fruit in the control treatment indicated that, using the entomological sleeves, the level of infestation was adequate to evaluate the resistance of the F1 materials obtained with the reciprocal crosses of cv. La Selva x cv. Castilla Larga Vida to the nematode M. incognita.

The F1 materials showed resistance to the moth N. elegantalis, probably due to an antixenotic effect in the adult insect, based mainly on physical and chemical factors caused by the presence of trichomes throughout the plant and especially in the fruits, which secrete terpenoids, alkaloids, phenolic substances and others. These can be olfactory or gustatory repellents for the insect (^{Aragão, Dantas, & Gandolfi, 2000}; ^{Cardona, & Mesa, 2011}; ^{Simmons, & Gurr, 2005}). However, it is necessary to continue with the observations of the lulo crop under commercial conditions.

Morphoagronomic behavior of hybrids

Hybrid plants of the reciprocal crosses proved to be vigorous, with an average height of 145 cm and a stem diameter of 4.5 cm, a good structure, large leaves, a length of 40.6 cm, a width of 41.5 cm, a petiole of 14 cm, and adapted to full solar exposure.

First flowering began 60 days after transplantation, with an average of four inflorescences per plant, which formed ca. 13 fruits, with an average diameter of 5.6 cm, similar to the lulo cv. Castilla Larga Vida. Fruit maturation and harvest occurred 155 days after the beginning of flowering.

Fruit epidermis was yellow and the pulp green, without dehiscence (cracking), and with pleasant aroma and flavor. However, plants showed abundant and pronounced thorns in stems, branches and leaves (figure 9), a condition that is classified as unfavorable by producers, since it hinders cultivation work.

Figure 9 F1 plants, four and a half months after transplantation to free sun exposure locations. a. Crossing of lulo cv. La Selva x cv. Castilla Larga Vida: stems and leaves with very strong thorns are observed in large quantity and with basal ramifications; b. Crossing of lulo cv. Castilla Larga Vida x cv. La Selva: shows thorns in stem and leaves.