The germination and viability of pollen are characteristics required for fecundation when individuals of different or the same species are crossed. For this reason, assessing these parameters in selected individuals to be used in breeding programs will increase the chances for the obtainment of new progeny. In this study, pollen from different accessions of the red-purple tree tomato (Solanum betaceum Cav.) was used: local cultivar (Morado Puntón), two commercial varieties (Large Red and Oratia Red) and six segregants [(Solanum unilobum x Solanum betaceum) x Solanum betaceum]. Three types of flowers were taken (A-day of anthesis, B-one day after anthesis, and C-two days after anthesis). The pollen was conserved in two temperatures (4° and 22° C) and four storage times (0, 5, 10, 20 days). The percentage of germination and pollen viability of the selected individuals were evaluated. It was observed that the commercial materials showed higher germination percentages than the segregants in flower A and B at a temperature of 4 ° C at all storage times, except for the segregants GT7P47 and GT7P48 at the same temperature on day 0. In addition, high percentages of viability were obtained both in flowers A and B, at both temperatures and at all storage times. However, the immediate use of pollen after it is collected is recommended because better germination is achieved. This study is helpful to improve breeding procedures in the initial stages of directed crosses.

Keywords. pollen quality, red-purple, tree tomato, Solanum betaceum, germplasm

Tree tomato (Solanum betaceum Cav.), native to the Andean region of South America¹, belongs to the Solanaceae family, is popular in this region for its consumption in juices and as fresh fruit. This fruit is characterized by its slightly bitter, sour, and astringent taste with a particular aroma^{2, 3}. In Ecuador, the cultivated area is around 2,000 ha and produces 28,512 tons with a yield of 13.79 t ha^{-1 (4)}. The cultivation of this fruit is carried out by small and medium producers⁵, but this fruit has excellent possibilities of positioning in the world market due to its excellent organoleptic characteristics, exotic aroma and flavor, and nutraceutical properties⁶.

In Ecuador, tree tomato production is based mainly on local cultivars, which are the product of natural crosses⁷. However, breeding processes have generated segregants, originating genotypes with significant phenotypic heterogeneity reflected in agronomic traits, fruit quality, and productivity^{5, 8}. Ecuadorian cultivars are not kept pure due to cross pollination that occurs in the production plots, generating fruits of different color, shape and size; being the most representative cultivars: Gigante Anaranjado, Anaranjado Puntón, Rendondo Anaranjado, Gigante Morado y Morado Puntón⁹; however, the cultivar Gigante Anaranjado is the one with the highest demand, consumption and production⁷.

Breeding on this fruit crop is carried out because tree tomato presents yield limitations mainly due to the attack of pests^{10, 11}. On the other hand, this fruit has excellent nutritional and commercial value, causing desired relevant demand in the national and international market¹². Furthermore, from the nutritional point of view, tree tomato fruit is an excellent source of vitamins A, B6, C, and E, and minerals such as iron; it also has a low carbohydrate content and less than 40 calories per 100 g¹³; it is a valuable source of pectins that favor the preparation of jellies and jams¹⁴. In addition, it has compounds with antioxidant capacity such as lycopene, anthocyanins, and high content of polyphenols such as 3-O-caffeoylquinic acid and rosmarinic acid^{3, 14}.

For those mentioned above, the Fruit Program of the National Agricultural Research Institute (INIAP) has evaluated tree tomato segregants with different levels of resistance and fruit quality (physical and chemical traits)^{5, 8}, to select elite materials to be used as parents in future crosses. Plant breeding through directed crosses is essential for generating new genotypes with better characteristics; consequently, pollen quality studies are essential in artificial hybridization techniques^{15, 16} to guarantee fertilization and the generation of new progeny.

Determining the quality of pollen is of great importance in breeding¹⁷ because it has a tremendous impact in the efficacy of the genetic improvement practices¹⁸. This characteristic is essential to define the direction of a cross and have bases for the success of controlled hybridizations that guarantee the generation of new hybrids19, originating superior individuals with better productivity, fruit quality, and obtain resistance or tolerance pests.

Currently, the Fruit Program has identified tree tomato individuals with superior characteristics associated with fruit quality, including soluble solids content and red-purple mucilage, a characteristic related to a more significant amount of antioxidant compounds³. These individuals will be used as parents in future crosses in the breeding program; therefore, it is necessary to study the pollen quality to guarantee an appropriate starting material to continue with the breeding of this fruit crop successfully.

The research was carried out in the Laboratory of the Tumbaco Experimental Farm. The pollen belonged to 1 local cultivar (CMP4 –Morado Puntón), 2 commercial varieties (NZLRP5 – Large Red and NZORP7 – Oratia Red), and 6 segregants of tree tomato (GT7p47, GT7p48, GT9p18, GT20p2, GT20p7, and GT33p5) coming from the cross [Solanum unilobum x Solanum betceum] x Solanum betaceum ^5,8. These individuals were selected for their fruit quality characteristics (Table 1, Figure 1).

The plants used for this study were grafted onto Nicotiana glauca, with an age of 4 years. They were sown in the Tumbaco Experimental Farm of the INIAP at an altitude of 2348 masl, with maximum temperatures of 27 ° C and minimum of 5 ° C, precipitation of 800 mm per year, average relative humidity of 70.86%, and geographical location of latitude: 00° 13 ’00» South and longitude: 78 ° 24′ 00″ West.

The pollen was stored in two temperatures (4° C and 22° C) and four periods of time (0, 5, 10, and 20 days) for its conservation. For the 0-day storage period, storage was performed for 8 hours at the two temperatures. Pollen extraction was carried out in the morning (8:00 – 10:00 am), taking fully open flowers in 3 states (Figure 2): where flower A (day 0) is the day of anthesis, flower B (day 1) one day after anthesis and flower C (day 2) two days after anthesis. To extract the pollen, the tips of the anthers were cut, and the pollen of the flower was obtained using light strokes and shaking.

A 1 mg sample of pollen was placed in Eppendorf tubes for each treatment. For the evaluation of pollen viability, the staining technique based on acetocarmine glycerol gelatin²⁰ was used. This test measures the integrity of the cytoplasm; that is, the pollen grains turn red when the cytoplasmic membrane is intact²¹. Viable pollen was considered to be those grains that did not show deformations and had intense staining. While for the evaluation of pollen germination, the pollen was sown for 24 hours in a sucrose medium described by Rodríguez and Dafni²². Germinated pollen grains were considered to be those that showed the pollen tube with a length greater than or equal to the pollen diameter^{23, 24, 25}. In both cases, 250 pollen grains were counted using an optical microscope (Olympus, SX40).

A completely randomized design was used, with a 9 x 2 x 4 x 3 factorial arrangement with a total of 216 treatments with 5 observations. An analysis of variance was performed, and the 5% Tukey test was used to determine differences between means.

Research carried out on pollen quality is of great importance for the success of breeding programs aimed at the generation of hybrids with better agronomic characteristics that allow increasing the efficiency of genetic improvement through directed crosses¹⁷.

To simulate pollen development in the gynoecium in vivo, the germination tests of pollen grains in vitro are established; this is achieved by placing the grains in a germination solution that must present similar conditions to the stigma of the female organ²⁶. In terms of germination percentage (Figure 3), significant differences were observed in flower type and temperature as storage days progressed (Table 2, 3, and 4). In the tree tomato, the size and production of pollen grains varies^7, and the type of flower to be collected is of utmost importance before pollen collection²⁷. In this research, it was observed that flowers A and B showed values ​​greater than 55% both at 22 ° C and 4 ° C on day 0, while in subsequent conservation times, values ​​greater than 40% were observed only at 4 ° C. On the other hand, in flower C, low values ​​(less than 13%) were obtained in the day 0 and the percentages were considerably reduced later until reach 0% in various individuals at 10 and 20 days. Consequently, while the storage time increases, the germination percentage decreases. This trend was also reported by Gonzáles et al. ¹⁶ who found that in potato (S. tuberosum), pollen stored at a temperature of 17 ° C loses its germination percentage rapidly, but if the pollen is stored at 4 ° C it had up to 20% germination. Araméndiz et al. ¹⁷ also reported this behavior in a study carried out on eggplant (S. melongena), where the highest germination percentage was obtained at 0 days of storage and it decreased as the storage time progressed and the temperature increased.

Statistical differences between the tree tomato individuals were observed, highlighting the commercial individuals CMP4, NZLRP5, and NZORP7 with values ​​in a range of 54 to 59% at 22° C and 53 to 64% at 4° C at 0 days of storage in flower A. The germination percentage was not optimum (less than 70%), even at 4° C, which is a temperature commonly used to conserve pollen grains in some species. Later these values ​​decreased, but CMP4 and NZLRP5 showed higher percentages, around 42%, at 20 days (Table 2). Only the GT7P47 and GT7P48 segregants showed germination percentages higher than 53% at 22° C and 4° C at 0 days and around 42% at 5 days at 4° C; while at 20 days, their germination decreased to values ​​of 16 and 26% respectively (Table 2). For flower B, the trend was similar to that observed in flower A with germination percentages in a range between 50 and 59% at 0 days and decreasing considerably until 20 days (Table 3). In flower C, only the segregant GT20P7 had values ​​of 12% (22 °C) at 0 days and 9% (4 °C) at 5 days, while the rest of the individuals showed shallow values ​​or did not have germination (Table 4). According to Revelo et al. ²⁸, this response is because commercial cultivars or varieties are better adapted to climatic and soil conditions, showing a more significant number of fruit set per inflorescence, corroborating that pollen from commercial materials has a superior germination capacity.

The determination of pollen viability allows making reliable fertility estimates, besides being used for incompatibility studies in crosses²⁹. In fruit crops, staining with 2% acetocarmine and observation by optical microscope are used to determine the viability of the pollen grain²⁶ (Figure 3). Similar to germination, significant differences were observed for the type of flower and temperature as the storage time advanced.

Flower A and B showed high percentages of viability (greater than 95%) (Table 2 and 3) in the commercial materials CMP4, NZLRP5, and NZORP7, as well as in the segregants GT7P47 and GT7P48, both at 22° C and at 4° C and at all storage times (0, 5, 10 and 20 days). While in flower C, only the segregating GT20P7 obtained 12% viability at 22° C at 0 days and 8% at 4° C at 5 days, while the rest of the individuals showed shallow values ​​, most of which did not have viability (Table 4). The results obtained in flower C corroborate what Montaner et al. ³¹ found that pollen loses its viability after its anthesis period.

In other Solanaceae species, using the same method of this research, viability percentages have been reported chiefly over 80% and reaching up to 98%^{26, 30}. On the other hand, in a study of pollen quality in chirimoya (Annona cherimola), it was reported that in storage at 7 °C, higher viability percentages are achieved as the conservation time advances³². This study obtained stable percentages throughout the pollen storage time at the two temperatures evaluated (Table 2 and 3).

The commercial materials (CPM4, NZLRP5, and NZORP7) showed high percentages of viability and relatively acceptable germination percentages (greater than 40%) in flower A and flower B at different storage times. On the other hand, the segregants GT7P47 and GT7P48 had a similar result only on day 0 of storage; after that, although they had high viability percentages, their germination was less than 40% even at 4 ° C. When pollen is preserved correctly, it maintains its hydration capacity and regular morphology. However, it loses its potential to germinate over time, so morphological stains are not a good technique for measuring actual fertility. It is an excellent tool to visualize the quality of pollen²⁷ because the staining technique only allows observing the pollen morphology related to its viability²². In addition, there are differences between the needs of pollen grains to germinate, and it is necessary to establish the environmental conditions (primarily temperature and humidity) that favor the development of the pollen tube^{26, 30}. In pollen studies carried out in Physalis peruviana, it has been reported that some accessions did not show pollen germination despite having good viability²⁶.

The authors thank the National Institute for Agricultural Research (INIAP) and the Spanish Agency for Development Cooperation (AECID) for their support and funding to carry out this research and Gabriela Altaminate for her participation in data collection.