Introduction

Tomato (Solanum lycopersicum L.) is one of the three most important vegetables grown in the world (^{Hernández et al., 2014}; ^{Calero et al., 2019}), with China being the main producer with 34% of the total production (^{FAOSTAT, 2021}). The United States is the main importer of tomatoes and Mexico is the main exporter of this vegetable, allocating 99.7 % of its exports to that market. (^{SIAP, 2020}). Per capita consumption in the world in 2021 was 22 kg/year/inhabitant, and these quantities have been increasing in 2020, consumption in Mexico was 13 kg/year/inhabitant (^{SADER, 2022}).

Tomato is the most important vegetable at the national level due to its wide consumption, the harvested area, and the economic value of production (^{Abera et al., 2020}). Cultivation in the open field has decreased, while cultivation under cover has increased, largely due to the fact that tomato production under the latter system (shade net or greenhouse) has brought an increase in yield per unit area; in fact, tomato cultivation in the open field is becoming increasingly difficult due to adverse environmental conditions such as frost, rain, hail, dew and the incidence of pests and diseases (^{Gatahi, 2020}; ^{Juárez-López et al., 2012a}). On the other hand, high levels of quality, food safety, and product certification are achieved under protected conditions (^{Bojacá et al., 2009}).

Native species have developed tolerance to adverse conditions, which has allowed them to gain different traits to face biotic and abiotic threats, with the environment being one of the factors that most influence biological variability (^{Álvarez-Hernández et al., 2009}; ^{Sanjuan-Lara et al., 2014}). A decrease in genetic variation in modern tomato cultivars has limited crop improvement (^{Fridman et al., 2000}; ^{Bai & Lindhout, 2007}). The diversity available in wild relatives can be used to improve traits of interest; for example, the sensory traits of cherry tomatoes could be used to generate hybrids (^{Lecomte et al., 2004}; ^{Salgado-Meraz et al., 2018}).

It is worth mentioning that there are fundamental studies evaluating native tomato species in different states of the country (^{Estrada-Castellanos et al., 2011}; ^{Ríos-Osorio et al., 2014}; ^{Sanjuan-Lara et al., 2014}; ^{Maldonado-Peralta et al., 2016}); however, there are not reports comparing the adaptability of native tomato to a production system under cover for marketing purposes, since this genetic resource continues to maintain a wide use being found in local markets in different regions of Mexico (^{Bonilla-Barrientos et al., 2014}; ^{Magdaleno-Hernández et al., 2016}). The hypothesis of this research was that there are tomato landraces that could be adapted to an intensive production system under cover while maintaining their fruit quality features. Hence, the present research was carried out with the aim of evaluating the agronomic performance of 23 Mexican populations of native tomatoes and two commercial hybrids as a control, established under greenhouse and open field conditions.

Material and Methods

Genetic material

The native tomato (Solanum lycopersicon L. ) came from the states of Campeche (C1, C2, C3, and C4), Estado de México (M1 and M2), Guerrero (G1, G2, and G3), Oaxaca (O1, O2, O3, O4, and O5), Puebla (P1, P2, P3, P4, P5, and P6), Yucatán (Y1 and Y2) and Veracruz (V1), all collected by Dr. Porfirio Ramírez Vallejo^†; as a result of the Integral Evaluation Project of the Diversity of Native Populations of Mexican Tomato (CONACYT), in the Postgraduate Program in Genetic Resources and Productivity, Colegio de Postgraduados. It is worth mentioning that these tomatoes were preserved by the producers since they are sold in local and regional markets in Mexico (Figure 1).

The crop was established from February to June 2018 under greenhouse and open field conditions, where 23 populations of native tomatoes with indeterminate growth habits were evaluated and the commercial hybrids El Cid and Reserva (Harris Moran^® and Nunhems^®, Saladette type), referred to as H1 and H2 in this study, were used as controls.

Figure 1 Different shapes and sizes of Mexican populations of native tomatoes represent the place of collection. 

Results

With the exception of the variable locule number, the results in Tables 1 and 2 showed that most of the variables were significantly different (p ≤ 0.05) for the factors: population (Pob), production system (Sis), and the Pob x Sis interaction, demonstrating the presence of a significant genetic variation within the total population, a different agronomic behavior of the populations between the two production systems evaluated and a different efficiency of the production systems among the variables. However, the proportion of genetic effect or environmental effect differed between variables. Among the agronomic variables (Table 1), the number of fruits per plant varied mainly due to the effect of the Pob factor, with a proportion of 55 % with respect to the total variation of the treatments. In contrast, the Sis factor accumulated a greater proportion of the variation in the agronomic variables, plant height (61 %), number of leaves (81 %), weight per fruit (64 %), and yield (57 %). For these variables, the Pob x Sis interaction showed significant effects, with proportions of variation ranging from 9 to 38 %, which shows that not all populations respond to the same production system. This suggests that the behavior of the populations is highly dependent on the conditions they receive, as shown by the variable number of fruits and yield.

The fruit quality variables (Table 2) mainly influenced by the Pob factor were: polar diameter (42 %), equatorial diameter (43 %), and locule number (93%). Whereas, pericarp thickness (49 %), equatorial diameter (49 %), and TSS (43 %) showed a greater effect caused by the Sis factor. The interaction Pob x Sis was also significant for all variables, but with small magnitudes, ranging from 7 to 30 %.

Comparing the means of the Pob x Sis combinations (Tables 3 and 4), plants grown in the greenhouse (i=greenhouse) were taller than those grown in the field (c=field), with the control (H1) having the tallest plants in both growing environments and G2c having the smallest size. The number of leaves was higher in the native genotype V1c with 329 leaves; in the protected conditions a range of 27 to 40 leaves was obtained (G2i and Y1i).

The number of fruits ranged from 12 to 55 fruits per plant, with the O3c population having the highest number in the study with a value of 57 fruits per plant. The fruit weight under protected conditions was the highest of the G3i genotype with 179 g, while the lowest fruit weight under field conditions was obtained in the C3c genotype with 12 g. In this situation, the native genotype G3i produced the highest yield under protected conditions with 4.4 kg/plant. The yields obtained in the open field were very variable, the average production ranged from 0.4 to 2 kg/plant; the lowest yield corresponded to the genotype P4c and the highest to the genotype G3i.

Obtained values of the variables polar and equatorial diameter, locule number, TSS, and pericarp thickness from the fruits produced under protected conditions were higher than those obtained under field conditions. The largest polar diameter was presented by the control (H2i), with 7.3 cm, and the equatorial diameter by the O4i population, with 8.3 cm; it is worth mentioning that H2i presented elongated fruits, while those of O4i were wider, a variable that determines the packaging to be used for shelf-life. Similarly, the locule number was higher in the native genotype O3i, with 11.5 locules, and in the field, it was the control H2c, with only 2 locules. The highest TSS value was 5.5 % in genotype P1i and the lowest was 2.9 % in the native population O4c; and the greatest pericarp thickness was obtained by the native genotype P3i, with 10.6 mm, tomato produced in the greenhouse, while in field conditions the fruits showed the lowest thickness, with 1.5 mm (C1c).

In the averages of the agronomic and fruit quality variables in the protected and open field conditions, statistical differences (Tukey, 0.05) were found in most of the evaluated variables (Table 5). In the greenhouse, eight variables stood out, the most important being fruit weight with a difference of 200 % with respect to that observed in the open field, yield with 131%, and pericarp thickness with 92 %; in the open field, only two variables were superior to those obtained in the greenhouse: leaves number (434 %) and fruits number (21 %).

Discussion

The native tomato populations were significantly affected (p ≤ 0.05) by the production environment in which they were cultivated (Tables 1 and 2), as evidenced by the observation of significant differences in the factors system, genotype, and system-genotype interaction for most of the variables. These results indicated that some populations improved or decreased their yield and fruit quality attributes under certain production system conditions. In this sense, ^{Zanne et al. (2014}) suggested that changes in the morphology and physiology of certain populations could be a consequence of evolutionary adaptations to environment in which plants grow.

The production system factor explained to a greater extent the variation in the effects on the variables plant height, number of leaves, weight per fruit, fruit yield, equatorial diameter, pericarp thickness, and TSS, while the population-production system interaction influenced to a greater extent the number of fruits. According to ^{Bhandari et al. (2021}), this is explained by the climatic condition which is a factor affecting quality and yield in tomato production. However, such a situation could be better managed in the greenhouse by partially controlling climatic variables to favor conditions for plant growth (^{Holcman et al., 2017}).

In the mean comparison analysis, genotypes superior to the control were identified in several variables measured in the greenhouse and in the field. Plant heights in genotypes H1i, G3i, H2i, P2i, and P1i were the highest, with values ranging from 2.2 to 1.9 m at 150 days, indicating that some native materials had similar growth vigor compared with the control, an important feature in a canopy production system. Similar results were reported by ^{Núñez et al. (2012}), who evaluated the tomato cultivar Beatrice and reported that plants reached an average height of 2.3 m at 150 days. In contrast, ^{Carrillo & Chávez (2010)}, who evaluated native tomato populations in the greenhouse, reported a plant height of 1.3 to 1.8 m. Besides, in the present study it was observed that most of the materials evaluated in the open field showed low height values. According to ^{Juárez-López et al. (2012b}), these differences between the behavior of materials in greenhouse and open field could be an opportunity for growers who prefer medium-sized genotypes, which allows them to reduce labor in cultural tasks of the crop.

The highest number of leaves at 150 days of development was recorded in most of the genotypes grown in the open field, with V1c having the highest value with 329 leaves. According to ^{Keller et al. (2021}), mature leaves are important for an open field production system as they produce up to 80% of the photosynthetic carbon. On the other hand, in the greenhouse, the number of leaves was less than 38 after 150 days, and genotype P2i had the highest number of leaves. These values were lower than those reported by ^{Núñez et al. (2012}), who counted 53 mature leaves under a canopy system.

Fruit weight is closely related to several parameters, such as the number of ovary carpels, the position of the fruit in the cluster, and the prevailing environmental conditions during the fruit growth phase (^{Delgado-Vargas et al., 2018}); in addition to the supply of photoassimilates and water, which determine its growth and development (^{Quinet et al., 2019}). The fruits with the highest weight were obtained under protected conditions, being genotypes G3i and G1i, which showed the highest values, 179 and 176 g, respectively. These genotypes are within the ranges of high-quality caliber, as indicated by ^{PHG (2012)} in its classification of tomato fruits, which considers them as large-sized when they weigh between 121 and 180 g.

According to ^{Diouf et al. (2018}), the ability of plant material to have a high yield is given by the ability to accumulate biomass in the organs to be harvested. In this study, the best yield was achieved by the native genotype G3i with a yield of 4.4 kg/plant, followed by genotypes P5i and P3i. However, under field conditions, the yield of G3i was 2.8 kg/plant. This result suggests that this genotype had a better capacity for biomass accumulation under greenhouse conditions; while, exhibiting poor yield under field conditions. On the other hand, in previous work by ^{Maldonado-Peralta et al. (2016}) with the same genotypes, the G3 population reported a lower yield (0.39 kg/plant) than the obtained yield in this research. In general, the differences in the agronomic performance of the populations studied in the two production systems are due to the fact that open-field populations grow under uncontrolled conditions, whereas growing in favorable environments allows their traits to be better expressed (^{Kissoudis et al., 2015}).

Tomato populations native to Mexico show a wide morphological diversification, fruit quality, and a high degree of adaptation (Table 4), which allows for finding well-defined morphological features in local and regional cultivars. In this study, the native populations fruits were wider than long (globosa), compared to the controls (H1 and H2), which presented elongated fruits. The populations from Oaxaca (O4i) had a kidney shape and those from Guerrero (G3i) had a ball with ribs. According to ^{Van der Knaap et al. (2002}), samples of the genus Lycopersicum vary in color intensity, fruit shape, and size, as well as growth habit and leaves morphology. In the production systems, the native genotypes had fruits with the highest number of locules, the O3i population presented 11.5 and kidney shape. ^{Vázquez-Ortiz et al. (2010)}, found that fruits with this shape have an average of 11 locules, a variation that induces the particular shape they take. However, shape transformation has pleiotropic effects on fruit architecture, such as those related to increasing the number of carpels and locules (^{Lippman & Tanskley, 2001}).

^{Diez (2001}) mentions that tomato fruit for industry should have an average total soluble solids (TSS) between 4.5 and 5.5 %. ^{Foolad (2007}) mentions that tomatoes for fresh consumption require a value of 4.6 %. The TSS of the five best genotypes presented values higher than 5.3 % of (TSS) in protected conditions (Table 4), acceptable values for industry and fresh consumption. According to ^{Ruiz et al. (2005}), TSS is composed of 65% reducing sugars, such as fructose, glucose, and traces of sucrose, while the rest is represented by citric and malic acids, minerals, vitamin C, and other compounds in low concentrations. Besides a direct relationship between TSS levels and flavor was reported (^{Quinet et al., 2019}); the higher the TSS, the better the flavor intensity (^{Cheng et al., 2020}). Therefore, given the TSS levels in the present study, the evaluated materials are a potential source of germplasm to increase flavor in breeding programs.

^{Quinet et al. (2019}) mention that a higher flesh thickness (mesocarp) increases the weight and the edible part, thus improving the fruit quality, which was consistent with the P3i, G3, and P5 populations, where materials with high flesh thickness showed high fruit weight. This situation could be advantageous, as greater pericarp thickness allows the fruit to have a longer post-harvest life, an essential factor for marketing (^{Gan et al., 2022}).

Overall, the traits measured were higher in the protected conditions, showing that these protected conditions, compared to those obtained in the field, improved fruit quality in terms of fruit size, pulp thickness, and total soluble solids. This agronomic native populations performance within the under-canopy system could be explained by the fact that the materials have been selected by farmers for years in an open field production system, which exerts a high natural selection pressure. Thus, subjective selection by farmers over generations has been responsible for indirectly selecting for traits associated with better fruit size and quality (^{Ríos-Osorio et al., 2014}); at the same time, these traits make consumers identify native populations as attractive and are willing to pay a higher price for them (^{Casals et al., 2012}). This was clearly observed in the variable of yield, where tomato populations increased their yield by 130 % more compared to open field conditions, in agreement with ^{Kanwar (2011}), who reported a yield increase of 136.12 % more compared to open field cultivation. Such a result is due to the protected conditions that create a more favorable environment for production (^{Tao et al., 2016}).

Conclusions

Mexican native tomato populations responded differently to the two production systems and there were genotypes that presented better quality and fruit yield under a specific system. Under protected environmental conditions, the Guerrero 3, Puebla 5 and 3 populations that presented a higher fruit yield than the control. In the greenhouse, its weight and average fruit yield increased by 200 and 130 %, respectively. For the pericarp thickness and TSS variables of the fruit, the protected environment also promoted higher quality values than those in the open field. These results show that some native materials could be exploited in a better way under greenhouses and others in the open field.