Heterosis in okra hybrids obtained by hybridization of two methods : traditional and experimental

High heterosis values can be obtained using hybrid okra seeds. However, despite high heterotic values, it is very difficult to produce hybrid okra seeds. The objective of this study was to evaluate an experimental methodology of hybridization. Two different methods of hybridization were used: traditional manual hybridization with complete emasculation followed by hand pollination (artificial) and, an experimental methodology, performing manual hybridization with incomplete emasculation followed by entomophilous natural pollination between two inbred lines (UFU-QB-040D and UFUQB-107G).These inbred lines were obtained from three selection cycles. Three hybrid types were obtained from the two hybridization methods: F1 THM(UFU-QB-040D x UFU-QB-107G), F1 IHM(UFUQB-107G x UFU-QB-040D) and F1 IHM(UFU-QB-040D x UFU-QB-107G). The resulting hybrid seeds and their parents were evaluated via an experiment set up in a randomized block design with four repetitions. We evaluated fruit production per plant, fruits per plant, number of branchess per plant, superiority of hybrids relative to parents (h) and heterosis (%). At the end of the cycle, morphological characterizations was done of the leaves (pubescence, length and width) and fruit (length, diameter and shape). Heterosis resulting from the incomplete emasculation followed by natural insect pollination was similar to that from the traditional hybrid method, being a viable alternative for the production of hybrid okra seeds without losses in heterosis.


Scientific communication
Hortic.bras., v35, n.1, Jan. -Mar.2017 O kra (Abelmoschus esculentus)   belongs to the Malvaceae family and originates from Africa.The species performs well in tropical regions and is especially well adapted to the tropical regions of Brazil.Santa Cruz 47 is the cultivar most commonly used by Brazilian growers.Currently, varieties predominate over hybrid seeds in fruit production.Nevertheless, the estimated use of hybrid okra seeds rose 29% from 2010 to 2011 (ABCSEM, 2012).
Improving productivity is one of the main goals of crop breeding programs.Hybrid seeds are extensively used because they provide one of the fastest ways to increase productivity (Paterniani, 1974).Okra heterosis has already been exploited for decades (Sing & Singh, 1978).Current research has also shown the heterotic potential of hybrid okra seeds in targeting greater productivity (Mehta et al., 2007;Weerasekara et al., 2007;Jaiprakashnarayan et al., 2008;Jindal et al., 2009).Hybrid seeds are more expensive than varieties; however, hybrid plants are more productive and produce greater income (Medagam et al., 2012).
Despite the superior genetic potential

Received on August 29, 2015; accepted on April 22, 2016
of hybrid okra seeds, seed production remains an obstacle.Okra is considered 100% self-pollinated (Malerbo-Souza et al., 2001) with high outcrossing rates that vary from 18.75% (Purewal & Rhandhawa, 1947) to as much as 42.2% (Mitidieri & Vencovsky, 1974), provided by frequent visits from insects, especially Apis melífera (Mitidieri & Vencovsky, 1974;Malerbo-Souza et al., 2001).This dynamic reproductive behavior means that female flowers need to be protected immediately after emasculation to avoid contamination.Some studies show that gametocides can induce male sterility (Deepak et al., 2007), which would reduce the cost of producing hybrid okra seeds and eliminate the need to emasculate and protect flowers.Nevertheless, gametocides use is still undefined, uncommon, and may be toxic to plants, which results in reduced seed yield.In fact, there is little research on economical ways of producing hybrid okra seeds.A traditional method is currently used in Brazil for large scale production of hybrid okra seeds.The method consists of emasculation of the female parent followed by manual crosses, leading to high production costs (Nascimento, 2014).Higher costs have limited greater exploitation of the heterotic potential of okra hybrids and reduced seed availability for growers.
The objective of this study was to evaluate an experimental methodology of hybridization for production of hybrid seeds of okra.

MATERIAL AND METHODS
The experiment was carried out at a university experimental station in Monte Carmelo, Brazil (EE da UFU, Monte Carmelo-MG) (18º42'43"S, 47º29'56"W, elevation of 873 m).The regional climate is humid temperate with hot summers and dry winters.
The genetic material consisted of inbred lines (UFU-QB-040D and UFU-QB-107G) obtained after three selection cycles from the okra germplasm bank at the Monte Carmelo campus of the Federal University of Uberlandia and maintained at LAGEN-UFU (laboratory for the analysis of seeds and plant genetic resources).
Two inbred lines (UFU-QB-040D and UFU-QB-107G) were sown on May 13, 2013 and used for a fourth selfing to produce hybrid seeds.Polystyrene trays (200 cells) were filled with a commercial coconut fiber substrate and used to produce 200 seedlings for each line.The seedlings were planted in the field 35 days after sowing.
The plants were distributed 1:1 in twenty rows with 20 plants per row.The plants were spaced 1.2x0.8m, covering a total area of 384 m 2 .Both inbred lines started flowering approximately 65 days after transplanting.Hybridization commenced at the peak of flowering for both lines, 92 days after transplanting.This strategy guarantees okra seeds of higher physiological quality than seeds produced from the initial phase (Purquerio et al., 2010).Apis melifera was the most common insect present during flowering.During this step, two different hybridization methods were used (traditional and new) on the two inbred lines.Traditional hybridization method (THM) consisted of emasculating the female parent flowers to expose the flower stigma before anthesis and eliminate the presence of pollen.Next, pollen from the male parent was moved by hand to the stigma of the previously emasculated female flower.After pollination, the female flowers were protected with Kraft paper to avoid contamination from insect pollination (Nascimento, 2014).This process resulted in the THM F1 hybrid (UFU-QB-040D x UFU-QB-107G).The new hybridization method (IHM) consisted of emasculating the female flowers and then allowing pollination to occur via insects that were naturally present at the experiment site.The number of steps was reduced in this process by eliminating the need to protect the emasculated female flowers or collect male flowers.This new method was adopted because of the results obtained by Malerbo-Souza et al. (2001), which showed that cross pollination of unprotected okra flowers without any kind of emasculation produced fruits that were 65% heavier, 26% wider and 28.3% longer.Two reciprocal hybrids [F1 IHM (UFU-QB-107G x UFU-QB-040D) and F1 IHM (UFU-QB-040D x UFU-QB-107G)] were obtained using the new methodology.
Fifteen flowers were emasculated from the female parent for both methods (THM and IHM).Originated fruits of hybridization were harvested when they reached physiological maturity and then the seeds were extracted separately according to hybridization method (THM and IHM).After extraction, the seeds were placed in pre-labeled Kraft paper bags and stored.Just after the first experiment, another experiment was conducted using the hybrid seeds originated from the first experiment (F1 THM(UFU-QB-040D x UFU-QB-107G), F1 IHM(UFU-QB-107G x UFU-QB-040D), F1 IHM(UFU-QB-040D x UFU-QB-107G) with their parents (seeds from the inbred lines).This experiment was set up in randomized blocks design (RBD) with four replications and used the following statistical model: The five treatments were sown on November 21, 2013: F1 THM(UFU-QB-040D x UFU-QB-107G), F1 IHM(UFU-QB-107G x UFU-QB-040D), F1 IHM(UFU-QB-040D x UFU-QB-107G), including both inbred lines used in the hybridizations.
The fruits were harvested every three days from February 2, 2014 until June 27, 2014, totalling 41 harvests.The following agronomic characteristics were determined: production per plant (g/plant), number of fruits/plant and average number of branches/plant.
Production per plant (g/plant): obtained from the sum of all samples harvested from the plot divided by the number of plants in the plot.A precision scale (four decimal places) was used.
Number of fruits/plant: obtained from the sum of all harvested fruits from the plot divided by the number of plants in the plot.

Average number of branches/ plant:
The branches were counted after the last harvest of the experiment.
The hybrid combinations were evaluated, resulting from the hybrid seed production methods (THM and IHM), by estimating the superiority of the hybrids in relation to parents [h=F1-(P1+P2)/2] and heterosis (%).We used the expression proposed by Fehr (1987) to evaluate the heterosis.
Data for each of the agronomic characters were submitted to analysis of variance (F test).Means were compared by the Tukey test (p<0.05)using the GENES (Cruz, 1997) software.Orthogonal contrasts [(hybrids versus strains), (THM versus IHM) using the Scheffé test (p<0.01)]were used to strengthen the h and h% values and the heterotic potential of the new method of producing hybrid okra seeds (IHM) relative to the traditional hybrid method (THM).
Only UFU-QB-107G presented shorter fruits than average value (Table 1).Conversely, mean fruit diameter (21.5 mm) of the same strain (UFU-QB-107G) was significantly greater than that of the other genotypes.There were no significant differences in fruit shape between the UFU-QB-107G parent and the three experimental hybrids [F1 THM(UFU-QB-040D x UFU-QB-107G), F1 IHM(UFU-QB-107G x UFU-QB-040D) and F1 IHM(UFU-QB-040D x UFU-QB-107G)].These fruits tended to be cylindrical and significantly different from the angular UFU-QB-040D parent (Table 1).An understanding of the morphological characteristics of parents (male and female) and the consequent hybrids is useful in plant breeding.Hybridization can be more confidently confirmed by analyzing a combination of these morphological characteristics from parents and hybrids.Pedrosa et al. (1983) stated that the morphological characterization of okra genotypes is needed to take advantage of characteristics that are highly correlated with productivity.With this information, heterosis in okra can be exploited regardless of the method used to obtain hybrid sees.Significant differences in production per plant (PP) and number of fruits per plant were observed in the different hybrids regardless of hybridization method (THM and IHM) (Table 2).Production per plant (9985.0,8278.7 and 10,157.5 g/plant) did not differ significantly between the experimental hybrids F1 THM(UFU-QB-040D x UFU-QB-107G), F1 IHM(UFU-QB-107G x UFU-QB-040D) and F1 IHM(UFU-QB-040D x UFU-QB-107G).Production per plant (6657.3 and 3622.2 g) from the parents (UFU-QB-040D and UFU-QB-107G) was lower than that of the F1 THM(UFU-QB-040D x UFU-QB-107G), F1 IHM(UFU-QB-107G x UFU-QB-040D) and F1 IHM(UFU-QB-040D x UFU-QB-107G) hybrids.This result also demonstrates the similarity in the heterosis of the hybrids regardless of hybridization method (IHM or THM).Number of fruits per plant (NFP) was also similar between the newly proposed hybridization method for okra (IHM) and the traditional method (THM) (Table 2).Furthermore, prolificacy (number of fruits per plant) was not statistically different between F1 IHM (UFU-QB-107G x UFU-QB-040D) and F1 THM (UFU-QB-040D x UFU-QB-107G), which had been hybridized by IHM and THM respectively (Tukey, p<0.05).Singh & Singh (1978) observed genotype expression in number of fruits per plant and productivity per plant and found that there were genes with no additive effect and that most of these genes were dominant and epistatic.Some reports show that the number of fruits per plant is related to the action of other types of gene actions such as epistatic (additive x additive) that allow transgressive segregation (Kulkarni et al., 1978).In fact, the hybrids obtained in this study showed superiority to their parents regardless of the hybrid seed production method (Table 2).
The number of branches per plant was statistically different between the genotypes evaluated in our study.
Similarities in PP and NFP resulting from two methods of producing hybrid okra seeds (IHM and THM) were confirmed by h and h% estimates.Satisfactory increases were obtained regardless of seed hybridization method (IHM versus THM).
Other contrasts were created, C4 [F1 (IHM UFU-QB-107G x UFU-QB-040D) vs parents], C5 [F1 IHM (UFU-QB-040D x UFU-QB-107G) vs parents] and C6 [F1 THM (UFU-QB-040D x UFU-QB-107G) vs parents], to confirm the individual superiority of each hybrid over their respective parents.Not only did the C4, C5 and C6 contrasts show the superiority of these three hybrids, regardless of seed hybridization method, they also demonstrated the heterosis of the hybrids and the use of new method of producing hybrid okra seeds.These results reflect the h and h% results obtained between the hybrids (Table 1).
Heterosis resulting from the incomplete emasculation followed by natural insect pollination was similar to that from the traditional hybrid method, being a viable alternative for the production of hybrid okra seeds without losses in heterosis.