INTRODUCTION

Global yam production is estimated at more than 253.5 million tons, with more than 7.4 million hectares established annually, and yams are ranked fourth in importance worldwide among plant roots and tubers (^{Borges et al., 2011}). Colombia is the third-largest producer in the Americas and one of the main yam exporting countries to the United States, with a yield of 10.05 t ha^-1 planted and a total of 108,067 ha planted in 2014 (^{Pérez and Campo, 2016}). This yield is due to factors within the production process, such as the type of seed, soil and planting techniques (^{Acevedo et al., 2015}).

Greater production is seen in the departments of Bolivar, Cordoba and Sucre (^{Álvarez, 2000}), where at least 20,000 families are dedicated this crop (^{Doncel and Pérez-Cordero, 2017}) and where the diet of the inhabitants relies on yams because the main component is starch, which contains high amounts of carbohydrates (^{Pinzon et al., 2013}).

Additionally, the pharmaceutical industry uses this tuber (Dioscorea) for the treatment of diabetes, dysmenorrhea, and inflammation, etc., (^{Ramos et al., 2015}).

The following species are cultivated: creole yam (D. alata L.), hawthorn yam (Dioscorea rotundata Poir), papo yam (Dioscorea bulbífera), sugar yam (Dioscorea esculenta) and ñampin (Dioscorea trifida), among more than 100 species of the genus Dioscorea; D. alata L. and D. rotundata Poir., stored by farmers for four months on average before commercialization, which directly affects quality and causes a significant amount of tubers to deteriorate (^{Dramani, 2013}) because of a lack of adequate storage, packaging, and logistics, with vulnerability to physiological and pathological factors that affect the income of producers (^{Tanye, 2016}).

During storage, transpiration and respiration cause desiccation and weight loss, aggravated by the sprouting process, which is characterized by a decrease in the starch content and an increase in reducing sugars, which make consumption and marketing unacceptable, generating losses for producers (^{Afoakwa and Sefa-Dedeh, 2001}).

Losses under normal storage conditions can reach up to 10% after three months and 25% after five months (^{Osunde and Orhevba, 2009}). To counteract this, tuber dormancy must be maintained, for which growth-inhibiting substances are needed (^{Hamadina, 2011}). Abscisic acid (ABA) seems to control dormancy through the tuberization process, and it has been suggested that both ABA and paclobutrazol (PBZ) protect against water loss (^{Rodríguez et al., 2000}).

Technological alternatives that help reduce weight loss in yam tubers and reduce sprouting must be tested. Therefore, the purpose of this study was to evaluate ABA and PBZ as treatments to reduce postharvest losses during tuber storage and estimate which concentrations of the exogenous applications saw better responses for prolonged storage of yams under optimal conditions, favoring commercialization as a product with excellent quality in the market.

MATERIALS AND METHODS

RESULTS AND DISCUSSION

The analysis of variance showed highly significant differences between the species (P<0.01) for tuber weight loss (TWL) during 90 and 120 d of storage after applying ABA, while, at 30 and 60 d, there were no differences. Likewise, the ABA concentrations and the Species×ABA interaction were significant at 30, 60, and 90 d. The additional treatments with PBZ, when contrasted with the factorial part, were only significant at 90 d. The contrast between the two species treated with PBZ showed significant differences at 30, 60, 90, and 120 d (Tab. 1).

In D. alata, the lowest TWL was recorded with ABA concentrations of 10 and 15 mg L^-1 at 30 d. At 60 d, there were no differences between the ABA levels, and the PBZ did not have an influence. At 90 d, the lowest TWL was registered with the ABA at 10, 15, and 20 mg L^-1, overcoming the effect of PBZ and the non-application of ABA. At 120 d, there were no differences in the TWL with any of the concentrations for the two growth regulators. In D. rotundata, the lowest TWL were recorded with 10 mg L^-1 of ABA at 30, 60, and 90 d, and at 30 and 90 d with the PBZ. As in D. alata, at 120 d, there were no differences in the TWL with any of the concentrations for both growth regulators (Tab. 2).

The applications of ABA and PBZ decreased the TWL in both species in the periods over three months. In both species, a gradual decrease in the TWL was observed between 30 and 120 d of storage, 24.4% on average. These results corroborate the report by ^{Dramani (2013)}, who indicated that yams can lose about 20% of their weight after three months and reach 50 to 80% weight loss after six months of storage, depending on the species. These losses are due to the decrease in substrates as a result of respiration and loss of water from transpiration, reducing the quality of products in terms of appearance, freshness, nutritional content, and loss of antioxidants (^{Rodríguez et al., 2006}).

The reduction in tuber size and weight during storage could be explained by the use of carbohydrate reserves for respiration, accompanied by transpiration, since some products present greater cell disintegration and weak cell walls and membranes during storage, allowing water to escape through transpiration at a high rate (^{Kays, 2004}).

ABA and PBZ have been suggested for protection against water loss; yam plants treated with PBZ showed greater resistance to desiccation than untreated plants (^{Rodríguez, 2000}). A similar result was reported by ^{Pantoja (2015)} in potatoes (Solanum phureja). In D alata, the lowest tuber sprouting (TS) was obtained with 20 mg L^-1 of ABA at 30 and 90 d, and with 15 mg L^-1 of PBZ at 60 d, with percentage differences concerning the non-application of ABA, 33.2, 28.1, and 21.1% (Tab. 3). This indicates that both growth regulators exerted effective control over ST during the first 90 d.

At 120 d, with all ABA and PBZ concentrations, 100% TS was reached in both species, indicating a decreasing loss of sprouting control over time. In D. rotundata, the lowest sprouting was estimated with 15 mg L^-1 of ABA at 30 and 60 d, and with PBZ at 90 d of storage, with percentage differences of 28.1, 28.2, and 24.5%, respectively, as compared with no ABA application.

These results could be explained in terms of the relationship between the growth-inhibiting substances and the dormancy state of the tubers. Hormones such as abscisic acid (ABA), gibberellins (GAs), ethylene, and cytokinins (CKs) are present in the dormancy period. ^{Suttle (2004)} pointed out that both ABA and ethylene are necessary for the initiation of tuber dormancy but only ABA maintains the latent state because it is involved in the inhibition of DNA and RNA synthesis, while GAs are involved in the acceleration of DNA and RNA synthesis, that is, while GAs promotes shoot growth, ABA maintains latency (^{Ortiz and Flórez, 2008}).

In this experiment, the ABA and PBZ, at the concentrations used, slowed down the ST process for up to a storage period of no more than 90 d. However, the decreased ABA concentration at which sprouting starts must be known (^{Claassens and Vreugdenhil, 2000}). When there are changes in the concentrations of GA and ABA, ABA biosynthesis appears to be involved in the induction and maintenance of tuber dormancy (^{Suttle, 2004}).

However, authors such as ^{Biemelt et al. (2000)} disagreed, not having found a correlation between the ABA levels and sprouting during storage. This indicated that shoot growth requires the supply of carbohydrates obtained from soluble sugars or starch. Also, these authors suggested that studies should focus more on the mobilization and transport of sucrose than on phytoregulators to understand the sprouting process in tubers.

^{Villar (2015)} stated that the delay in sprouting that is achieved with regulators such as ABA can be related only to the time necessary for the tuber to metabolize the extra dose of circulating ABA, and that subsequent attempts at maintaining high concentrations of this hormone for dormancy are not effective.

For the effect of PBZ, a decrease of up to 50% in length and number of shots of potato tubers stored and treated with PBZ was reported (^{Cheema, 2010}). Likewise, it has been found that periodic applications of PBZ at low concentrations favor the accumulation of starch, with a tendency to increase the number of tubers (^{Flores et al. 2016}).

^{Pantoja (2015)}, on the other hand, warned about the efficient behavior of PBZ in the regulation of sprouting Solanum phureja tubers, reporting sprouting percentages lower than 3.33% for treatments with doses of 30 mL kg^-1 under unspecialized storage conditions, up to 9 d, with 98% of the tubers without the presence of shoots, which may be due to the maintenance of low levels of GAs that are involved in the growth of shoots because PBZ inhibits GA biosynthesis (^{Hartmann et al., 2011}). Similar results have been found with PBZ foliar sprays on potatoes, delaying the appearance of new plants and the beginning of tuberization (^{Kianmehr et al., 2012}; de ^{Araújo et al., 2020}).

Contrary effects have been reported in the date palm (Phoenix dactylifera L.), with an increase and multiplication of shoots (^{Awadh et al., 2019}), which could indicate that the action of PBZ will not only depend on the concentration used but on the studied species, the form of application, the absorption and the phenological stage in which the application was made (^{Mabvongwe et al., 2016}).

The starch percentages on a dry basis (SD) in D. alata were higher with 10 and 20 mg L^-1 of ABA at 30, 60, and 90 d of storage, while, in D. rotundata, there were no differences between the different concentrations of growth regulators at 30 and 60 d; however, at 90 d, the highest SDs were obtained with 15 and 20 mg L^-1 of ABA and 15 mg L^-1 of PBZ. Likewise, D. rotundata always presented a higher starch content during the entire storage period (Tab. 4).

These results indicate that the ABA and PBZ favored the preservation of a high SD content in both species, thereby increasing its importance as a food ingredient of great energy value and as a possible alternative source of commercial starch (^{Techeira, 2008}). The SD percentage in both species fluctuated between 73.7 and 80.5% for D. alata and 80.8-83.2% for D. rotundata in the first 90 d, similar to the report by ^{Techeira (2008)} for D. alata.

Plant treatments with PBZ have caused an increase in the number of enzymes such as starch synthase responsible for the biosynthesis of starch, increasing its content in tubers (^{Mabvongwe et al., 2016}). However, other authors have pointed out that the percentages of starch in tubers may decrease when treated with growth regulators such as PBZ, which could be due to a momentary dilution of the starch content as the result of an increase in the water content of tubers, depending on the conditions in which they are maintained (^{Araujo et al., 2020}) or the presence of a high GA content that prevents the accumulation of starch in tubers (^{Mabvongwe et al., 2016}).

The contents of reducing sugars (RS) in D. alata were not significantly different between the ABA levels during the first 90 d of storage but, at 120 d, the PBZ was higher(1.47%) with ABA at 20 mg L^-1 (Tab. 5). For D. rotundata, the treatment with ABA at 10 mg L^-1 had a significant influence only at 90 and 120 d, expressed in a lower content of RS (1.23 and 1.05%, respectively), than in the other levels of ABA and PBZ.

These results indicate that the RS content was not related to the sprouting of tubers since the non-application of growth regulators was expected to influence a higher content of reducing sugars as the result of the interconversion of carbohydrates for the production of oxidizable substrates needed in cell division for the generation and growth of new shoots. For potatoes, ^{Pantoja (2015)} reported that growth inhibitors did not have a significant effect on the content of reducing sugars, always maintaining low concentrations both in tubers treated with inhibitors and in untreated ones.

For Lilium rubellum, ^{Xu et al. (2006)} reported an increase in the sugar content in tubers with the highest sprouting, related to the possible combined effect of a decrease in free ABA, the accumulation of sugars, and the decrease or increase of some factors unknown in bulbs. Likewise, ^{El-Sherbiny et al. (2017)} reported high percentages of total sugars in Solanum tuberosum L., similar to those detected by ^{Mahto and Das (2014)} in potato tubers.

^{Araujo (2020)} reported a decrease in RS in potato tubers treated with PBZ. The accumulation of soluble sugars could be related to the release of endodormancia from the sprout, possibly by reducing the osmotic potential that triggers the events that lead to dormancy of the buds or sprouts in potatoes (^{Pantoja, 2015}) and in dormant onion bulbs (^{Benkeblia et al., 2008}). An increase in sugars is not only part of the nutrition that maintains growth but also signals control of sprout development (^{Chao and Serpe, 2010}).

CONCLUSIONS

Abscisic acid (ABA) and paclobutrazol (PBZ) exercised effective control over sprouting and tuber weight loss during the first 90 d of storage in the yam species D. alata and D. rotundata, and their use is suggested as a management alternative to reduce losses in quantity and quality during post-harvest conservation.

The best results for sprouting reduction in both species were obtained with ABA concentrations of 20 and 15 mg L^-1 of PBZ during the first 90 d, with percentages up to 33.2% in D. alata and 28.2% in D. rotundata. The lowest weight loss in both species was recorded with ABA concentrations of 10 and 15 mg L^-1 during the first 90 d.

The ABA, at concentrations of 10, 15, and 20 mg L^-1, and PBZ, at 15 mg L^-1, favored the conservation of high starch contents in both species.

The applied concentrations of ABA and PBZ did not influence the content of reducing sugars. Only up to 90 and 120 d of storage, a decrease in reducing sugar content was observed in D. rotundata with concentrations of 10 mg L^-1 of ABA.