Introduction

Banana (Musa spp.) is the main crop in humid and warm regions. Its production is distributed in 20 countries, out of which Mexico holds the 10^th place (^{FAOSTAT, 2012}). The State of Tabasco is the second domestic producer of banana (^{SIAP, 2015}). Black Sigatoka disease -caused by Mycosphaerella fijiensis Morelet fungus- is a phytosanitary problem that affects the banana commercial production in most of the producing regions. Fruit losses can be total and represent up to 30 % of production costs (^{Marín et al., 2003}). A strategy to solve this problem is to select natural genotypes or those coming from breeding programs that are resistant or tolerant to this disease, and present good agronomic characteristics (^{Dadzie, 1998}). These programs have included the use of the Yangambi km 5 dessert banana cultivar (AAA). In the Democratic Republic of Congo, its country of origin, its name means “many small fruits” (^{Daniells and Bryde, 1995}; ^{Daniells et al., 2001}). Another cultivar is Pisang Awak (ABB) -which means bananas in Malay-; its peel is light green and waxy when it is unripe, and pale yellow and slightly waxy when it is ripe. These distinctive colors can often be enough to identify the Pisang Awak fruits (^{Wang and Kepler, 2009}) that are used in all East African countries to brew beer (^{FHIA, 2003}). The genetic improvement program of Fundación Hondureña de Investigación Agrícola (FHIA) has had great success in the creation and scattering of hybrids of all types, resistant or tolerant to Fusarium oxysporum f. sp. cubense (Panama disease) and Mycosphaerella fijiensis Morelet (Black Sigatoka) fungi, and to Pratylenchus spp. and Radopholus similis nematodes. Some of those hybrids are: dessert bananas (FHIA-17, FHIA-23), French type plantains (FHIA-20, FHIA-21), sweet and sour bananas (FHIA-01, FHIA-18, FHIA-26), and cooking bananas (FHIA-03, FHIA-25). The FHIA-18 is consumed both fresh and processed. A special characteristic of this hybrid is that the unripe and mature pulp does not turn brown, making it ideal for processing. When cooked green, it has an attractive creamy color, with excellent texture and flavor. Fruit processed while it is still unripe (tajaditas) has a creamy color, is crunchy, and has a concentrated bittersweet flavor (^{FHIA, 2014}).

Banana has a special place in the daily diet of millions of people, and is a basic food consumed fresh or cooked. This fruit can be transformed into products that can be packed and stored for various periods, such as banana purée, dehydrated bananas, flour, starch, banana chips, juice, flakes, cereal, vinegar, alcoholic beverages, canned slices, essence, bread, cookies, cakes, jam, jelly, and crystallized cubes (^{Mohapatra et al., 2011}).

The production of green banana flour is an excellent alternative to minimize postharvest losses and to retain the nutritional value of fresh bananas. This flour is rich in resistant starch and dietary fiber that helps to keep the colon healthy (^{Pragati et al., 2014}). Flour can be made from banana pulp or skin as a result of the drying procedure. The flour keeps its original vitamin, protein, and mineral salt properties intact, and its weight decreases by one third (which facilitates its transportation and storage). This product does not depend on crop seasonality, and losses caused by excess ripening are avoided. In addition, its powdery form facilitates its industrial handling (^{Robles, 2007}). Banana flour is incorporated into cookies (^{Aparicio-Sanguilán et al., 2007}), high-fiber bread (^{Juárez-García et al., 2006}), and edible films (^{Rungsinee and Natcharee, 2007}), making them slowly digestible. According to the Colombian Technical Standard 2799 -published by the Instituto Colombiano de Normas Técnicas y Certificación (^{ICONTEC, 2002}) for the banana flour industries-, the use of sweeteners, flavoring, or coloring, and any other additive is not allowed in the flour. Flour must be free of dust particles or any other strange pollutant material, and it must be free of toxic materials (such as pesticides and herbicides).

Resistant banana cultivars face a lack of acceptance by consumers as fresh fruit, because the peel of some of them presents unpleasant spots, and the taste of others is different to commercial varieties. Therefore, it is necessary to find processing alternatives that encourage its consumption. In Mexico, the sanitary or nutritional requirements of banana flour are not regulated. Consequently, the objective of this study was to evaluate the quality of banana flour of three cultivars in physiological maturity that are resistant to black Sigatoka disease (Yangambi km 5, FHIA-18, Pisang Awak).

Materials and Methods

In this study, banana bunches in physiological maturity from FHIA-18 (AAAB), Yangambi km 5 (AAA), and Pisang Awak (ABB) cultivars were used. The plot where they were obtained was located in the community of Cumuapa (Cunduacán, Tabasco, Mexico), 18° 01’ 03” N, 93° 07’ 39” W, at 10 m altitude. The research was conducted in two harvest cycles (September 2014 and October 2015). The plants were set up in a fully randomized block design with four replicates, and all the cultivars were randomly assigned in the plots within the blocks. Each plot had six plants with 2.5 m x 3 m spacing between them, and a local cultivar was used between rows. Fertilization was carried out twice a year using Triple 17. The bagging was carried out two weeks after the appearance of inflorescence. No shoring of the plants whatsoever was required.

The evaluation of the postharvest characteristics of the fruit in physiological maturity was carried out using five bunches per cultivar. In order to determine the bunch weight, the hands were separated from the stalks, and each stalk and each hand were weighed in a Torrey digital control scale (Model L-EQ 5/10, 10 kg capacity, Mexico). The hands were placed in tanks with chlorine water (125 mg L^-1 sodium hypochlorite) to favor the coagulation of the latex that flows from the breaking point, as well as its elimination (^{Arias and Toledo, 2000}). In order to evaluate the fruit characteristics, the middle finger of the second hand was selected to reduce the variation. Pulp and peel were separated and weighed in order to determine their ratio. The finger was subject to a cross section at the central point, the fruit was peeled, and the peel and pulp thickness were measured using a digital caliper (Caldi-6MP, China) (^{Dadzie and Orchard, 1997}).

Moisture, dry matter (DM), total soluble solids, pH, and titratable acidity were determined in order to characterize the quality of the fruit in physiological maturity. The samples were placed in an oven (Yamato Model DX400, USA), with temperature control, and air circulation, at 100 °C for 24 h in order to determined its moisture, the samples were cooled to 26 °C in a desiccator, and weighed on an analytical balance (Adam Equipment, USA) (^{Dadzie and Orchard, 1997}).

The total soluble solids were measured based on the refractive index of the pulp juice using a digital refractometer (Hanna Instruments, USA) following the instructions described in the method 940.31 of ^{AOAC (2000)}. For this evaluation, 30 g of pulp tissue -taken from a cross section of the middle part of the fruit- were homogenized for 2 min, in 90 mL of distilled water (ratio 1:3, p:v), using a hand blender (Cuisinart Model HB-154PC, France); subsequently, the juice was filtered through a sieve (^{Dadzie and Orchard, 1997}).

The titratable acidity was measured according to ^{AOAC (2000)}, 30 g of pulp tissue were weighed, 90 mL of distilled water were added, the mixture was homogenized for 2 min, and the juice was filtered. Twenty-five mL of the filtrate were transferred to a 125-mL Erlenmeyer flask; then 25 mL of distilled water and 5 drops of 1 % phenolphthalein indicator in alcohol solution were added. The acidity of the filtrate was titrated with 0.1 N NaOH. The results were expressed as malic acid meq per 100 g of sample.

The pH of the pulp juice was measured with a pH-meter (Hanna Instruments, USA). In order to achieve this, 30 g of banana pulp were weighed, 90 mL of distilled water were added, the mixture was homogenized for 2 min, and it was filtered to obtain the juice (^{AOAC, 2000}).

To make the flour, a hand in the first stage of maturity was weighed and separated from each banana cultivar, the fingers were peeled, and the peel was eliminated. The pulp was cross cut into 3-4 mm thick slices (average) and placed in the trays of the tray dryer (Polinox, Mexico, capacity 8 kg) with integrated automatic temperature control. Internal temperature was controlled at 60 °C for 18 h and the dried slices were weighed to calculate the yield. A Siemens pulverizing mill (with a 5 HP motor, Germany) was used to reduce the particle size. A No. 70 sieve (212 µm) was used to homogenize particle size (method 925.22, ^{AOAC, 2000}). Then, the flour was packed in heat sealed cellophane bags to keep it from absorbing moisture from the environment while it was kept in storage awaiting analysis.

The sample was prepared for its microbiological analysis in accordance with ^{NOM-110-SSA1-1994}. In order to quantify aerobic mesophilic bacteria, the samples were incubated for 24 h at 31±1 °C, using Tryptone Glucose Yeast Extract Agar (Bioxon Standard Method Agar). Total coliform bacteria were quantified in plates with 30-300 colonies (^{NOM-092-SSA1-1994}), using Violet Red Bile Dextrose Agar, (VRBD, Bioxon), incubating the sample for 24 h at 31±1° C in a bacteriological oven (Thelco, USA) (^{NOM-113-SSA1-1994}). The fungi and yeasts were quantified on Potato Dextrose Agar (PDA, Bioxon), incubating the sample for 48 h at 25 °C (^{NOM-111-SSA1-1994}).

In the proximate chemical analysis of the flour, the percentage of total nitrogen was determined by the Kjeldahl method using a factor 6.5 to obtain the percentage of crude protein (^{AOAC, 2000}). Ash determination was carried out by the incineration method (^{AOAC, 2000}) using a Felisa muffle furnace (Model Fe-340, Mexico). The ethereal extract was determined using the Soxhlet technique according to the official method of ^{AOAC (2000)}. Crude fiber was quantified using the procedure of the ^{NMX-F-090-S-1978} Standard.

The experimental design was fully randomized with three treatments: one per each of the three banana cultivars from which the flour was made. All the evaluations were done in triplicate. The data were analyzed using an ANOVA with the GLM procedure and the means were compared with the Tukey test (p=0.5) using SAS 9.2 for Windows (^{SAS Institute, 2009}). In addition, the Pearson test (p≤0.05) was used to correlate the physical and chemical variables of the three treatments.

Results and Discussion

Conclusion

The characteristics of banana bunches in physiological maturity resulted in a flour with good quality parameters, fulfilling most of the chemical and microbiological requirements established in the Colombian Technical Standard 2799. These values are a little higher than those allowed for crude fiber and total aerobic mesophilic bacteria. This opens a potential for the industrial use of these three cultivars whose consumption as fresh fruit is not fully accepted. Based on the results obtained, the FHIA-18 cultivar has the best characteristics for flour production.