Fermentative and nutritive quality of banana by-product silage for goats

Abstract

This study was carried out to ascertain the ensiling capacity of fresh banana packaging industries by-products and the chemical composition and fermentative characteristics of the silage. Ensilability parameters were analysed in five different combinations [100% green banana (GB), 100% bunch (b), 30% GB–70% b, 50% GB–50% b and 100% ripe banana (RB)]. Subsequently, according to ensilability results, three combinations were selected [100% GB, 50% GB–50% b, 50% GB–50% b + additives, 10% molasses and 7% beet pulp] in order to assess the quality and stability over time of the experimental banana microsilages. The 50% GB–50% b combination showed the acceptable silage characteristics but requires certain additives to improve the fermentative process. The combination with better results, when assessed for quality and stability during the preservation process, was 50% GB–50% b + additives allowing to increase the initial dry matter (DM) and soluble sugars. This combination showed a good aptitude for ensilage, having a pleasant smell and good visual characteristics. The banana by-product showed moderate silage DM (20–30%) and organic matter (85–94%) content and low protein and fibre values. The chemical composition did not vary significantly along the conservation process. The three combinations of banana by-product silage studied presented a moderate DM content and a low protein value, with lower values in the 50% GB–50% b.

Keywords:

• by-product
• banana waste
• silage
• fermentation parameter
• ruminant
• goat

1. Introduction

The climatic conditions of arid areas generate low agricultural production and a low availability of forage for ruminants. For this reason, the Canary Islands (Spain) present a marked dependence on foreign countries related to livestock feed, one of the serious bottlenecks in the livestock sector. In recent times, this problem has worsened greatly due to increasing prices of concentrate and fibrous materials. Farmers have been significantly damaged, and their profit margins have been reduced. A good economic alternative for livestock feeding is to produce good quality silage using intensive crops by-products or using by-products of the food industry.

The Canary Islands are the most important producers of bananas in Europe (over 345,000 tonnes of bananas per year). Banana by-products consist on the banana crop residues (stem and leaves) and the classification and packaging process wastes (bunch and bananas) and are often used fresh. Ten per cent of the weight of banana fruits (40–50 kg) can be considered waste product in the packaging, conditioning and classification process before banana commercialization. Of this 4–5 kg, about 3 kg would correspond to the bunch and 2 kg to the banana fruit waste. Therefore, the production of banana packaging by-products is estimated around 35 million kg of fresh matter (Álvarez et al. 2011b).

Despite the high amount of this by-product, its utilization presents two main limiting factors. The first one is that individual fractions constitute unbalanced forages being highly fibrous in the case of stem, leaves and bunch or with high soluble carbohydrate content in the case of wasted bananas. The second limiting factor refers to the seasonality of supply (availability restricted to the banana harvesting period). Moreover, the high moisture content of this by-product means that it is spoilt, and they are often wasted.

Ensilage allows harvesting the forage at an optimal vegetative stage and could be the most appropriate way for preserving high moisture by-products for long periods. Ensiling by-products is a simple and a low-cost option, which can preserve feeds that are seasonally abundant for later feeding during periods of feed shortage. Incorporation of additives such as molasses to silage is optional; nevertheless, this is an excellent option to ensure a good conservation and enhance high quality silage of any ensiled feed resource (Kayouli & Lee 2000).

Large amounts of agro-industrial by-products and crop residues produced in the Mediterranean basin (citrus pulp, grape marc, tomato pulp, olive cake, wheat bran, date palm, etc.) have been successfully ensiled in different ways as sole ingredients or in different associations. Such silage-making practices are extensively practised by numerous farmers and replace the conventional feedstuffs, including imported concentrates (Hadjipanayiotou 1993; Kayouli et al. 1993; Vasta et al. 2008).

During ensilage process, lactic acid bacteria ferment carbohydrates of the fresh forage or by-products, producing lactic acid and volatile fatty acids (VFAs) which lower the pH and so help preserve the material (Meneses et al. 2007).

This study was carried out to ascertain the ensiling capacity of fresh banana packaging industries' by-products and their silage chemical composition and fermentative characteristics, oriented towards a better utilization of banana by-products to be used for feeding small ruminants in order to improve the benefit of local farmers.

Furthermore, the banana by-product silage could be a very interesting material for certain sub-Saharan African countries as Senegal with huge shortages of animal feed that are gradually increasing banana production.

2. Materials and methods

2.1. Silage samples

Taking into account that the ability of a product to be silage depends on the values of dry matter (DM), water soluble carbohydrates (WSC), buffering capacity (BC) and nitrate, five combinations representing the product [100% green banana (GB), 100% bunch (b), 30% GB–70% b, 50% GB–50% b and 100% ripe banana (RB)] were taken from various banana packaging industries in Tenerife (Canary Islands, Spain).

2.2. Chemical analyses

All samples selected were analysed in duplicate for ensilability characteristics, including DM determined by drying in an air-forced oven at 102°C for 24 h, WSC following Hoffman (1937), BC using the methodology described by Playne and McDonald (1966) and nitrate. The ratio of fermentability (FR) was calculated according to Schmidt et al. (1971). According to the results obtained, three combinations were selected (100% GB, 50% GB–50% b, and 50% GB–50% b + additives) in order to assess the quality and stability over time of the experimental banana microsilages. In the third combination, 10% molasses (47% soluble carbohydrates) and 7% beet pulp (10% moisture) were used as additives. For this purpose, all material was ensiled in laboratory silos made of PVC cylinders provided with bun valves to allow for gas losses and glass containers to store the evacuated effluent. These laboratory silos have a capacity of 4 dm³ and a forage density of 650 kg m⁻³, according to Martínez-Fernández and De la Roza-Delgado (1997).

Chemical composition and silage suitability were evaluated after 45 days, 3 and 6 months. The containers were individually opened. Three samples were taken from the top, the centre and the bottom of each ensiled mass and were frozen at –18°C until analysis.

The fermentation process was evaluated from final metabolites in silage juice in the Animal Nutrition Laboratory in SERIDA (Asturias) for pH, soluble-N and ammonia-N, WSC, lactic acid and VFAs including acetic, propionic and butyric acids (VFAs).

Ammonia-N was determined by ultraviolet-visible spectroscopy (UV-Vis Ammonium test. Merck, Germany) and soluble-N was determined by the Kjeldahl procedure. Lactic acid and VFAs analyses were performed by high-performance liquid chromatography (HPLC) with a Water Alliance 2690 instrument equipped with a Waters 996 Photodiode Array Detector Module (Waters. Milford, Massachusetts) monitoring at 206 nm and driven by Millenium software. The juice extract were analysed on a Shodex RSpak KC-811 column (Waters) using a mobile phase with 0.025% phosphoric acid. Flow rate was 1.0 ml min⁻¹ and column temperature for analyses was kept at 40°C.

The chemical composition was determined after being freezer dried, and the milled samples were passed through a 1-mm sieve, for DM, ash, organic matter (OM) and crude protein (CP) according to the Association of Official Agricultural Chemists (AOAC 1984), and acid detergent fibre (ADF), neutral detergent fibre (NDF) and acid detergent lignin (ADL) according to Van Soest et al. (1991).

2.3. Statistical analyses

The software package SPSS version 15.0.1 (SPSS Inc. Chicago, IL, USA) was used for statistical processing of the results. Analyses of variance were performed for comparing the fermentative and chemical values after different ensilage times. Post hoc multiple analyses by Tukey's test were used for the ripening time factor and the combination factor.

3. Results and discussion

The quality of silage fermentation is due to two factors: the nature of the starting material and the proper development of the technique. The ensilability depends mainly on the DM content, WSC, resistance to acidification or buffering capacity and nitrate content.

Table 1 shows the silage suitability (ensilability characteristics) results of bananas packaging by-products from five different combinations of GB, b and RB. The sample of 100% RB showed the best aptitude for ensiling: high sugar content and low nitrates values. In addition, DM and buffering capacity exceeds the minimum levels required for proper fermentation. Despite the good quality of this combination, it was not considered suitable for later use due to the small quantity available in the packaging industry and its complex management because of its high FR and rapid putrefaction. Acceptable aptitude for ensiling showed 100% GB combination with adequate DM, WSC and BC and 50% GB–50% b combination, despite their low WSC content. The 100% b and 30% GB–70% b combinations behaved worst with low values of DM, WSC and FR ratio and excessively high values of BC (Table 1).

Table 1. Ensilability characteristics of banana by-product (five combinations).

Combinations	DM (%)	WSC (%)	BC	Nitrite (ppm)	Nitrate (ppm)	FR
100% GB	21.36	17.27	258	no	<10	26.9
100% b	7.63	11.29	645	no	<25	9.1
30% GB/70% b	11.43	8.59	385	no	<25	13.3
50% GB/50% b	15.86	9.58	279	no	<25	18.8
100% RB	21.27	50.50	320	no	<10	34.7

DM, dry matter; WSC, water soluble carbohydrates; BC, buffering capacity (meq NaOH kg^–1MS); FR, fermentability ratio; GB, green banana; b, bunch; RB, ripe banana.

Based on these results, the experimental combinations for studying the quality and stability over time were 100% GB, 50% GB–50% b alone and 50% GB–50% b + additives (molasses and beet pulp). The latter combination was composed with the purpose to increase the initial DM and WSC content which in turn means an improved FR coefficient.

The fermentative characteristics of banana by-products silages for the fermentation periods considered are shown in Table 2. As expected, the better combination result was GB/b + additives. Soluble-N and ammonia-N values below 50% and 10%, respectively, denoted adequate protein degradation during the fermentative process (INRA 1981). Silage is considered excellent when the NH₃-N/N content is below 7 g/100 g of the total N and considered good when the NH₃–N/N content is between 7 g and 10 g/100 g of the total N (Romero 2004; Lima et al. 2010).

Table 2. Final metabolites in silage juice of banana by-product (three combinations) in different fermentation periods (45, 90 and 180 days).

		Combinations
Silage fermentation parameters	Fermentation periods (days)	GB	GB/b	GB/b + additives
pH	45	5.27 ± 0.08^a	5.15 ± 0.04^a	3.99 ± 0.03^b
	90	4.87 ± 0.31^a	5.24 ± 0.29^a	3.98 ± 0.02^b
	180	4.96 ± 0.02^b	5.14 ± 0.01^a	3.96 ± 0.03^c
NH3-N (% tot N)	45	19.13 ± 3.91^a	7.26 ± 0.60^b,y	6.75 ± 0.34^b
	90	20.83 ± 1.54^a	11.04 ± 0.17^b,x	6.90 ± 0.20^c
	180	25.61 ± 0.07^a	9.76 ± 0.03^b,x	5.94 ± 0.97^c
Soluble-N (% tot N)	45	37.02 ± 4.92^a	17.04 ± 2.25^b	43.94 ± 2.47^a
	90	36.35 ± 1.21^b	17.27 ± 1.12^c	45.24 ± 1.95^a
	180	39.56 ± 0.19^a	20.75 ± 0.32^b	45.91 ± 4.17^a
WSC	45	0.10 ± 0.05^a	0.16 ± 0.04^a,x	2.71 ± 0.74^b
	90	1.05 ± 0.95	0.15 ± 0.02^x	1.96 ± 0.12
	180	0.23 ± 0.13^b	0^b,y	2.90 ± 0.70^a
Lactic acid	45	0.81 ± 0.98^a	1.25 ± 0.37^a	6.75 ± 0.21^b
	90	1.37 ± 1.52^b	0.33 ± 0.07^b	6.45 ± 0.22^a
	180	0.10 ± 0.01^b	0.73 ± 0.01^b	6.39 ± 0.38^a
Acetic acid	45	2.37 ± 0.81	1.59 ± 0.30	1.48 ± 0.05^x
	90	3.55 ± 0.38^a	1.84 ± 0.10^b	1.68 ± 0.03^b,y
	180	3.80 ± 0.04^a	2.24 ± 0.03^b	1.87 ± 0.01^c,z
Propionic acid	45	0.33 ± 0.17	0.76 ± 0.63	0
	90	0.34 ± 0.01^a	0.34 ± 0.10^a	0^b
	180	0.52 ± 0.01^a	0.44 ± 0.01^b	0^c
Butyric acid	45	1.36 ± 0.83^a	1.22 ± 0.41^ab	0^c
	90	2.21 ± 0.63^a	1.89 ± 0.09^a	0^b
	180	2.44 ± 0.16^a	1.20 ± 0.05^b	0^c

WSC, water soluble carbohydrates; GB, green banana; b, bunch.

^a–cValues in the same row without a common superscript letter are significantly different (P < 0.05).

^x–zValues in the same column, within the same parameter, without a common superscript letter are significantly different (P < 0.05).

Moreover, high amounts of lactic acid in these samples indicated an optimal conversion of carbohydrates contributing to the reduction of pH and the stability of silage. Lactic acid is the most effective acid for reducing silage pH thereby conserving the high quality of the forage. The ensiled forages with higher lactic acid concentrations tend to be more nutritive, more palatable and sweet smelling, which results in greater nutrient intake (Meneses et al. 2007). The lactic acid concentration usually falls when the forage water content and the amount of oxygen trapped in the ensiled forage during filling of the silo are high, resulting in the growth of facultative aerobic micro-organisms, such as coliforms (Stefanie et al. 2000).

Certain additives, such as molasses with high levels of soluble carbohydrate, are likely to be able to produce sufficiently high levels of lactic acid improving the fermentation and quality of silage (Fagbenro & Jauncey 1998; Evers & Carroll 1998; Machin 2000).

According to Table 2, the additives increased the lactic acid content and decreased acetic and butyric acid content. Furthermore, the combined silages with additives resulted in a lower pH and ammonia content than the silages without additives. These results are consistent with the results of other experiments (Umaña et al. 1991; Ojeda & Montejo 2001; Lima et al. 2010) in which addition of molasses to forage resulted in silages with lower pH, increased lactic acid concentration and lower ammonia content.

The acetic acid contents in GB/b and GB/b + additive silages were below 2%. Above this level, undesirable fermentations occur, such as those resulting from the action of the Coliforme genus bacteria, heterofermentative lactic bacteria and Clostridic bacteria (Allen et al. 1995).

Propionic and butyric acid were not detected in GB/b + additive combination. The percentage of propionic acid did not vary with the conservation period, while acetic and butyric acid increased till 90 days. In the present study, the butyric acid amounts in GB/b + additive combination were compatible with good-quality silage, according to Allen et al. (1995), who reported 0.1% content level as a critical concentration for an adequate fermentation process. GB and GB/b samples showed higher contents (>0.1%) so they were classified as bad-quality silage.

Moreover, the absence of butyric acid and a low content of VFAs expressed as (acetic + propionic + butyric) characteristic of the GB/b + additive combination has an excellent quality silage with a high stability over time. The low pH obtained by this combination, which is usually accomplished through the fermentation of sugars in the crop to lactic acid by lactic acid bacteria, decreases plant enzyme activity and prevents the proliferation of clostridia and enterobacteria (Woolford 1984). The pH value is a key criterion to evaluate silage fermentation. Proteolysis is inhibited more strongly by pH than by osmolarity (Bickel et al. 2005). Generally the lower the pH, the better preserved and more stable is the silage. The critical pH that affects growth clostridium depends on the water activity of the fermentation medium; when decreasing water activity, the sensitivity of these micro-organisms to acidity increases. Therefore, the DM content of the forage affects the pH required for quality silage (Revuelta et al. 2008).

In this study, GB and GB/b presented excessively high rates of ammonia-N, VFAs and specially butyric acid. The results obtained classified these combinations near the limit considered as poor quality silage (Martínez-Fernández et al. 1999). The presence of butyric acid in the silage material is related to the depressive effect of the Clostridium genus bacteria, with significant losses in silage quality and consequently reduced acceptability and intake (Wilkinson 1983). By contrast, their stability over time was considered acceptable as the values remain relatively constant throughout the conservation process.

The three combinations of banana by-product silage studied presented a moderate DM content and a low protein value (Table 3), with lower values in the GB/b combination without additive. The OM content (93–94%) and ash (6–7%) may be considered acceptable and similar to other silage based in agro-industrial by-products in the GB combination being significantly lower in the other two combinations. The fibrous fractions content is also low but higher in those samples that include GB and bunch together.

Table 3. Chemical composition of banana by-product (three combinations) in different fermentation periods (45, 90 and 180 days).

		Combinations
Chemical composition	Fermentation periods (days)	GB	GB/b	GB/b + additives
DM	45	30.04 ± 1.61^a	19.82 ± 4.09^b	26.57 ± 1.38^a,b
	90	31.05 ± 1.12^a	20.70 ± 1.15^c	26.75 ± 0.14^b
	180	28.09 ± 0.59^a	20.19 ± 0.86^b	27.96 ± 1.15^a
OM	45	92.45 ± 1.66^a	87.36 ± 2.66^b	85.48 ± 0.18^b,x
	90	93.46 ± 0.61^a	88.04 ± 3.13^b	84.87 ± 0.09^b,y
	180	94.13 ± 0.01^a	91.79 ± 0.19^b	84.75 ± 0.18^c,y
Ash	45	7.05 ± 1.66^b	12.64 ± 2.66^a	14.52 ± 0.18^a,x
	90	6.54 ± 0.51^b	11.96 ± 3.13^a	15.13 ± 0.09^a,y
	180	5.87 ± 0.01^c	8.20 ± 0.19^b	15.25 ± 0.18^a,y
CP	45	5.14 ± 0.01^b	4.81 ± 0.30^b	6.21 ± 0.11^a
	90	5.42 ± 0.15^b	4.47 ± 0.41^c	6.27 ± 0.03^a
	180	5.04 ± 0.13^b	4.29 ± 0.14^c	6.56 ± 0.22^a
NDF	45	11.41 ± 5.18	21.39 ± 5.45	16.32 ± 1.96
	90	10.53 ± 0.49^c	18.78 ± 0.531^b	15.89 ± 0.23^b
	180	8.15 ± 1.07^b	13.42 ± 2.58^a	17.17 ± 0.43^a
ADF	45	8.52 ± 3.08	14.61 ± 4.14	11.83 ± 1.47
	90	7.88 ± 0.39^b	13.67 ± 3.12^a	11.67 ± 0.36^ab
	180	6.69 ± 0.42^b	11.78 ± 1.84^a	13.80 ± 0.57^a
ADL	45	3.38 ± 1.88	3.81 ± 1.62	2.23 ± 0.28^x
	90	3.25 ± 0.40	4.16 ± 1.92	3.00 ± 0.30^y
	180	3.05 ± 0.16	2.47 ± 0.47	2.69 ± 0.25^xy

GB, green banana; b, bunch; DM, dry matter; OM, organic matter; CP, crude protein; NDF, Neutral detergent fibre; ADF, acid detergent fibre; ADL, acid detergent lignin.

^a–cValues in the same row without a common superscript letter are significantly different (P < 0.05).

^x–zValues in the same column, within the same parameter, without a common superscript letter are significantly different (P < 0.05).

The use of additives increases the initial cost of the silage, although its better fermentation and nutritive quality offset this higher price. Recent studies guarantee the inclusion of this feed in the diet of dairy goats, showing any negative effects on the performance of the animals, similar milk yield and better milk quality to standard diets (Álvarez et al. 2011a). Furthermore, cheese evaluations showed textures, odours and flavours that correspond with the typical characteristics of Canarian goat raw milk cheeses (Álvarez et al. 2011a).

4. Conclusion

The banana packaging industry by-products showed a good aptitude for ensiling especially when some additives were added to the raw material to improve DM and soluble carbohydrates content. Its fermentative composition was acceptable, and low changes in chemical composition during silage were detected, although its moderate nutritional value recommends a limited inclusion in ruminant diets. However, further studies are required in order to assess palatability and to observe feed intake. In addition, the influence of banana silage diets on milk and cheese production must be checked.

Funding

This work was supported by the Spanish Government (project RTA2008-00108) and EU Transnational Cooperation Programme (project MAC/3/C188 GANAFRICA) both with FEDER funds (European Union).

Additional information

Funding

Funding: This work was supported by the Spanish Government (project RTA2008-00108) and EU Transnational Cooperation Programme (project MAC/3/C188 GANAFRICA) both with FEDER funds (European Union).