The feeding value of deseeded pods from Moringa stenopetala and Moringa oleifera as evaluated by chemical analyses and in vitro gas production

This study evaluates the nutrient composition and in vitro fermentation characteristics of deseeded green pods of Moringa stenopetala and Moringa oleifera cultivated at low and moderate altitudes. Crude protein (CP) content (g/kg DM) varied from 103 in Moringa oleifera to 135 in Moringa stenopetala. The CP contents for Moringa stenopetala cultivated at low and moderate altitudes were 135 and 127 g/kg DM, respectively. The CP values for Moringa oleifera were 103 and 105 g/kg DM at low and moderate altitudes, respectively. Low values of neutral detergent fibre, acid detergent fibre, and cellulose were found in Moringa stenopetala. High concentrations of Ca, P, K, Mg, Mn, and Cu were observed in Moringa oleifera. Significantly high values of metabolizable energy (ME), organic matter digestibility, and short-chain fatty acids were found in Moringa stenopetala. These values were also significantly high at moderate altitude. The ME values were 7.35 and 5.80 MJ/kg DM for Moringa stenopetala and Moringa oleifera, respectively. In conclusion, deseeded pods of the Moringa tree could be used as an alternative, cheap source of home-grown energy supplements for low quality crop residues of tropical livestock while using the seeds for human consumption.


INTRODUCTION
The livestock sector plays a significant economic role in most developing countries, and is essential for the food security of rural populations.The productivity of farm animals in most tropical countries is generally low, mainly due to poor quality and inadequacy of available feeds.Moreover, conventional feed resources (grains, cereals, legumes, etc.) for animal production are scarce and highly expensive in many parts of the world.Thus searching for alternative unconventional feed sources that may have valuable components of animal diets is indispensable.For instance, feeding by-products from agricultural and food processing industries to livestock can be one of the solutions (Negesse et al., 2009;Szumacher-Strabel et al., 2011;Zhou et al., 2012).The use of tree parts as alternative feed resources for ruminant livestock is becoming increasingly important in many parts of the tropics and sub-tropics (Silanikove, 2000;Melesse et al., 2009).Moringa trees are multi-purpose trees of economic importance with several industrial and feeding values.The genus Moringaceae is represented by 14 species to which Moringa stenopetala and Moringa oleifera belong.M. stenopetala has a wide range of adaptation from arid to humid climates and can be grown in a various land use patterns.It is cultivated both for human food and animal feed in Southern Ethiopia and has been recently distributed to other regions of Ethiopia beyond its original sites.M. oleifera is native to the sub-Himalayan tracts of north-west India, Pakistan, Bangladesh and Afghanistan (Makkar and Becker, 1997).It is a pan-tropical multipurpose tree and is characterized by high biomass yield and can tolerate unfavourable environmental conditions (Foidl et al., 2001).
Although studies have reported the chemical composition of leaves, seeds and seedpods of both Moringa species (Makkar and Becker, 1996;Melesse et al., 2009), no information has been available on the chemical and mineral compositions as well as ruminal fermentation characteristics of deseeded green pods as an alternative animal feed sources in the tropics.The objectives of this study were thus: 1. to assess the altitudinal variations in chemical and mineral compositions of deseeded pods from M. stenopetala and M. oleifera cultivated at low and moderate altitudes; 2. to evaluate the potential of this Moringa tree part as a ruminant feed using an in vitro gas production method.

Sampling sites
Samples of whole green pods from M. stenopetala and M. oleifera were collected from nursery sites of the Southern Agricultural Research Institute located in districts of Hawassa (altitude 1700 m a.s.l.) and Arbaminch (altitude 1100 m a.s.l.) representing moderate and low altitudes, respectively.The Arbaminch site (low altitude) lies between 06 o 03' latitude north and 37 o 33' longitude east and has a warm agro-ecology with an average rainfall of less than 900 mm (with a range of 640-1130 mm), that peaks during the months of July and August (National Metrological Agency).Similarly, the Hawassa site (moderate altitude) lies between 07 o 03' latitude north and 38 o 28' longitude east and has a sub-moist warm agroecology with an average rainfall of less than 1000 mm (with a range of 700-1200 mm), the peak being during the month of August.The mean monthly maximum and minimum temperatures of the Hawassa site are 27.5 and 13.3 o C, respectively.The corresponding temperature values for the Arbaminch site are 29.9 and 16.8 o C (National Metrological Agency).

Sample collection and preparation
Samples were collected from nursery sites during the dry season (December) in 2009 from 6-year-old Moringa trees cultivated at low and moderate altitudes.From each Moringa species, three trees were randomly sampled per altitude from which whole green pod samples were collected.Deseeded pods were then prepared by manual removal of seeds from each whole green pod, chopping the pods with a knife, and partially sun-drying them to reduce the moisture content.Samples collected from three trees were handled separately and dried at 65 o C for 48 h and ground to pass a 1 mm sieve.Ground feed samples were kept in sealed plastic containers and transported to Hohenheim University (Germany) for analysis.

Chemical analysis
Analyses of proximate nutrients and fibre fractions were performed as outlined by Verband Deutscher Landwirtschaftlicher Untersuchungs-und Forschungsanstalten (VDLUFA, 2006).The samples were analysed for DM, ash, crude protein (N multiplied by 6.25), petroleum ether extract, and crude fibre.Neutral detergent fibre (NDF) assayed with a heat stable amylase and acid detergent fibre (ADF) and acid detergent lignin (ADL) were analysed according to VDLUFA (2006) and were expressed inclusive residual ash.Cellulose and hemicellulose were computed as ADF minus ADL and NDF minus ADF, respectively.The nonfibre carbohydrate (NFC) content was calculated as 100-(NDF + CP + crude fat + ash) according to NRC (2001).
For mineral analysis, samples were incinerated at 550 o C and the remaining ash was treated with 6 mol per l HCl.Minerals were determined from filtered ash solutions using an Inductively Coupled Plasma spectrometer (ICP-OES) (Rodehutscord and Dieckmann, 2005).
All chemical analyses were conducted in duplicate on each individual sample.

In vitro measurement protocols
Gas production was determined according to the procedure of the VDLUFA official method (VDLUFA, 2006) and Menke and Steingass (1988).About 200 mg of feed sample were weighed in two replicates and transferred into 100 ml calibrated glass syringes, fitted with pistons.To prepare the inoculum, rumen fluid was collected before the morning feeding from two rumen-fistulated, nonpregnant and non-lactating Holstein-Friesian cows.The cows were fed twice a day at 8 a.m. and 4 p.m. and received a daily quantity of 8 kg meadow hay and 2 kg of a commercial dairy concentrate (18% CP, 7.6 MJ Net Energy for Lactation).Water and a mineral lick were available ad libitum.Rumen fluid was manually pumped directly into pre-warmed thermo flasks and taken immediately to the laboratory.
The rumen fluid was then filtered through two layers of cheesecloth and diluted with buffered mineral solution, which was maintained in a water bath at 39°C under continuous flushing with CO 2 .In each run, three incubation units each with Hohenheim gas test standard hay and standard concentrate were included, serving as control units for successful incubation.Moreover, six parallel glass syringes that contained rumen fluid-media mixtures without substrate were used as blanks.A total of 30 ml incubation medium (consisting of 10 ml rumen fluid, 5 ml of bicarbonate buffer, 5 ml of macro-mineral solution and 10 ml of distilled water) was dispensed into pre-warmed glass syringes containing the respective experimental feeds (200 mg), reference standards and blank syringes.After gently shaking the syringe and removing air bubbles, the clip on the silicon tube attached to the tip of the syringe was closed, initial reading recorded, and the syringe was placed in a temperature-controlled incubation rotor set at 39 o C. Incubation was completed in duplicate within each run; runs were replicated two times yielding four observations per sample.The gas volume was recorded at 2,4,6,8,10,12,14,18,24,30,36 and 48 h of incubation according to the time described by Blümmel and Becker (1997).The volume of gas in a glass syringe was readjusted to 30 ml whenever it exceeded 60 ml.

Statistical analysis
The experimental procedure was a completely randomized design having two factors, in which factor 1 was Moringa species with two categories (M.oleifera and M. stenopetala) and factor 2 was altitude with two categories (low and moderate altitudes).From each Moringa species, three trees were randomly sampled per each altitude consisting of a 2 x 2 factorial ANOVA design with a total sample size of 12. Data were subjected to ANOVA with Moringa species and altitudes as main effects and all two-way interactions using General Linear Models (GLM) Procedure of Statistical Analysis System (SAS, 2002).Differences of means were separated by Duncan Multiple Range Test.All statements of statistical differences were based on P<0.05 unless noted otherwise.Time series measurements of gas volumes from 2-48 hrs of in vitro incubation were used for the curve fitting to mathematically express gas production over incubation time.Model fitting for gas production kinetics and parameter estimation were done according to Beuvink and Kogut (1993) described in detail by Boguhn et al. (2008) by using the software GraphPad Prism 4.02 for Windows (GraphPad Software, Inc. La Jolla, CA, USA).

Chemical compositions
The chemical compositions of deseeded pods of both Moringa species at moderate and low altitudes, including two-way interactions, are presented in Table 1.Except for hemicellulose, the effect of Moringa species was significant for all investigated nutrients.Similarly, altitude had a significant effect on all nutrients except CP, ADL, and hemicellulose.Significant interactions of Moringa species x altitude were also observed in all compounds except for ADL and hemicellulose contents.
The ash content was generally higher for M. oleifera cultivated at both altitudes than for M. stenopetala with significant altitude × Moringa species interactions.The highest and lowest CP values were noted for M. stenopetala and M. oleifera cultivated at low altitude, respectively, with significant altitude × Moringa species interactions.The fat content was generally highest for M. stenopetala cultivated at low altitude, but lowest for M. oleifera at moderate altitude, as indicated by the highly significant altitude-by-species interactions.The NFE and NFC contents of M. stenopetala were generally higher than those of M. oleifera.The contents CF, NDF, ADF and cellulose were generally lower, however, in M. stenopetala than in M. oleifera at both altitudes.Generally, nutritive value parameters in M. oleifera were little affected by altitude, whereas large changes in ash, fat, CF, NDF, ADF, cellulose and NFC due to altitude were observed in M. stenopetala.As a result, the contents of ash, fat, CP, CF, NDF, ADF, ADL and cellulose in M. stenopetala were generally higher at low altitude than those at moderate elevation.In M. oleifera, however, the contents of ash, CP, CF, NFE, NDF, ADF, ADL and NFC were similar between altitudes.

Mineral concentrations
The mineral composition of both M. stenopetala and M. oleifera as affected by altitude and their two-way interactions is presented in Table 2.The effect of Moringa species was significant for all investigated minerals except for Zn.The effect of altitude was significant for P, Na, Fe, Zn and Cu.The interactions between species and altitude were significant for all investigated minerals except for P and Mn.
The Ca, P, K, Mg, Mn, and Cu concentrations for M. oleifera were generally higher than those of M. stenopetala cultivated at both altitudes.The Ca concentration (g/kg DM) ranged from 2.47 in M. stenopetala to 3.82 in M. oleifera grown at moderate altitude.The highest Zn concentration for M. stenopetala was noted at moderate altitude and the lowest at low altitude.The Ca, K, Mg, Fe and Cu concentrations for M. stenopetala were generally higher at low altitude; whereas those of P, Na, Mn and Zn concentrations were higher at moderate altitude.In contrast, Ca, P, K, Mg, Fe, Mn, and Zn concentrations for M. oleifera were generally higher at moderate altitude than at low altitude.

In vitro gas production characteristics and estimated parameters
Among the investigated Moringa types, M. stenopetala produced a significantly higher in vitro gas volume than M. oleifera (Figure 1).The increase in gas volume was highest during the initial phase of incubation and consistently increased thereafter.
As presented in Figure 2, although not significant, the in vitro gas production of deseeded pods cultivated at moderate altitude was higher than that of at low altitude.In general, although the in vitro gas volume increased with advancing time of incubation, the greatest proportion of gas production occurred in the first 24 h of incubation (Figures 1 and 2).The corrected 24 h gas volume, in vitro estimated values of ME, OMD, and SCFA as affected by altitude and Moringa species are presented in Table 3.The values of these parameters corrected 24 h gas volume and values of ME, OMD and SCFA were significantly higher at moderate altitude than at low altitude.Moreover, there were significant differences among Moringa species in the corrected 24 h gas volume, and estimated parameters of ME, OMD, and SCFA values.Accordingly, M. stenopetala had significantly higher corrected 24 h gas volume and values of ME, OMD and SCFA than those of M. oleifera (Table 3).As presented in Table 4, at low altitude, the average plateau value of gas production (parameter b) for M. stenopetala and M. oleifera was 28.0 and 27.03 ml, respectively, whereas the corresponding values at moderate altitude were 37.54 and 21.90 ml.At both altitudes, the average rapid gas production rate during early stages of fermentation (parameter mr) was generally higher for M. stenopetala than M. oleifera.Nonetheless, the average values of slower gas production rate during the later stages of fermentation (parameter ms) from M. oleifera were comparatively higher than obtained from M. stenopetala cultivated at both altitudes (Table 4).
The average maximum gas production rate was also generally higher for M. stenopetala at both altitudes.The average maximum gas production rate ranged from 1.484 ml/h for M. oleifera to 4.35 ml/h for M. stenopetala cultivated at moderate altitude.At both altitudes, M. oleifera had a longer lag time and point of inflection time than M. stenopetala (Table 4).means within a column with no common superscripts differ significantly (P<0.05) 1 b -the plateau value of gas production (ml); μr -the rapid gas production rate (ml/h) during early stages of fermentation; dr -the fractional decay constant for mr; ms -the slower gas production rate (ml/h) during later stages of fermentation; ds -the fractional decay constant for ms DISCUSSION Chemical compositions.To the author's knowledge, no work has been published on the chemical composition of deseeded green pods of M. stenopetala and M. oleifera.Some studies on the edible portion of M. oleifera pods have been reported, but comparisons with our results (expressed on a DM basis) are difficult.The CP content of seedpods of M. stenopetala reported by Melesse et al. (2009) was much lower than that found in the current study.The NDF and ADL values for M. stenopetala are similar to those reported for lucerne hay (Bueno et al., 2010), but the NDF, ADF, and ADL contents for M. stenopetala were generally lower than those reported for four varieties of cassava leaves by Oni et al. (2010).
Considerable differences in the morpho-physiological traits of plants were observed when they were grown at different altitudes (Todaria and Purohit, 1979).In the present study, a significant interaction between altitude and Moringa species was noted for most chemical and mineral compositions.Singh et al. (2010) observed significantly higher ash and CP contents in foliages of Celtis australis L. grown at high altitude than those from low altitude.The present study suggests that altitude significantly influenced the chemical composition but less the mineral components of both Moringa species.
Diets containing high K levels are known to reduce Mg absorption (Jittakhot et al., 2004) and cause an increase in the percentage of Mg excreted in the faeces.Thus, the high K in M. oleifera might affect Mg absorption.High levels of Fe have been observed and may interfere with the absorption of Zn, Cu and Mn, as reported by Gengelbach et al. (1994).The P, K and Ca concentrations showed a strong positive correlation with elevation range of foliage.On average, high altitude foliage exhibited comparatively higher values for P, Ca and K (Singh et al., 2010).Similarly, the concentrations of Ca, P, K, Mg, Fe, Mn, and Zn were higher in M. oleifera cultivated at moderate altitude.
In vitro fermentation characteristics and estimated parameters.Gas volume produced at 24 h from M. stenopetala was comparable with that of Abas et al. (2005) for wheat straw and Oni et al. (2010) for four varieties of cassava leaves.The incubated feed samples from M. stenopetala produced significantly more gas than those from M. oleifera.The explanation could be that M. stenopetala contained more ME and NFC than M. oleifera, and these components are positively associated with the gas production potential of feedstuffs (Tylutki et al., 2008).The low gas volume observed in M. oleifera might be caused by the presence of highly fibrous substances, as shown in Table 1.High fermentation rates indicate high nutrient availability for ruminal microorganisms, while lower rates may be the result of greater NDF and ADF contents (Table 1); the chemical components of NDF and ADF may slow down the speed of substrate fermentation (Fievez et al., 2005).Thus, the extent of in vitro fermentation in M. stenopetala suggests that it is of higher nutritional value than that of M. oleifera.
The calculated high maximum gas production rate of 4.4 ml/h for M. stenopetala cultivated at moderate altitude was comparable to that of lucerne leaves (4.1 ml/h) reported by Bulang (2005), but the lag time (time for microbial attachment and start of degradation) in the present study was more than tenfold smaller than obtained from lucerne leaves (2.6 h).Hence it is most likely that the degradability and consequent nutrient availability, especially nitrogen in ruminants, will be increased at a faster rate.The observed shorter lag time for M. stenopetala from both altitudes might be related to the high content of fermentable carbohydrates (relatively high NFC content, Table 1).The lag time reported by Makkar and Becker (1996) for M. oleifera leaves (0.23 h) was twofold lower than found in the present study of the same Moringa species.The overall lag times observed in the present study are generally lower than those reported by Oni et al. (2010) for four varieties of cassava leaves.
M. oleifera cultivated at both altitudes had the longest lag times.This would indicate slower nutrient availability from deseeded pods of M. oleifera.Nonetheless, the influence of high structural carbohydrates (NDF, ADF and cellulose) and low concentrations of soluble carbohydrates (notably low NFC value) in M. oleifera on the gas production profile should be taken into account when explaining the long lag time.
The livestock feed resources in tropical and subtropical regions are usually poor quality for the majority of any feed year and are deficient in critical nutrients for efficient microbial growth in the rumen.Low quality feed resources with high fibrous contents are considered the primary reasons for high methane gas emissions by ruminants.The green pods might contain phytochemicals, like tannins and flavonoids, that may have antimethanogenic properties that could act as chemical inhibitors of ruminal methane formation, as suggested by Szumacher-Strabel et al. (2010, 2011).Phenolic compounds (e.g., tannins), may reduce rumen methanogenesis in sheep and cattle (Boadi et al., 2004), while flavonoids have been demonstrated to modify microbial metabolism in the rumen (Broudiscou and Lassalas, 2000).It would be thus helpful to further study the presence of anti-nutritional factors (tannins, phenols, etc.) in deseeded pods of both Moringa species and its effect on the performance of farm animals.
In vitro rumen gas production has been extensively used to accurately predict the ME content of a wide variety of feeds (Melesse et al., 2009).These authors reported 5.1 MJ/kg DM of ME for M. stenopetala seedpods, which is comparable to that of M. oleifera in the current study.In agreement with the present findings for M. oleifera, Magalhaes et al. (2010) reported a ME value of 5.66 MJ/kg DM for elephant grass.
A high correlation has been reported between in vivo OMD and in vitro gas volume at 24 h (Menke et al., 1979).The OMD value obtained from M. oleifera in the present study was comparable with that of lucerne hay (44.7%) and wheat straw (45.2%) reported by Sallam et al. (2008) and Abas et al. (2005), respectively.Similarly, the calculated OMD values for grass and vetch hays reported by Abas et al. (2005) were in good agreement with the current findings for M. stenopetala.These findings suggest that deseeded pods from both Moringa species, particularly from M. stenopetala, could be used as an alternative energy supplement to tropical ruminants when the availability of improved forages and grasses becomes scarce during the dry season.
Since the SCFA content is an indicator of the energy value of diets, its prediction from in vitro gas measurements is useful under circumstances where laboratories lack gas chromatography equipment, especially in developing countries.The high production of gas and predominance of SCFA in M. stenopetala vs M. oleifera could probably describe an increased proportion of acetate and butyrate but a decrease in propionate production.Moreover, M. stenopetala contains more fermentable carbohydrate (notably NFC and NFE) which is a vital substrate for growth of ruminal microorganisms (Van Soest, 1994).In general, the ME, OMD, and SCFA values obtained from the present study were higher than those reported by Negesse et al. (2009) for agro-industrial and kitchen waste products.

CONCLUSIONS
M. stenopetala had a better chemical composition and in vitro degradability of nutrients, while M. oleifera had comparatively high mineral concentrations.The moderate altitude appears to have favored nutrient compositions and in vitro degradability characteristics, suggesting its suitability for Moringa cultivation for better utilization.The present findings further suggest that deseeded green pods may be used as alternative and sustainable sources of home-grown energy supplements for low quality crop residues in livestock feeding under smallholder farmer conditions, while using the seeds for human consumption.The effects of feeding deseeded green pods to ruminants on their performance require further studies, however.

Figure 1 .
Figure 1.Pattern of in vitro gas production in deseeded pods of M. stenopetala and M. oleifera measured over 48 h of incubation

Figure 2 .
Figure 2. Development of in vitro gas production in deseeded pods cultivated at low and moderate altitudes measured over 48 h of incubation

Table 2 .
Concentrations of macro and trace minerals in deseeded pods of M. stenopetala and M.

Table 3 .
Corrected 24 h gas production, in vitro estimates of ME, OMD and SCFA in deseeded pods as affected by Moringa species and altitude (n=12 each, means and pooled standard error)