Studies to assess the suitable conservation method for tapioca leaves for effective utilization by ruminants

R Murugeswari, V Balakrishnan* and R Vijayakumar

Department of Animal Nutrition, Madras Veterinary College, Chennai, Pin 600 007 India
Tel: 91-044-25381506, Fax: 91-044-25362787
drbalakrishnanphd@yahoo.co.in

Abstract:

An experiment to assess the effect of processing of two varieties of tapioca (Manihot esculenta Crantz) leaves viz. White Rose (H226) and Mulluvaadi (MVD-1) on chemical and HCN composition was conducted.

The proximate principles of the leaves of both the varieties collected from six different fields were comparable to that of lucerne. The crude protein content of 20.2 and 20.7% in dry matter (DM) was recorded for White Rose (H226) and Mulluvaadi (MVD-1), respectively. It was also observed that the tapioca leaves had less structural carbohydrates than lucerne. The crude fibre and neutral detergent fibre content varied between varieties. The White Rose (H226) had significantly higher (P<0.05) crude fibre (20.6%) and neutral detergent fibre (48.3%) than Mulluvadi (MVD-1) containing 14.7% and 42.7% respectively. The mineral profile of leaves of both the varieties were well above the critical level. However, a wide Calcium:Phosphorus ratio 9.12:1 and 9.19:1 were observed in both White Rose (H226) and Mulluvaadi (MVD-1), respectively which is suggestive of the need to supplement a phosphorus source along with tapioca leaves. Tapioca leaves contained high level of HCN 1934 mg/kg DM (White Rose (H226) and 1143 mg/kg DM (Mulluvaadi (MVD-1)). Wilting for 24 hours reduced HCN level to 193 mg/kg DM for both the varieties with White Rose (H226) decreasing the HCN content more rapidly than Mulluvaadi (MVD-1) which is suggestive of shorter duration for wilting of White Rose (H226) over Mulluvaadi (MVD-1). Good hay and good silage could be prepared from tapioca leaves of both varieties. While HCN (mg/kg DM) content of White Rose (H226) silage (51.8) was significantly (P<0.01) lower than Mulluvaadi (MVD-1) silage (81.7), the hay made from Mulluvaadi (MVD-1) had significantly (P<0.01) lower (72.4) HCN content than White Rose (H226) hay (106.77) suggestive of White Rose (H226) suitability for ensiling and Mulluvaadi (MVD-1) suitability for hay making.

Key words: Cassava, hay, HCN, leaves, silage, tapioca

Introduction

Tapioca or Cassava (Manihot esculenta Crantz) was introduced in India during the later part of the18^th century. Tapioca is a crop of economic importance, both as a food and feed and as a raw material for industrial products. It is considered to be the cheapest source of carbohydrates amongst the cereals, tubers and root crops. In India, the yield of fresh tapioca roots is around 23,200 kg/hectare/year. At the time of harvest of the roots, abundant quantities of leaves are produced (Ravindran and Rajaguru 1988), but are under-utilized and are usually put directly on to the soil as compost (Khang and Wiktorsson 2000).

Tapioca leaves have a high crude protein content varying from 16.7 to 39.3% in dry matter (DM) (Allen 1984). A factor limiting its utilization as a feedstuff is the high content of cyanogenic glucosides, which give rise to the cyanide (HCN) toxin. Cyanide concentrations of nearly 1000 mg/kg DM have been reported in tapioca foliage (Man and Wiktorsson 2001; 2002). There are reports indicating that feeding large amounts of tapioca products without treatment could result in death of the animals, particularly non-ruminants (Hill,1973). In cattle and sheep, HCN can be lethal at 2 to 4 mg HCN/kg body weight (Kumar 1992).

It is in this context that a study was undertaken to evaluate the nutritive value of tapioca leaves as well as presence of anti-nutritional factors in order to suggest corrective measures and to explore the possibility of preserving tapioca leaves for usage during scarcity periods.

Materials and methods

Leaves from two commonly cultivated varieties of tapioca, namely White Rose (H226) and Mulluvaadi (MVD-1),  were collected randomly from six different fields in Tamil Nadu  after harvesting the roots..

The samples were divided into four groups from each collection site. One group was shade-dried and ground to pass through a 1 mm sieve in a Willey mill and preserved in airtight container for chemical analysis. The second group was made into hay, the third group was ensiled and the fourth group was allotted to a study of the effect of wilting at periodic intervals.

Chemical composition of Tapioca leaves

Samples were analysed for crude protein (CP), ether extract (EE), crude fibre (CF), nitrogen free extract (NFE) and total ash (TA) contents as per AOAC (1980). The neutral detergent fibre (NDF), acid detergent fibre (ADF) cellulose and lignin were analysed as per the method described by Goering and Van Soest (1970). The minerals in tapioca leaves viz. Calcium, Magnesium, Iron, Copper, Manganese and Zinc were determined using the Atomic Absorption Spectrophotometer (Perkin Elmer Model 3110) as per the procedure outlined in the reference book. Phosphorus was determined colorimetrically using ammonium vanadate (AOAC 1980).

Effect of conservation on the quality of tapioca leaves

Hay making

Tapioca leaves of both the varieties from each area of collection were dried in the shade for two days. Frequent turning was done to facilitate even drying. The tapioca hay thus prepared was then brought to the laboratory and stored in a safe place for further use. The physical quality of hay was evaluated according to a score card developed by Vough (2002). The evaluation of hay quality was carried out by six Animal Nutritionists. The evaluation included grading for physical characters such as maturity, leafiness, colour, odour, condition and presence of foreign material. Grading was done according to a 100 point score card. Samples of both the varieties of hay were analysed for proximate composition, fibre fractions and mineral profile as described earlier. The HCN content of the leaves was estimated as per the method of AOAC (1980).

Silage making

Tapioca leaves of both the varieties from each area of collection were chopped into pieces of 4-5 cm length and allowed to wilt under shade for three hours. Laboratory silage was prepared in a desiccator by adding 2% molasses and 1% salt to the chopped tapioca leaves which were then tightly packed and compressed to prevent any air space. The lid of the desiccator was sealed airtight with wax and tightly secured with thread. The desiccators were labeled and stored in a dark room. The desiccators were opened after two months. The tapioca silages were subjected to chemical analysis for proximate composition, fibre fractions, mineral profile and HCN content as described earlier. Organoleptic as well as biochemical characteristics were measured: pH, ammonia nitrogen, lactic acid, propionic acid, butyric acid. Silage pH was determined as per the method of Wilson and Wilkins (1972). Ammonia nitrogen concentration in the silage was estimated colorimetrically as per the method described by Weatherburn (1967). Total and individual volatile fatty acid concentrations were measured by gas chromatography as per the procedure of Chase (1990). Lactic acid in the silage was estimated colorimetrically as per the procedure of Barker and Summerson (1941). The "flieg" index was then calculated as per Zimmer (1966).

Studies on the effect of wilting on HCN content of tapioca leaves

The samples allocated for studies on the effect of wilting were immediately divided into five groups at the time of collection. Each group of leaves was then wilted for specified periods of time (0, 6, 9, 12 and 24 hours) and the HCN content estimated as described earlier. The HCN content at "0" hour, i.e at the time of harvest was carried out on fresh wet samples, while the rest were carried out at the end of the respective hours of wilting.

Statistical analyses

The data obtained for the different parameters were subjected to statistical analysis as per the procedure of Snedecor and Cochran (1967).

Results and discussion

Chemical composition of tapioca leaves (unprocessed)

Several workers (Reed et al 1982; Gomez and Valdivieso 1984; Ravindran and Ravindran 1988; Ravindran 1993; Sankaravinayagam et al 1999; Phuc et al 2000; Khang and Wiktorsson 2000; Man and Wiktorsson 2001) have reported the proximate composition of tapioca leaves. The analytical values of both varieties of tapioca leaves in the present study (Table 1) were within the range reported by the above workers.

The crude protein content of the tapioca leaves in this study was similar to that in leguminous fodder. The crude protein content of lucerne for example is on an average 20% in DM (McDonald et al 1999) in the pre-budding stage. The high level of ether extract (9-10% in DM) infers that tapioca leaves are a rich source of energy among forages, with a considerable quantity of soluble ash suggesting good availability of the minerals. The fibre fraction of both the varieties of tapioca leaves were in agreement with reports by Ravindran and Ravindran (1988), Sankaravinayagam (1999) and Man and Wiktorsson (2001) but differs from that quoted by Reed et al (1982) and Khang and Wiktorsson (2000). Variety and stage of maturity might be the possible reason for the differences in the results. Tapioca leaves contained less structural carbohydrate than lucerne. The cell contents (Neutral Detergent Solubles) was reported to be 49.3% in lucerne (Banerjee1998) compared to 51.6% in White Rose (H226) and 57.3% in Mulluvaadi (MVD-1) varieties of tapioca leaves. The hemicellulose and lignin values of White Rose (H226) and Mulluvaadi (MVD-1)  leaves were similar to that in lucerne but the cellulose content was comparatively less in both varieties. A significantly (P<0.05) higher crude fibre and consequently significantly (P<0.05) lower nitrogen free extract was observed in the White Rose (H226) verity of tapioca leaves compared with Mulluvadi (MVD-1) variety. The higher crude fibre in White Rose (H226) was reflected in significantly (P<0.05) higher structural fibre (Neutral detergent fibre) in this variety over Mulluvadi (MVD-1). Though hemicellulose, cellulose and lignin content appeared to be higher in White Rose (H226) over Mulluvadi (MVD-1), the differences were not significant.

The observed values in both varieties of tapioca leaves for calcium, phosphorus and magnesium are in agreement with the reported values (Ravindran and Ravindran 1988; Gomez and Valdivieso 1984 and Ravindran 1993). The value for zinc in both varieties was lower and that for copper was higher than reported in the above cited references. Soil mineral status can influence the mineral status of the plants. It is possible that the differences would be due to the mineral profile of soil in the study area against that of values reported elsewhere. The ratio of calcium:phosphorus is not favourable in both varieties, and ranged from 9.12:1 in White Rose (H226) to 9.19:1 in Mulluvaadi (MVD-1). Hence it is advisable to include phosphorus supplements whenever tapioca leaves are fed.

Effect of conserving on the quality of tapioca leaves

Chemical composition of unprocessed compared to hay or silage made with tapioca leaves

The chemical composition of unprocessed leaves, hay and silage of White Rose (H226) and Mulluvaadi (MVD-1) variety of tapioca leaves was similar. Likewise the fibre fractions and mineral profile also did not show statistical variation within the same variety of unprocessed, hay and silage. Normally, changes in chemical composition are expected due to processing especially in crude protein, crude fibre and nitrogen free extract content. However, in the present study no such variations were noticed probably because the test materials were leaves and hence the shattering loss is not applicable. Though ensiling increases the nitrogen free extract content in silage proportionate to molasses supplementation, NFE was not significantly higher than unprocessed tapioca leaves of the respective varieties.

Physical characteristics of tapioca hay

The physical qualities, viz maturity, leafiness, colour, odour, presence of foreign material indicated that the hay of both the varieties was of good quality (Table 2).

Organolepticcharacteristics of ensiled tapioca leaves

The colour of the White Rose (H226) and Mulluvaadi (MVD-1) silage was greenish brown and greenish yellow, respectively (Table 2), while Chhay Ty et al (2001) reported that tapioca leaves silage as pale green to brown yellow at 2 months of storage. The smell was pleasant in both varieties of tapioca leaves and there was no mould growth. 

Biochemical characteristics of ensiled tapioca leaves

Pettersson (1988) categorized silage with DM of 250-350 g/kg and pH  below 4.5 as good and at pH of 4.5 to 4.8 as fair silage. The marginally higher pH value could be attributed to the reduced lactic acid production by microbes during ensiling (Thomas and Chamberlain 1982. Trace levels of butyric acid were found (Table 2), while Man and Wiktorsson (2001) reported that butyric acid was not found in tapioca leaf silage.  The higher Flieg index in Mulluvaadi (MVD-1)  silage could be contributed by the higher lactic acid content and lower butyric acid content  when compared to the White Rose (H226) silage.

Effect of wilting on HCN content of tapioca leaves

Several authors (Balasundaram et al 1976; Ravindran and Ravindran 1988; Ravindran et al 1987; Phuc et al 2000; Man and Wiktorsson  2001) reported values for HCN content of tapioca leaves ranging from 76 to 1881 mg/kg DM. Varietal difference could be a possible explanation for such difference. In addition to this, Ravindran (1993) reported that the HCN content varied due to the stage of leaf maturity and can also be influenced by the nutritional status of  the plant. Cyanide levels were reported tobe increased by fertilized nitrogen, whereas potassium and farmyard manure decreased HCN content. A different pattern of reduction in HCN was recorded in White Rose (H226) variety as against Mulluvaadi (MVD-1) variety. The HCN content in White Rose (H226) declined to about 1/3^rd of its HCN at 6 hours of wilting followed by 1/6^th level at 9 hours of wilting and thereafter declined at slow pace. On the other hand in Mulluvaadi (MVD‑1) variety the HCN level declined at steady pace without steep fall during early hours of wilting. It is quite possible that this difference could be due to variation in the quantity of linamerase present in the two varieties of tapioca leaves.

A similar significant reduction in HCN content was observed by Phuc et al (1996) on sun drying the tapioca leaves. Padmaja (1989) and Ravindran et al (1987) also postulated that wilting of tapioca leaves markedly reduced the HCN content. However, the HCN content in both the varieties did not fall below the safety level of HCN (50 mg/kg dry matter) as suggested by Phuc et al (2000) even after 24 hours of wilting. However under field condition, it has been observed that the tapioca leaves of these varieties that are wilted for 1-2 hours can be fed at the rate of about 1-2 kg of DM per head of cattle without any deleterious effect. It is quite possible that the rumen microbes in these animals might have developed the capacity to detoxify the HCN content as suggested by Majak et al (1990).

Good hay and good silage were prepared from tapioca leaves of both varieties. However, the HCN content was significantly lower (P<0.01) in Mulluvaadi (MVD‑1) than White Rose (H226). Such variation continued to exist even when they were made as hay (Table 1). However, ensiling narrowed the HCN content with Mulluvaadi (MVD-1) having (P<0.05) a significantly higher HCN level than White Rose (H226). This variable reduction in HCN effected by different preservation techniques between two varities of tapioca leaves needs in-depth analysis.

The HCN content of the silages [51.80 ± 2.89 mg/kg (White Rose (H226)  and 81.7 ± 9.83 mg/kg (Mulluvaadi (MVD-1)] are in agreement with that observed by Preston (1995) who reported values as low as 33 mg/kg dry DM in ensiled leaves. The variability in HCN reduction pattern between two verities of tapioca leaves could be attributed to variation in linamerase. Santana et al (2002) reported two different patterns of linamarase activity in two cassava  cultivars. Significant variability in the inhibition of linamarase enzyme activity between two cultivars of tapioca leaves and consequently variation in HCN content was reported by Nwosu and Onofeghara (1994). Furthermore, Limon (1991) and Nguyen Thi Loc et al (1996) reported that fermentation was an effective way of reducing the HCN content of tapioca. The fact that the HCN content was more rapidly reduced in White Rose (H226) than in Mulluvaadi (MVD-1) is suggestive of the need for shorter duration for wilting in White Rose (H226) over Mulluvaadi (MVD-1). HCN content of White Rose (H226) silage was significantly (P<0.01) lower than for Mulluvaadi (MVD-1), while the hay made from Mulluvaadi (MVD-1) had significantly (P<0.01) lower HCN content than White Rose (H226) hay, suggesting that White Rose (H226) is more suitable for ensiling and Mulluvaadi (MVD-1) for hay making.

Conclusions

• Hay making with leaves of both the varieties was adjudges as "good" based on their physical characteristics.

• Ensiling improved the products by reducing the HCN of the raw materials.

• The White Rose (H226) was more suitable for ensiling and Mulluvaadi (MVD-1) for hay making, according to their HCN characteristics.

References

Allen R D 1984 Feed stuffs ingredient analysis table. Feedstuffs:(USA) 25-30.

AOAC 1980 Official Methods of Analysis of Association of Analytical Chemists. 13^th Edition, Association of Official Analytical Chemists, Washington D.C.

Atomic Absorption Spectrophotometer 1994 Perkin Elmer model 3110 cook book. The Perkin - Elmer Corporation Norwalk, USA.

Balasundaram C S, Chandramani R, Muthuswamy P and Krishnamoorthy K K 1976. Distribution of Hydrocyanic acid in different fraction during the extraction of leaf protein from cassava leaves. The Indian Journal of Nutrition and Dietetics 13, 11-13.

Banerjee G C 1998 Green forages. In: Feeds and Principles of Animal Nutrition. Revised Edition, Oxford and IBH publishing company private limited, culcutta, India 15.

Barker S B and Summerson W H 1941 The colorimetric determination of lactic acid in biological material. Journal of Biological Chemistry 138, 535.

Chase L E 1990 Analysis of fatty acids by packed column gas chromatography. In: G.C., Bulletin 856, Division of Rohmand Hass, Suppelco, 1-12.

Chhay Ty, Ly J and Rodriguez L 2001 An approach to ensiling condition for preservation of cassava foliage in Cambodia. Livestock Research and Rural Development 13, 1-7. http://www.cipav.org.co/lrrd/lrrd10/3/sene103.htm

Goering H K and Van Soest P J 1970 Forage Fibre Analysis. Agriculture HandBook No.379. Agricultural Research Service United States Department of Agriculture Washington, D.C. 1-20.

Gomez G and Valdivieso M 1984 Cassava for animal feeding: Effect of variety and plant age on production of leaves and roots. Animal Feed Science and Technology 11, 49-55.

Hill D C 1973 Chronic cyanide toxicity in domestic animals. In: Chronic cassava toxicity. Proceedings of the Interdisciplanary Workshop, London, 105-111.

Khang D N and H Wiktomson 2000 Effect of cassava leaf meal on the Rumen environment of local yellow cattle fed urea - treated paddy straw. Asian-Australasian Journal of Animal Sciences, 1102-1108.

Kumar R 1992 Anti-nutritional factors, the potential risks of toxicity and methods to alleviate them. In: Legume trees and other fodder tree as protein sources for livestock. In: FAO Animal Production and Health Paper 102 (Editors A. Speedy and P.Pugliese.) FAO: Rome 145-160.

Limon R L 1991 Ensilage of cassava products and their use as animal feed. In: Roots, tuber, plantains and bananas in animal feeding FAO Animal Production and Health paper 95 (Editors D.Machin and Solvoig Nyvold) 99-109.

Majak W,  McDiarmid R E,  Hall J W and Cheng K J 1990 Factors that determine rates of cyanogenesis in bovine ruminal fluid in vitro. Journal of Animal Science 68, 1648-1655.

Man N V and Wiktorsson H 2001 Cassava tops ensiled with or without molasses as additive effects on quality, feed intake and digestibility by Heifer. Asian-Australasian Journal of Animal Sciences 14, 624-630.

Man N V and Wiktorsson H 2002 Effect of molasses on nutritional quality of cassava and Gliricidia tops silage. Asian - Australasian Journal of Animal Sciences 15, 1294-1299.

McDonald P,  Edwards R, Greenhalgh J F D and Morgan C A 1999 Grass and forage crops. In: Animal Nutrition. (Editor Addison Wesley Longman) Harlow, U.K. 434-450.

Nguyen Thi Loc, Olgle R B and Preston T R 1996 On-farm and on-station evaluation of cassava root silage for fattening pigs in Central Vietnam; M.Sc. Thesis submitted to Swedish University of Agricultural Sciences.

NwosuL A and Onofeghara F A 1994 A comparison of cyanide accumulation, leaf retention and linamarase activity of cassava varieties during water stress. Acta Hort. (ISHS) 380, 187

Padmaja G 1989 Evaluation of techniques to reuse assayable tannin and cyanide in cassava leaves. Journal of Agriculture and Food Chemistry 37, 712-716.

Pettersson K 1988 Ensiling of forages: Factors affecting silage fermentation and quality; Dissertation submitted to Swedish University of Agricultural Sciences.

Phuc B H N, Ogle R B and Lindberg J E 2000 Effect of replacing soya bean protein with cassava leaf protein in cassava root meal based diets for growing pigs on digestibility and Nitrogen retention. Animal Feed Science and Technology 83, 223-235.

Phuc B H N, Ogle R B,  Lindberg J E and Preston T R 1996 The nutritive value of sundried and ensiled cassava leaves for growing pigs. Livestock Research and Rural Development (8) 489-499.

PrestonT R 1995 Tropical Animal feeding: A manual of research workers. FAO Animal Production and Health paper 126. FAO: Rome 72.

Ravindran G and Ravindran V 1988 Challenges in the nutritional composition of cassava {Manihot esculenta Crantz} leaves during maturity. Food chemistry 27, 299‑309.

Ravindran V 1993 Cassava Leaves as Animal Feed: Potential and Limitation. Journal of Science of Food and Agriculture 61, 141-150.

Ravindran V and Rajaguru A S B 1988 Effect of stem pruning on cassava root yield and leaf growth. Srilankan Journal of Agricultural Science 25, 32-37.

Ravindran V, Kornegay E T,  Rajaguru S B and Notter D R 1987 Cassava leaf meal as a replacement for coconut oil meal in pig diet. Journal of Science of Food and Agriculture 41, 45-53.

Reed J D, McDowell R E, Van Soest P J and Horvath P J 1982 Condensed tannins: A factor limiting the use of cassava forage. Journal of Science of Food and Agriculture 33, 213-220.

Sankaravinayagam V  B,  Ravi R and Purushothaman M R 1999 Tapioca leaf meal for egg-type chicks. Indian Journal of Animal Sciences 69, 641-642.

Santana M A,  Vásquez V,  Matehus J and  Aldao R R  2002 Linamarase Expression in Cassava Cultivars with Roots of Low and High-Cyanide Content. Plant Physiology, 129, 1686-1694.

SnedecorG W and Cochran W G 1967 Statistical methods. 6^th Edition, Oxford and IBH Publishing Company private limited, Calcutta, India.

Thomas P C and Chamberlain D G 1982 Silage as a feedstuff in silage for milk production. (Editors J.A.F.Rook and P.C.Thomas) Technical Bulletin - 2 England.

VoughL R 2002 Evaluating hay quality in Fact Sheet 644. Department of Agronomy. University of Maryland at College park, U.S.A. 1-8.

WeatherburnM W 1967 Phenol - Hypochlorite reaction for determination of ammonia. Analytical Chemistry 39, 971-973.

Wilson R F and Wilkins R J 1972  An evaluation of laboratory ensiling techniques. Journal of Science of Food and Agriculture 23, 377-385.

Zimmer E 1966 Die nuufrassung das Gärfutter schlussels nach flieg. Wirtschaftseigene futer, 12:299-303.

Received 28 October 2005; Accepted 28 February 2006; Published 22 March 2006

Go to top