PREFACE

The Working Group on Cassava Safety (WOCAS) was formed under the umbrella of the International Society for Tropical Root Crops (ISTRC), during a meeting of the society's Africa Branch in Kampala, Uganda in 1992. The aims of WOCAS are to provide recommendations for the promotion of safe cassava based on current knowledge, to identify research needs and develop research strategies, to identify people working in this field and facilitate exchange of information and experiences. The first activity of WOCAS was the organization of this workshop, and these proceedings represent the first output.

Considering that reported toxic effects of cassava are relatively rare in relation to its wide use as a staple, we decided that cassava safety was a better working concept than cassava toxicity. Hence the name of the working group and of the workshop.

The workshop aimed to take stock of the present state of knowledge on safety issues related to cyanogenesis in cassava and to disseminate this information more widely among researchers in the field. It was organized around seven main themes: Biology of cyanogenesis; Analytical methods; Agronomic research; Cassava processing and cyanogen removal; Livestock feeds; Human health and nutrition; Socio-economic aspects. These themes are reflected in the chapters of these proceedings. Invited leading researchers in these fields prepared discussion papers for sessions on the theme relevant to their expertise. In each session, the discussion papers were presented to and discussed by all the workshop participants. The session chairperson and speakers were asked to meet in the evenings and to summarize the major findings of their session and to formulate any consensus and recommendations that emerged during the debates. Summary and recommendations from all sessions were debated in a final plenary session. We have tried to capture the essence of the discussions and to incorporate them in the draft recommendations emanating from the seven sessions. The section "Summary and Recommendations" reflects these, including any controversial issues as they were presented. It was notable how effectively researchers from different disciplines interacted. We believe that this interaction provided for a comprehensive review of the subject area.

Since the last workshop on health aspects from cassava cyanogenesis in 1982, many advances have been made in the understanding of cyanogen removal during processing, in improving analytical methods; in understanding the causal relationship between cassava cyanogenesis and human diseases, especially the understanding of the factors underlying toxic effects; in elucidating the genetic basis of the synthesis of cyanogenic glucosides; and in understanding the socio-economic mechanisms influencing cassava production.

These advances have established that the control of cyanogenesis in cassava should be approached by both genetic improvement and promotion of effective processing. The latter has been recognized as the most efficient way of controlling cyanogenesis in the short term. The dynamics of cyanogen removal from cassava and the factors controlling this dynamics have been elucidated. It can now be explained why apparently small modifications in length or sequence of steps in processing cassava roots can lead to large differences in the levels of residual cyanogens. These advances were made possible by the development of new methods for the determination of cyanogenic glucosides, cyanohydrins and hydrogen cyanide in cassava products. Advances in the molecular biology of cyanogenesis combined with conventional plant breeding now make it possible to develop powerful approaches to optimize the levels and distribution of cyanogenic glucosides in cassava. A gronomic research has shown that environmental factors can be as important as genetic factors in determining the levels of cyanogenic glucosides in cassava roots. The understanding of causal relationships between cassava cyanogenesis and associated human diseases has improved. Of special importance was the identification of similar underlying causes to the reported outbreaks of paralytic diseases and acute poisoning attributed to cyanogens in cassava. These outbreaks occurred in socio-economically deprived communities that relied on cassava for food security and which, due to food shortage, war or poverty, made short-cuts in their traditional processing methods. New socio-economic findings emphasized the importance of cassava processing, not only for the sake of safety, but also for expanding cassava production by improving shelf life, facilitating transport and introducing consumer-specific taste and texture of cassava products.

Several topics discussed at the workshop could not be settled and require further study. The reason for the use of bitter and toxic cassava varieties in communities where the risk of intoxications is great remain unclear. The levels of cyanogenic glucosides in fresh cassava roots currently used by plant breeders as target for developing cassava genotypes with low cyanogenic potential are not in agreement with the understanding of safety limits for cyanogens in cassava. An approach to be followed to establish safe levels of cyanogenic glucosides is proposed (see Rosling, this volume) but requires to be validated. WOCAS will undertake this task.

It has been hypothesized that long term exposure to sub-clinical amounts of cyanogens from cassava based diets may influence human biological fitness and microevolution (see Jackson, this volume). Evidence to support or reject this hypothesis is currently limited.

The relationship between bitterness of fresh cassava roots and their total cyanogen content needs further clarification. Although the correlation coefficient between the two is high, there is need to establish whether there is a cause-effect relationship.

Although some progress has been made on the understanding of the role of cyanogenic glucosides in resistance to pests, an irrevocable proof has not been obtained. The development of acyanogenic varieties by genetic engineering techniques that enable the silencing of gene(s) coding only for the biosynthesis of cyanogenic glucosides may provide such a proof by demonstrating that a variety that was otherwise resistant to pests becomes sensitive when it no longer does produce cyanogenic glucosides.

The terminology used in the scientific literature when reporting the concentration of various cyanogenic compounds found in cassava is very diverse, often confusing, and sometimes misleading. An agreement could not be reached during the meeting. Afterwards, advice was sought from the International Union of Pure and Applied Chemistry (IUPAC). Based on IUPAC's suggestions and current state of knowledge, and to foster a better understanding of safety issues in cassava, the following recommendations on terminology are being made. Firstly, it should be recognized that intact and fresh cassava tissues mainly contain the cyanogenic glucosides linamarin and lotaustralin. Processed or damaged tissues may contain varying amounts of cyanogenic glucosides, cyanohydrins and hydrogen cyanide. The recommended analytical procedure (see Essers, this volume) can determine the total amount of all three compounds (Fraction A), the total amount of cyanohydrins and hydrogen cyanide (Fraction B) or the amount of hydrogen cyanide (Fraction C). Fraction A should be referred to as "total cyanogen content", fraction B as "non-glucosidic cyanogen content" and Fraction C as "hydrogen cyanide content". The "cyanogenic glucoside content" is obtained by subtracting fraction B from fraction A, while the "cyanohydrin content" is obtained by subtracting fraction C from fraction B. The recommended unit to be used is "mg HCN equivalent kg^-1" and has been used throughout this volume. Authors should indicate whether their data are calculated on fresh or dry matter basis. The potential for a sample to produce HCN, expressed as total amount of HCN equivalent per weight of sample has been called HCN-potential, HCN-releasing potential, cyanide potential, or cyanogenic potential. The latter is preferred. Abbreviations such as HCNp, CNp, or CNP are discouraged.

Our aim has been to publish these proceedings rapidly. This has been possibly due to the goodwill of authors, most of whom were able to rapidly return their proofs. We are grateful to Professor Eric E. Conn for letting us publish his paper although he was not able to participate in the workshop. We gratefully acknowledge the literary editing done by Ms. Sue Scott-Paul on part of the manuscript and the secretarial assistance of Ms. Labake Sadiq.

This volume is, therefore, an update of research on cyanogenesis in cassava and its implications. We intend to address the definition of safe levels of cyanogenic glucosides in fresh and processed cassava products in a subsequent publication. We also plan to develop and publish guidelines and a manual to be used by development agents involved in the promotion of cassava processing. There are also plans to address other safety aspects of cassava, including mycotoxins. We would be interested to receive from our readership comments on this volume and on research and development related to cassava safety.

Wageningen, 30 August 1994.
Working Group on Cassava Safety:
Mpoko Bokanga
Sander Essers
Nigel Poulter
Hans Rosling
Olumide Tewe


SUMMARY AND RECOMMENDATIONS

1. Biology of cyanogenesis
   

   1.1 Conventional plant breeding for cassava cultivars (with low or high cyanogenic potential) that satisfy user needs and preferences in diverse socioeconomic and agroecological conditions, can be effectively complemented with biotechnological approaches. Current knowledge on synthesis, degradation, transport, and regulation of cyanogenic glucosides in cassava provides the possibility of developing new approaches and tools for optimizing the content and distribution of cyanogenic glucosides and associated enzymes in cassava.

   1.2 Three genes coding for key enzymes controlling the biosynthesis and degradation of cyanogenic glucosides have already been isolated and cloned. These are the gene for cytochrome P450, the enzyme catalyzing the rate-limiting conversion of the parent amino acid to the corresponding oxime in the biosynthetic pathway, the gene for linamarase, which catalyzes the degradation of linamarin to acetone cyanohydrin, and the gene for hydroxynitrile lyase, which catalyzes the degradation of acetone cyanohydrin to hydrogen cyanide and acetone. Important genes which have not yet been isolated include those coding for glucosyltransferase which converts linamarin to its transport form linustatin, simultaneous diglucosidase, which would split linustatin in the first step of its metabolism to a non-cyanogenic compound, and root- and leaf-specific promoters effective in cassava.

   1.3 An efficient transformation system for cassava is urgently needed in order to introduce relevant genes already available and those to be developed. Progress in cassava transformation and regeneration (not reviewed at this meeting), appears likely to permit transgenic cassava plants, using currently- and shortly-available genes, to be ready for controlled testing within two to three years.

   1.4 The presently observed demand among cassava farmers, processors, and consumers for cassava cultivars with a high content of cyanogenic glucosides and/or with bitter taste may reflect a tight genetic coupling of the cyanogenic character and bitterness to other, beneficial characteristics. In spite of this tight coupling, these associations may be broken by continued traditional breeding using better analytical methods (e.g., antibodies and cDNA probes) enabling the desired varieties to be selected with a higher degree of accuracy.

   1.5 The control of cyanogenesis could be achieved in several ways:

   1. by transforming cassava with the introduction of an anti-sense construct of cytochrome P450 under the control of a strong constitutive promoter to produce a-cyanogenic plants,
   2. by inserting tissue-specific promoters or developmentally-controlled promoters in front of the cytochrome P450 gene to limit production of linamarin to certain tissues and at specific periods of plant growth,
   3. by introducing a strong promoter in front of the linamarase gene to increase the breakdown of linamarin during processing,
   4. by preventing linamarin conversion to the transport metabolite linustatine by increasing the conversion of linustatin to asparagine instead of its conversion back to linamarin. This last approach may result in the accumulation of protein nitrogen in the roots.

      The plants so obtained would constitute an ideal research material for specifically testing the relationship, if any, between cyanogenic glucoside content and desired properties such as starch quality, insect resistance, bitterness and others.

   1.6 If any models or experimental approaches prove to be useful, plant breeding can be used to transfer the desired cyanogen metabolism into appropriate varieties. When genetic transformation of cassava has been extended to a wider range of genotypes, it may be possible to confer the desired cyanogenesis phenotype on varieties improved for quantitative traits (i.e. for traits more difficult to define genetically), by using transformation of elite selections. This option would relieve the plant breeder of a set of selection objectives for cyanogenesis and thereby permit faster progress for quantitative traits such as yield and environmental adaptaion.

   1.7 Molecular biology research may produce experimental data within five years to answer some of the questions that cannot be otherwise answered. However, it is not expected to produce results transferable to cassava farmers in the next ten years. It is therefore important that these efforts are combined with continued and enlarged efforts to expand our knowledge of effective and practical processing techniques to reduce cyanogen levels in cassava products.

   1.8 In the long term, molecular biology can offer more than mechanisms for removal of cyanogenic glucosides. Combined with plant breeding and other disciplines, it provides a new and potentially powerful approach. It can circumvent the loss of any desirable functions that may be found to be associated with cyanogens by the introduction of nutritionally less problematic factors, including plant protective agents and quality factors.

   1.9 Similarly to the large resources being spent for research on temperature crops by industrialized countries, it is imperative to strengthen basic research on cassava with respect to nutritive value, productivity, tolerance to biotic and abiotic stresses, etc., if this crop is to provide an economic as well as food security resource to tropical countries.

2. Analytical methods

   2.1 There is a wide variation in the levels of cyanogenic glucosides and linamarase activity within the same tissue of a cassava plant. In addition, variations occur between varieties, and also between different plants of the same variety in apparently similar environments, and between plants from the same variety in different environments. Sampling procedures are therefore critical for statistically valid analysis of results, no matter which chemical methods are used. There is a need to standardize sampling protocols, depending on the kind of the material and the purpose of the measurements. Investigators should be aware that handling and storage of the fresh and processed cassava collected for analysis, as well as extracts obtained from these, must be standardized and validated since considerable losses may occur. Simple and mobile equipment for homogenization and extraction needs to be developed.

   2.2 Linamarin and to a lesser extent lotaustralin are the cyanogenic glucosides found in the cassava plant. It is only when the tissue is damaged, mainly by mechanical or microbial action, that the cyanogenic glucosides can decompose to cyanohydrins that may further hydrolyze to hydrogen cyanide. Therefore, at harvest intact cassava tissues only contain cyanogenic glucosides and no hydrogen cyanide. However, processed products may contain varying amounts of cyanogenic glucosides, cyanohydrins and hydrogen cyanide. The development of analytical methods for separate determination of these three types of cyanogenic compounds have advanced the understanding of the dynamics of cyanogen removal during processing. Simplistic reference to cyanide or total cyanide content in cassava continues to hamper this understanding and should be avoided.

   2.3 Different chemical assay methods are needed since no single technique serves all requirements. The method should be chosen depending on resources available and objectives of the analysis. Many laboratories in developing countries with limited resources require robust, low-cost and simple methodologies. Simple, specific and relatively sensitive methods for use in field surveys are much needed. Of importance for both qualitative and quantitative techniques is reproducibility. Particularly important in enzyme mediated methods is the standardization of pH and thus the use of effective buffer systems.

   2.4 Autolytic methods (that rely on endogenous enzyme for glucoside hydrolysis) are unreliable for processed products in which the endogenous enzyme may have been inactivated. The use of exogenous linamarase is recommended but is currently constrained by high cost. One alternative is to produce crude linamarase preparations from cassava leaves or root cortex. Alternative sources of low cost but effective enzymes should be explored. The immobilization of linamarase to allow its repeated use provides an alternative approach.

   2.5 HPLC techniques allow the separate measurement of cyanogenic glucosides and cyanohydrins, but are costly and complicated. Potentiometric methods using cyanide electrodes have limited sensitivity and reproducibility.

   2.6 The safe handling of reagents used in the estimation of cyanogens needs to be taken into consideration. The pyrazalone/pyridine reagent used as a color reagent is a highly toxic and volatile reagent. Its use requires appropriate safety equipment which may not be readily available in developing countries. The reagent has to be prepared daily. The use of isonicotinate and 1,3-dimethyl barbiturate as a color reagent is a better alternative. Picrate and tetrabase can be used in qualitative and semi-quantitative assays although their potential health hazard should be investigated should be taken into consideration.

   2.7 Interfering compounds occur in samples rich in oils, fats, proteins and phenolic compounds that provide significant problems in extraction, recovery and colorimetric estimation. The importance of estimations of the prepared foods consumed by people (which may contain high levels of added oils and proteins makes it necessary to solve this problem.

3. Agronomic research

   3.1 Cassava varieties show a very wide range of cyanogenic glucosides levels in the storage roots. New findings in the health and food sciences now calls for a revision of the safe levels established more than 40 years ago.

   3.2 Cyanogenic potential is a heritable trait. The polygenic and recessive nature of its inheritance, coupled with inadequate sampling strategies, has slowed progress in conventional breeding this trait. Recent improvements in sampling methodology will facilitate more rapid progress. There is a very large environmental influence on the expression of cyanogenic potential in cassava. A thorough review of existing knowledge and a focus on identifying genotypes with greater stability over environments and at key developmental stages is essential. The contribution of field cultural practices in modulating cyanogenesis needs to be addressed.

   3.4 There is preference for bitter or potentially toxic varieties in some cassava growing communities. Observational studies indicate that bitterness or toxicity may play an important role in the prevention of damage to the crop by animals that feed on the roots. Current evidence suggest good correlation between cyanogenic potential and bitterness. However, some varieties with extreme expression of either trait do not follow the general trend. The compound responsible for bitterness needs to be identified and the reasons for preference of bitter or toxic varieties in some farming communities needs to be established. The results of these studies will determine the value of developing varieties with a low cyanogenic potential and a bitter taste should the bitter taste turn out to be the factor deterring animals from feeding on the crop.

   3.5 It remains to be established whether acyanogenesis is a viable option for addressing cassava safety. A recent study has suggested a role in resistance to pests for cyanogenic glucosides. However, further investigations should be conducted to confirm this role. The possibility of partitioning the cyanogenic glucosides into non-edible plant parts, whilst maintaining pest resistance, needs to be explored.

   3.6 Available data on relationships between cyanogenic potential and morphological/ agronomic traits show many inconsistencies. Molecular markers to be used for detecting genotypes for low or high cyanogenic potential are needed for accelerating breeding efforts to control cyanogenesis.

4. Cassava processing

   4.1 Cassava processing can reduce the cyanogenic content of roots and leaves of even the most potentially toxic varieties to safe levels. However, there is a myriad of processing methods, but these are not equally effective in achieving cyanogen reduction. The effectiveness of several processing techniques need to be verified for different cassava cultivars.

   4.2 Most of the principles of cyanogen removal during processing are now understood. Our current knowledge indicate that plant cells contain cyanogenic glucosides, mainly linamarin. Disintegration of the cells brings linamarin into contact with the endogenous enzyme linamarase, resulting in hydrolysis of linamarin into glucose and acetone cyanohydrin. Acetone cyanohydrin breaks down into acetone and hydrogen cyanide by the action of the enzyme hydroxynitrile-lyase or spontaneously at increased rates at higher pH's. The latter pathway appears to be the principal pathway. The volatile HCN (boiling point 25.7°C) can escape into the air.

   4.3 Effective cyanogen reduction is achieved in two steps. In the first step, the hydrolysis of glucosides is facilitated through disintegration of cells which is achieved by grating, crushing, microbial fermentation, enzymic action or a combination of these. In the second step, high pH facilitate the spontaneous breakdown of cyanohydrin. Higher temperatures achieved through heating and reducing moisture contents during drying also influence this effect. The factors determining cyanohydrin stability require further clarification. will facilitate more rapid progress. There is a very large environmental influence on the expression of cyanogenic potential in cassava. A thorough review of existing knowledge and a focus on identifying genotypes with greater stability over environments and at key developmental stages is essential. The contribution of field cultural practices in modulating cyanogenesis needs to be addressed.

   4.4 Processing methods which involve effective disintegration followed by heating or drying results in the highest removal of cyanogens. Examples of these methods include mechanical grating followed by roasting in the production of gari and farinha, and microbial fermentation followed by drying or steaming as in the production of lafun and chickwangue. Incomplete disintegration will result in residual cyanogens, particularly linamarin, and incomplete drying or heating may result in residual cyanohydrin. Whether in some cultivars, reduced linamarase activity may be a limiting factor in the removal of cyanogens is not known. Similarly, the role played by hydroxynitrile lyase requires further elucidation.

   4.5 Direct sun-drying of whole fresh roots only achieves partial removal of glucosides. Slower drying extends the effect of linamarase activity but simultaneously allow microbial growth. Chipping of fresh roos involving extensinve mechanical tissue damage will facilitate glucoside breakdown. But slicing of roots with minimal tissue damage followed by rapid drying will result in a high retention of glucosides. In sun-dried cassava pieces, an inverse relationship seems to exist between cyanogen and microbial content. Possibilities for optimizing cyanogen removal while minimizing microbial contamination during sun-drying should be further explored. The issue of mycotoxin contamination in sun-dried cassava, as well as in cassava molded on purpose needs to be addressed. Studies have identified mycotoxins in some sun-dried root products, but so far not in purposely molded cassava.

   4.6 Ineffective processing methods found in some communities can lead to cassava products with residual cyanogen levels. Appropriate improvements to these methods may reduce levels of within safe limits. It is recognised that in communities where intoxications due to cassava consumption occur, this minimalist intervention represents a powerful approach to solving the problem.

   4.7 As commercial cassava processing intensifies and the scale of cassava operations increase, safety issues become of greater critical importance. For example, inhalation of hydrogen cyanide vapors from roasting cassava and other occupational hazards should be minimized by good ventilations and accident prevention. Disposal of effluents from cassava processing are expected to become an increasing problem. More problems may arise from high biological oxygen demands (BOD) due to high solid content in the effluents rather than from cyanogens.

   4.9 There is a lack of detailed knowledge about why people process cassava the way thet do and the factors leading to this. Recommendations regarding processing should therefore take into account the quality characteristics of the raw materials and the end products as they relate to the wider socio-economic and cultural environment. The relationship between sensory characteristics, namely bitterness, and cyanogenic content as thet relate to cassava processing needs further elucidation.

5. Cassava in livestock productio

   5.1 Effective processing techniques for removal of cyanogens exist for preparation of dried cassava chips for animal feeding. A total cyanogen level of less than 100 mg HCN equivalent/kg dry cassava for inclusion in balanced compounded animal feed is economically acceptable in intensive livestock production systems.

   5.2 It has been established that cyanogens in feeds can increase the requirement for sulfur compounds, iodine, zinc, copper and selenium. Optimal levels of these compounds per unit cyanogen need to be determined for various livestock species.

   5.3 Sporadic deaths attributed to cyanogens in cassava have been reported in livestock production systems. Cassava roots, leaves and wates are often used components of livestock feed in rural farming communities. It is necessary to substantiate these claims and to develop safe handling stategies incorporate cassava into livestock feed, particularly in smallholder systems.

   5.4 Problems of cassava toxicity in livestock appear to be also due to contaminations as a result of poor handling and humid climate. Efforts to the safety of cassava-based feed, should also address its microbial quality.

6. Human health and nutrition

   6.1 Hydrogen cyanide is rapidly lost during processing and it probably does constitute the main source of dietary cyanide exposure from processed cassava. The main sources appear to be residual linamarin and cyanohydrin that are broken down in varying degrees to cyanide in the body. A substantial proportion of ingested linamarin is absorbed from the gut excreted unchanged in the urine. Thus, the dietary cyanide exposure can considerably lower than that expected from the total amount of cyanogens ingested Cyanide release from linamarin in the gut may depend on whether active are in ingested from cassava or foods simultaneously consumed provides. presence in the gut of active B-glucosidases from cassava, other foods, or from microflora, will influence the release of cyanide from ingested linamarin.

   6.2 It is reasonable to conclude, although there are few published reports, that cyanide exposure from insufficiently processes cassava can cause poisonings. Such acute poisonings occur when food shortage and social induce shortcuts in established processing methods. They may also occur cassava varieties with high glucoside levels are rapidly introduced into lacking efficient processng methods. Hospitals expected to receive such should be provided with rapid analytical methods and cyanide anti-dotes. This save patients, verify the cause of the intoxication and thereby avoid unnecessary sensationalism. The importance of gari in West Africa and the attribution of acute poisonings to short-cuts in gari processing in Nigeria provide a strong justification to study whether short-cuts in gari processing can yield products with lethal cyanogen levels.

   6.3 It is well established that the thiocyanate load resulting from dietary cyanide exposure from cassava can aggravate iodine deficiency disorders (IDD), expressed mainly as goiter and cretinism. This aggravating effect only occurs in populations with low iodine intake, and it is of secondary importance to the global IDD problem. Supplementation and fortification with iodine now receives high international priority and can counteract the effect of thiocyanate from cassava on the thyroid gland.

   6.4 The evidence is strong, although not conclusive, for a causal role of cyanide exposure from cassava in the paralytic diseases konzo and tropical ataxic neuropatthy (TAN). The pathogenic mechanisms are unknown. These diseases only occur in populations with severe socioeconomic problems, monotonous diet and food insecurity. The acute onset of konzo is attibuted to several weeks of high cyanide exposure due to shortcuts in cassava processing and concomitant low protein intake that reduces the rate of cyanide to thiocyanate conversion. The gradual onset of TAN is linked to several years of moderate cyanide exposure combined with low intake of protein and some B-vitamins.

   6.5 The proposed association between dietary cyanide exposure and malnutrition related diabetes as well as tropical pancreatitis remain speculative as no epidemiological data support a casual role of cyanide exposure. The suggested aggravating role of cyanide exposure from cassava in protien-energy malnutrition still lacks supporting data.

   6.6 Animal models can further elucidate the mechanisms involved and clarify the casual factors of the diseases associated with cyanide exposure from cassava. Long-term follow-up studies of population known to have had high dietary cyanide exposure in combination with various dietary deficiencies can provide new information on safe cyanogen levels in cassava products. The cyanogenic potential of cassava cultivars, together with residual levels of cyanogenic compounds in cassava poducts, linamarin intake and cyanide exposure should be studied in cassava eating communities where no related diseases are found. Such studies will advance the understanding of safety limits for cyanogens in the diet.

   6.7 Future studies are facilitated by recent development of new sensitive, specific and rapid analytical methods for blood or urine levels of linamarin, cyanide, thiocyanate and the alternative cyanide metabolites amino-thiazoline-carboxylic acid and cyanate.

   6.8 Due to sustainable production on marginal land and in drought conditions, cassava is crucial for food security in areas where toxic effects are reported. Affected populations state that bitter and potentially toxic varieties provide the best food security. Given the constraintss to agriculture in such areas, these varieties may paradoxically have an overall positive effect on human survival. Prevention of toxicity should not be attempted through banning of certain varieties but through positive actions like introduction of new varieties and promotion of effective processing.

   6.9 The majority of the 400 million people that consume cassava daily is not at risk of the diseases described above. From a public health perspective, the linkages between cassava and these toxico-nutritional diseases are similar to the linkages between monotomous rice and maize diets, and the nutritional diseases beri-beri and pellagra, respectively. The main reason for public health concern regarding cassava-related disease is that the underlying causes, severe social instability, agro-ecological crisis, and food insecurity, are becoming more common in parts of Africa.

   6.10 Human disease linked to cassava cyanogenesis are entirely preventable. Preventive actions include the promotion of effective processing, iodine supplementation and dietary improvements. The diseases can also be prevented by measures against underlying factors like food shortage, socio-economic deterioration and market distortions for cassava. Introduction of high yielding cassava varieties with low glucoside levels may be a long-term prevention in farming and food systems where cyanogenesis is not indispensable for food security. However, promotion of such varieties should only be done when proven to perform well under stress in the local farming system conditions.

   6.11 A cyanogen level of 300 mg HCN equivalent kg^-1 dry weight (10 mg/100g wet weight) has been used as upper limit for “low cyanide” in breeding programs since 1954. This levels is 30 times higher than the 10 mg HCN equivalent kg^-1, dry weight defined by FAO/WHO as safe level for cassava products in codex Alimentarius. A revision of these levels should be made based on a conceptual framework relying on new current knowledge from several disciplines. The estimations should be based on cyanide detoxification rates in humans, necessary safety margins for natural toxins, degree of cyanide release from ingested cyanogens, expected daily consumption and degree of cyanogen removal during processing. Theoretical levels should be compared with empirical measurements of the content of cyanogenic compounds in processed and fresh products consumed without effects by human populations according to general principles for natural substances in food.

7. Socio-economic considerations

   7.1 For the majority of cassava consumers, cyanide intoxication is not a concern. In some communities, particularly those facing nutritional deficiency and economic hardship, long term exposure to dietary cyanide from cassava has been reported to be an aggravating factor for diseases attributed to chronic cyanide intoxication. In situations of war, social distress, drought or economic instability, populations may be forced to survive on cassava as the sole food for extended period of time since cassava is usually the only food that remains. The shortage of food may lead to shortcuts in processing methods to obtain food more quickly. Such shortcuts result in high residual cyanogen levels which cause acute intoxications among consumers. In cases of cassava-related intoxications, intervention strategies should recognize social, cultural and economic peculiarities in order to find appropriate approaches and for effective implementation.

   7.2 Problems of cassava intoxications are linked to situation of economic depravation. Exploration and development of transnational and multiregional markets is a mechanism for enhancing local economies. The ability of cassava products to enter new markets will depend on product quality with respect to convenience, performance and safety. Building of rural infrastructure and amplifying trade relationships between and among various indigenous communities are part of the development of the cassava market.

   7.3 Cassava varietal dissemination is currently largely a local farmer-initiated event. Therefore, we need to better understand the various local rationale for farmer preference, in particular regarding adoption of new cultivars. The rate of introduction of germplasm of improved varieties with the desirable characteristics, as defined by local farmers, should be increased. Cassava cultivars with higher levels of cyanogenic glucosides than those already used should never be introduced without vigorous simultaneous promotion of appropriate processing methods. New varieties with very low levels of cyanogenic glucosides that hold promise of good performance in areas affected by toxic effects from cassava should be immediately introduced in those areas as a matter of priority. Varietal characteristics should be linked with particular processing methods.

   7.4 A broad diversity of ecologically, socioculturally, and technologically appropriate cassava processing techniques should be evaluated in a range of socio-economic and ecological settings and disseminated. Since many traditional forms of cassava processing are gender skewed (towards women and children), it is important that labor saving technologies should be promoted to reduce women's workload and increase productivity without compromising their access to income.

   7.5 The development and consumption of supplementary foods, both indigenous and introduced, in conjunction with various cassava food products should be promoted. We recommend the exploration of new uses of cassava to improve the economy of cassava growing communities.

   7.6 Cassava safety can best be improved by distribution of desirable varieties, promotion of effective processing techniques and diversification of markets for this root crop.
   

Distinguished participants and dear colleagues,

It is a real pleasure to welcome you to the International Institute of Tropical Agriculture (IITA) and to this International Workshop on Cassava Safety.

Cassava roots form a staple food for an estimated 500 million people in the tropics, and the leaves are commonly consumed as a vegetable in several areas. In Sub-Saharan Africa, cassava is currently the major staple food for 40% of the population. The presence of cyanogenic compounds in cassava and cassava products has for long cast some shadow over its use. Over time, more and more information and evidence have been gathered to show that cyanogens from cassava can play a role in certain health problems, while its relationship with other diseases is questioned or rejected.

Two international workshops conveyed by the International Development Research Centre (IDRC) in 1973 and in 1982 have focused on the goitrogenic effect of the cyanogens in cassava, and on neuropathies that were described in Nigeria. Since then, new concern on cassava-related food intoxications has been raised by the occurrence of several outbreaks of a paralytic disease named "konzo" in several parts of Africa, including Mozambique, Tanzania and Zaire. There is increasing evidence that konzo may be due to cyanogen intake from poorly processed cassava roots. In the mid-to-late eighties, numerous reports appeared in the Nigerian press alleging intoxications with sometimes fatal incidence from cassava consumption. However, a workshop convened by two leading Nigerian universities and held at IITA in December 1989 provided some answers but did not put the controversy to rest.

The possibility of intoxication from cassava is a matter of serious concern for IITA. The Institute has the mandate for the genetic improvement of cassava in Africa. Cassava is relatively new to Africa; its cultivation is continuing to expand to new areas and to populations formerly not familiar with the crop. In most cases, cassava cultivation is done by resource poor farmers on marginal lands, particularly during drought periods. Current evidence suggests that under these circumstances (poor soils, inadequate rainfall), the cyanogenic potential of cassava increases. Populations new to cassava cultivation may not be familiar with the toxic potential of cassava varieties they are using, may lack the appropriate processing technology for detoxifying cassava, and may therefore be at high risk of intoxication.

The problem of cassava toxicity is highly complex. It is one that mixes aspects of biochemistry, toxicology, food technology, anthropology and economics. It is my hope that this gathering of outstanding experts convened by the Working Group on Cassava Safety will succeed in clarifying the complex web of issues related to the safety of cassava consumption and will provide answers and recommendations that will guide researchers and users of cassava around the world.

Once again, I welcome you to IITA and I wish you very fruitful deliberations.


OPENING COMMENTS

Robert Asiedu

It is indeed an honor to have this opportunity to make A few remarks at the OPENING of this important meeting.

It is good to see that cassava, the mainstay of millions of people in the tropics and subtropics of the world, is at last getting some of the attention it deserves from the scientific community. It is only through unbiased scientific investigation that many of the myths surrounding this valuable crop will be dispelled. But even more important are the opportunities for expansion in utilization, and consequently production, of cassava throught the exchange of relevant and accurate information on its quality attributes and its versatility in utilization.

Most of the confusion and concern about the use of cassava in various food and feed products originates from the presence of cyanogenic glucosides in most tissues. The role of these glucosides in the plant, the genetic and environmental influences on their levels, the methodologies for estimating these levels, and the processes for the elimination of the glucosides from foods and feeds have all been subjects of some study and much controversy. The continuing controversies can partly be attributed to the absence so far of a coordinated approach to resolving the major issues, for instance through standardization of methodologies or agreement on terminologies. Questions remain on issues like the possibility of cyanide exposure for people engaged in cassava processing activities; the efficiency of field sampling procedures in the analysis of cyanogenic potential of cassava cultivars; the relative efficiencies for the available analytical methods applicable to field situations or to large numbers of samples; the mechanisms and efficiency of cyanogen removal from cassava during processing; the real or potential hazards of various forms of fresh cassava root consumption, and the problems and benefits of using cassava leaves as food; and farmers' needs and preferences.

The importance of safety in utilization of cassava was recognized many years ago at the International Institute of Tropical Agriculture (IITA). Screening for low cyanogenic potential started in 1973. The rapid but less accurate picrate leaf test which was used then was replaced in 1984 by a high-performance automated enzymatic assay. A sampling methodology designed to take into account the wide variability of cyanogenic potential in cassava has recently been introduced to improve the accuracy of the screening process. Our cassava germplasm base includes a broad range of materials from many African countries and also some recent introduction from Latin America through collaboration with the Centro Internacional de Agricultura Tropical (CIAT). Moreover, a serious effort in interspecific hybridization is intended to produce materials incorporating desirable levels cyanogenic glucosides and other useful traits. Out of this broad mix of materials and through cyclic selection and recombination of selected parents, we have succeeded in obtaining varieties with high and stable yields in several agro-ecological zones, with resistance to major pests and diseases and with culinary attributes desired by farmers, processors and consumers. The range of cyanogenic potential in those varieties extends from very low (below 20 mg HCN equivalent kg^-1, fresh weight basis) to very high (above 200 mg HCN equivalent kg^-1, fresh weight basis). This wide range allows us to offer to our collaborators in the national agricultural research systems and in non-governmental organizations a diversity of genetic materials from which they can select research findings to farmers is A task that is yet to be fully accomplished. farmers locate in remote areas where problems of intoxication have been fully reported, are in cases out of the reach of institutional extension services. We would be most grateful your comments on the relative efficiencies of our past and current approaches achievements during the course of this meeting.

Regardless of what we as scientists do, however, we should be aware that there millions of people who will continue to depend heavily on cassava as food and feed. For some of them, handling of cassava will be influenced by social, economic and circumstances that make it impossible or unlikely for them to handle cassava the way textbooks currently recommend. It is essential for us to appreciate the interplay biological and social sciences in addressing the issue of cassava safety.