Generation of a Safer Food Product for Subsistence Farmers
August, 2000

Many wild and domesticated crops produce secondary compounds that effectively reduce or deter herbivory by insects and animals. Some of these herbivore deterrents are relatively non-toxic; however, some herbivore deterrents are lethal if ingested. One of the more toxic classes of herbivore deterrent compounds is the cyanogenic glycosides. The cyanogenic glycosides are a group of nitrile-containing plant secondary compounds that yield cyanide following their enzymatic breakdown (cyanogenesis). It is estimated that between 3,000 and 12,000 plant species produce cyanogenic glycosides. Many important crops are cyanogenic, including sorghum, almond, lima beans (non-domesticated), and white clover. The most agronomically important of the cyanogenic crops is the tropical root crop cassava (Manihot esculenta, Crantz). Over 153 million tons of cassava root is harvested annually in the tropics, primarily for human consumption.

Cassava has many agronomic features that make it an ideal crop for cultivation on marginal lands in the tropics. It is drought tolerant, grows on nutritionally poor soils, and it produces large yields of starch. In addition, the presence of cyanogens in cassava has been shown to reduce herbivory. Recent studies also suggest that subsistence farmers in sub-Saharan Africa may rely on the toxicity of cyanogens to protect their crop from theft.

Various human health disorders have been linked to the consumption of poorly processed cassava containing residual cyanogens. Chronic cyanide exposure has been associated with the occurrence of several health disorders including hyperthyroidism and tropical ataxic neuropathy, particularly in sub-Saharan Africa. In addition, the consumption of unprocessed and highly cyanogenic varieties of cassava may cause permanent paralysis of the legs (konzo). The incidence of konzo is most frequent during crop failures or when less time is taken to process or remove cyanogens from cassava. Significantly, cyanide poisoning due to consumption of poorly processed cassava is exacerbated by insufficient consumption of the sulfur-containing amino acids, cysteine and methionine. The sulfur-containing amino acids are required for the detoxification of cyanide in humans by the enzymes rhodanese and/or ß-cyanoalanine synthase. Significantly, the conditions (e.g., drought) that may lead to the use of short-cut cassava processing practices often go hand-in-hand with reduced availability of protein (methionine or cysteine) in the diet.

Recently, the biochemistry and physiology of cyanogenesis in cassava has been elucidated. The generation of cyanide from linamarin (cyanogenesis) begins when the plant tissue is damaged. Rupture of the plant cell vacuole releases linamarin. Subsequently, linamarin is hydrolyzed by a cell wall ß-glycosidase known as linamarase. The products of linamarin hydrolysis are acetone cyanohydrin and glucose. Significantly, acetone cyanohydrin will spontaneously decompose to cyanide and acetone at pHs >5.0 or temperatures >35 °C. Acetone cyanohydrin also is broken down by the cassava enzyme hydroxynitrile lyase (HNL).

Until recently, it had been assumed that the only cyanogen present in processed cassava foods was linamarin. Several studies have shown that some portion of ingested linamarin is de-glycosylated in the body where cyanide is presumably produced from the spontaneous decomposition of acetone cyanohydrin. In 1992, however, Dr. Hans Rossling and colleagues demonstrated that the major cyanogen present in poorly processed cassava roots was not linamarin but acetone cyanohydrin. This result was quite unexpected since it was assumed that acetone cyanohydrin would be eliminated from foods by spontaneous (high pH and/or temperature) or enzymatic breakdown. Significantly, free cyanide has not been found in cassava foods, presumably due to its volatilization. To account for the presence of acetone cyanohydrin in cassava foods, it was hypothesized that the low pH conditions used during cassava processing (soaking and fermentation) reduced the rate of spontaneous acetone cyanohydrin decomposition. This hypothesis, however, did not address the issue of acetone cyanohydrin turnover catalyzed by HNL.

In 1996, we characterized the abundance, distribution, and kinetic properties of HNL in cassava roots, stems, and leaves. Our objective was to determine the biochemical basis for the unexpectedly high levels of acetone cyanohydrin in processed cassava food products. Significantly, we discovered that HNL was expressed and present only in leaves. Thus, it was apparent that the high acetone cyanohydrin levels present in processed cassava roots could be attributed to the absence of HNL in roots. This observation lead to the development of transgenic strategies to facilitate cyanogen elimination from processed cassava foods. A transgenic approach for expressing HNL in roots also was predicated on the fact that cassava produces few seeds and has a long life-cycle (one year seed-to-seed). Thus it would be difficult to use traditional breeding and selection strategies to identify plants expressing HNL in roots. Furthermore, cassava typically is propagated clonally as stem cuttings in the field.

Our strategy for reducing the cyanide toxicity of cassava food products was to express HNL in the roots of transgenic cassava. Using an Agrobacterium mediated transformation system we introduced a cassava cDNA encoding HNL into somatic embryos. Transgenic cassava plants were regenerated and verified by PCR screening for the integrated HNL cDNA. A total of eight transgenic plants have been obtained to date. The HNL activity of root extracts obtained from in vitro transgenic plants has been measured. HNL activities in transgenic plants range from 0 - 0.56 mmol CN/mg protein/hr. Significantly, the HNL activity of crude leaf extracts is 0.6 mmol CN/mg protein/hr. Thus, HNL activity rates comparable to those in leaves could be obtained in the roots of transgenic plants.

Currently, the transgenic plants are being grown under greenhouse conditions to produce full size roots. It is expected that the roots from transgenic plants expressing HNL will have reduced levels of acetone cyanide or cyanogenic potential relative to wild type plants following processing. Follow-up field trials will determine whether the expression of HNL in roots in fact reduces the cyanide toxicity of cassava food products. It is expected that the HNL-dependent acceleration and increased efficiency of cyanide removal from cassava roots will reduce the level of chronic cyanide exposure from cassava food products and facilitate the broader acceptance of cassava as a safe food product for the consumer.

Richard T. Sayre
Department of Plant Biology
Ohio State University