- Email: [email protected]
The targets of acetone cyanohydrin neurotoxicity in the rat are not the ones expected in an animal model of konzo
Neurotoxicology and Teratology 32 (2010) 289–294
Contents lists available at ScienceDirect
Neurotoxicology and Teratology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / n e u t e r a
Brief communication
The targets of acetone cyanohydrin neurotoxicity in the rat are not the ones expected in an animal model of konzo☆ Carla Soler-Martín 1, Judith Riera 1, Ana Seoane, Blanca Cutillas, Santiago Ambrosio, Pere Boadas-Vaello, Jordi Llorens ⁎ Departament de Ciències Fisiològiques II, Universitat de Barcelona, 08907 Hospitalet de Llobregat, Spain
a r t i c l e
i n f o
Article history: Received 13 October 2009 Received in revised form 12 November 2009 Accepted 12 November 2009 Available online 20 November 2009 Keywords: Konzo Cyanide Manihot esculenta Nitriles Acetone cyanohydrin
a b s t r a c t Konzo is a neurotoxic motor disease caused by excess consumption of insufficiently processed cassava. Cassava contains the cyanogenic glucoside linamarin, but konzo does not present the known pathological effects of cyanide. We hypothesized that the aglycone of linamarin, acetone cyanohydrin, may be the cause of konzo. This nitrile rapidly decomposes into cyanide and acetone, but the particular exposure and nutrition conditions involved in the emergence of konzo may favor its stabilization and subsequent acute neurotoxicity. A number of preliminary observations were used to design an experiment to test this hypothesis. In the experiment, young female Long–Evans rats were given 10 mM acetone cyanohydrin in drinking water for 2 weeks, and then 20 mM for 6 weeks. Nutrition deficits associated with konzo were modeled by providing tapioca (cassava starch) as food for the last 3 of these weeks. After this period, rats were fasted for 24 h in order to increase endogenous acetone synthesis, and then exposed to 0 (control group) or 50 μmol/kg-h of acetone cyanohydrin for 24 h (treated group) through subcutaneous osmotic minipump infusion (n = 6/group). Motor activity and gait were evaluated before exposure (pre-test), and 1 and 6 days after exposure. Brains (n = 4) were stained for neuronal degeneration by fluoro-jade B. Rats exposed to 50 μmol/kg-h of acetone cyanohydrin showed acute signs of toxicity, but no persistent motor deficits. Two animals showed fluoro-jade staining in discrete thalamic nuclei, including the paraventricular and the ventral reuniens nuclei; one also exhibited labeling of the dorsal endopiriform nucleus. Similar effects were not elicited by equimolar KCN exposure. Therefore, acetone cyanohydrin may cause selective neuronal degeneration in the rat, but the affected areas are not those expected in an animal model of konzo. © 2009 Elsevier Inc. All rights reserved.
1. Introduction The tuberous root of cassava (Manihot esculenta) is the main staple food in many tropical and sub-tropical regions of the world. Cassava contains nitrile derivatives of glucose, mostly linamarin (2-(beta-Dglucopyranosyloxy)-2-methylpropanenitrile), and smaller amounts of lotaustralin. These and other similar nitriles are known as cyanogenic glucosides, because they release cyanide upon metabolism [17,49]. Disruption of the cassava cells puts linamarin in contact with linamarase, a selective beta-glucosidase present in the tissue, which metabolizes the glucoside into glucose and the aglycone, acetone cyanohydrin (CAS # 7586-5). This nitrile then decomposes spontaneously or enzymatically into acetone and HCN; therefore, chewing cassava tubers that contain high concentrations of the glucosides will lead to cyanide intoxication. ☆ Parts of the present work were presented at the 6th Forum of European Neuroscience (Geneva, Switzerland, July 2008). ⁎ Corresponding author. Departament de Ciències Fisiològiques II. Universitat de Barcelona. Feixa Llarga s/n. 08907 Hospitalet de Llobregat. Spain. Tel.: + 34 93 402 4277; fax: + 34 93 402 4268. E-mail address: [email protected] (J. Llorens). 1 Contributed equally to this work. 0892-0362/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.ntt.2009.11.001
There are many varieties, or cultivars, of cassava, which all contain different concentrations of glucosides. The cultivars that are most frequently used as a staple food tend to be the most toxic, because this feature protects the crop against pests and theft [11]. Because of their cyanogenic potential, they are appropriate for consumption only after thorough processing. Several procedures are available for processing, and safe flour can be obtained from highly toxic tubers [12]. However, adequate processing may require 3 to 5 days, and the correct procedures may be ignored when socio-economic circumstances make immediate access to this staple food the overriding consideration [3,15,44,45]. Dietary use of cassava has been associated with konzo, a neurotoxic disease affecting populations in rural areas of Africa. Clinically, konzo is characterized by the abrupt onset of an isolated and symmetric spastic paraparesis which is permanent but non-progressive. Spastic gait stands out as the chief sign of the disease [2,46]. Prevalence is higher in women and children than in adult males. Epidemic outbreaks of konzo are associated with periods of agro-ecological crisis, including drought and war, but endemic occurrence has also been reported. Invariably, the disease is associated with consumption of insufficiently processed, highly toxic cassava as the main (or the only) food for a period of several weeks [3,15,20,24,44,45,47]. One striking feature of the disease is that it appears
290
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
in an abrupt form, within days or even hours, strongly suggesting that the degeneration of a particular set of neurons is occurring in an acute time frame, despite the fact that a previous high intake of toxic cassava products for several weeks seems to be required for konzo development. The causative agent(s) and the pathogenic mechanisms of konzo are unidentified. Insufficiently processed cassava flour contains significant amounts of linamarin, acetone cyanohydrin, and cyanide [44], and the association of exposure to cyanide and linamarin and the acute stages of konzo is well established [3,44]. However, konzo is invariably linked to a chronic pattern of cassava intake; it is a chronic disease with effects that differ markedly from those of acute cassava poisoning [1,25]. What is more, none of the known effects of acute or chronic cyanide exposure in either humans or animals, such as convulsions and delayed Parkinsonism [9,26,34] match with the singular features of konzo [2]. Linamarin is considered to have low intrinsic toxicity, besides its cyanogenic potential [10,32] and the possibility that it may be the ultimate toxic agent remains largely unexplored. One major reason for this is the lack of an affordable source of gram quantities of linamarin for animal experimentation. So far, two main non-exclusive hypotheses have been proposed. The first is that a cyanide metabolite may be the causative agent of konzo; these metabolites include cyanate [41,42], thiocyanate [39] and 2-iminothiazolidine-4-carboxylic acid [4]. The second is that deficient intake of sulphur aminoacids may compromise cyanide metabolism to thiocyanate by rhodanese [14]. The attempts to explore these hypotheses [40–42] have not provided an animal model of konzo. The aim of this study was to explore an alternative hypothesis, namely that acetone cyanohydrin is the causative agent for konzo. This nitrile is chemically unstable in aqueous neutral or alkali solutions [18], but cassava flours associated with konzo contain significant amounts of it, due to their low pH as a consequence of lactic acid fermentation [44]. The available toxicological information for acetone cyanohydrin is scarce, but generally cyanide-like effects are reported [16,21], and exposure to this compound has been assumed to be equivalent to cyanide exposure. However, this evidence does not rule out the possibility that acetone cyanohydrin has an intrinsic neurotoxic potential. In fact, its chemical instability offers a possible explanation of the acute characteristics of the disease in association with chronic exposure: Only under particular circumstances would target organ concentrations of acetone cyanohydrin reach toxic levels; in most circumstances no cyanohydrin accumulation would occur, and its breakdown would instead lead to subtoxic cyanide exposure, as revealed by high thiocyanate serum and urine concentrations [3,44]. The hypothesized neurotoxic potential of acetone cyanohydrin would also be congruent with the fact that a number of small alkyl nitriles do cause a variety of toxic effects in selected populations of neurons and sensory cells [7,13,23,31], which may not depend on cyanide release [5,6,19]. To explore the hypothesis that acetone cyanohydrin is the causative agent in konzo, we studied the neurotoxic effects of this nitrile in the rat. Here we report a number of preliminary observations and an initial experiment evaluating its effects in an exposure model including several conditions associated with konzo such as chronic oral exposure and poor nutrition, which may be important for the expression of acetone cyanohydrin neurotoxicity and for survival after the associated cyanide exposure. Other conditions included final exposure to high doses, and fasting, which is known to increase the circulating concentrations of acetone. Acetone is one of the products of acetone cyanohydrin breakdown, and the presence of high acetone concentrations may increase the nitrile's half life.
cula, CA, USA). Other chemicals were of analytical grade as obtained from common commercial sources. 2.2. Animals The care and use of animals were in accordance with the Law 5/ 1995 and Act 214/1997 of the Autonomous Community (Generalitat) of Catalonia, and were approved by the Ethics Committee on Animal Experimentation of the University of Barcelona. Male and female Long–Evans rats (CERJ, Le-Genest-Saint-Isle, France) of different ages were used for preliminary studies. For final experiments, females of three weeks of age on receipt were used. The rats were housed two to four per cage in standard Macrolon cages (280 × 520 × 145 mm) with wood shavings as bedding at 22 ± 2 ºC. At least 7 days were provided for acclimation before experimentation. The rats were maintained on a 12:12 L:D cycle (0700:1900 h) and given standard diet pellets (A04, U.A.R., France) ad libitum except when otherwise indicated. For histology, rats were anesthetized with 400 mg kg− 1 chloral hydrate and transcardially perfused with 50 ml of heparinized saline followed by 400 ml of 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). 2.3. Dosing and experimental design We initially explored the toxicity of acetone cyanohydrin in a variety of exposure paradigms, using small numbers of animals, 1 to 3 in each group, as detailed throughout the Results section. For dosing, acetone cyanohydrin was dissolved in acidified saline (pH 3.0–3.5, with HCl) immediately before use. Subchronic exposure through acidified drinking water was also evaluated. Finally, acetone cyanohydrin exposure was performed using osmotic pumps, with the aim of modeling the continuous exposure to acetone cyanohydrin that may result from the release of this nitrile from ingested cassava flour. The final experiment used 12 females of the Long–Evans strain, aged 35–40 days (114–123 g) at the beginning of the study. The experimental design is illustrated in Fig. 1. As we hypothesized that konzo is caused by acute high dose acetone cyanohydrin exposure in subjects chronically exposed to the same cyanogenic nitrile, we used two groups of chronically exposed animals finally receiving (“treated”) or not receiving (“control”) the final dose. Thus, all rats were exposed to 10 mM acetone cyanohydrin in drinking water for 14 days, followed by 20 mM acetone cyanohydrin for 42 days. For the last 21 days in this period, the rats were deprived of standard food and instead given cassava starch (tapioca, Riera Marsa brand, Nabisco Iberia, Barcelona, Spain), with the aim of modeling the unbalanced diet that is associated with konzo development. The rats were then starved for 24 h in order to induce an increase in the concentration of acetone in the body. Osmotic pumps were subsequently implanted in the animals, delivering 0 (n = 6) or 50 (n = 6) μmol/kg/h of acetone cyanohydrin for 24 h. At the end of this period, the pumps were removed and the animals were returned to free access to standard food pellets. The behavior of the animals was assessed at days −8 (pre-test), 1 and 6 after the end of the pump exposure period. On each
2. Methods 2.1. Chemicals and reagents Acetone cyanohydrin (99%) was obtained from Aldrich Química (Alcobendas, Spain), KCN (N98%) from Fluka Chemika (Buchs, Switzerland), and Fluoro-Jade B from Chemicon International (Teme-
Fig. 1. Design of the experiment evaluating the neurotoxic effects of acetone cyanohydrin in the rat in an exposure model mimicking the conditions associated with human konzo. For further details, see the methods section. ACH = acetone cyanohydrin.
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
of these days, we assessed locomotor and rearing activities in the open field and gait performance. One rat exposed to subcutaneous acetone cyanohydrin was killed at day 2 on ethical grounds; its brain was included in the histological assessment. Two other treated animals and one control were killed for histology assessment at day 7. As a control experiment, we assessed the central nervous system effects of 24 h minipump exposure to KCN, at 50 μmol/kg/h (n = 5) and 60 μmol/kg/h (n = 3). Survivors (4 and 1, respectively) were assessed for fluoro-jade B staining at 7 days after exposure. 2.4. Osmotic pumps Osmotic pumps (ALZET) were from ALZA Co. (Palo Alto, CA, USA). Models used were number 2001D, with a nominal release rate of 8 μl/h for 24 h, and 2ML1, with a nominal release rate of 10 μl/h for 7 days. The pumps were filled and used according to the manufacturer's instructions. The concentrations of acetone cyanohydrin were adjusted to deliver the desired dose considering the body weight of each animal, although on some occasions, an initial fixed concentration was used and the exact acetone cyanohydrin dose calculated afterwards. Once filled, the osmotic pumps were placed in saline at 37 ºC for 3 h (model 2001D) or 24 h (model 2ML1) before implantation. The animals were anesthetized with isoflurane, and the pumps implanted subcutaneously in the interscapular region. The incision was sutured using tissue adhesive (Vetbond). At the end of the exposure period, the rats were briefly anesthetized for pump removal through the reopening of the same incision used for placement. 2.5. Open field Open field behavior was assessed in a white wood 1 × 1-m arena divided into 20 × 20-cm squares by black lines, enclosed with 50-cm high side walls and illuminated with a 100 W light bulb placed 70 cm above the floor. In 5 min sessions, the number of rears and the number of square crossings were counted [23]. 2.6. Gait topography analysis Stepping movements were evaluated following the method of Parker and Clarke [27], as previously described [37]. Briefly, after marking the rat hind and forefeet with ink (red and black ink respectively), the animal walks across chart paper on an elevated pathway leaving a permanent record of its footprints. Stride length and stride width were measured. 2.7. Histology Brain and spinal cord tissues were removed from the perfusionfixed animals and stored in the same fixative at 4 °C for subsequent staining. To examine all brain regions and also the spinal cord, the whole brain and one slice sample from each the cervical and the lumbar regions of the spinal cord were cut in transverse sections (50 µm) using a Leica VT1000M vibrating blade microtome. Every third section was dried onto a microscopy slide for staining with fluoro-jade B to identify degenerating neurons in the central nervous system [35,36]. Brain nuclei containing degenerating neurons were identified by comparison of fluoro-jade B stained sections with the appearance of the corresponding structures in the normal brain, according to the atlases by Paxinos et al. [29,30]. 2.8. Statistics Open field and gait data were tested with repeated measures MANOVA — Wilks' criterion — with day as the within-subject factor, using the SPSS 12.0.1 for Windows program package. The α level was set at 0.05.
291
3. Results 3.1. Preliminary observations Two adult male rats administered (ip) 59 μmol/kg of acetone cyanohydrin showed acute effects, including hyporeflexia, prostration or recumbent posture with muscular hypotonia, depressed or superficial and accelerated ventilation, and clonic movements of the limbs. These alternating effects started within 1–2 min of administration, and lasted 30–60 min. Similar effects were observed with oral doses of 235 μmol/kg (n = 2), while one rat which was administered a higher ip dose (88 μmol/kg) died through convulsions. Adult male rats exposed to acetone cyanohydrin in drinking water (n = 7, one animal per concentration from 0.68 to 43.72 mM, with doubling-up increments) showed no effect or only a small transient loss of body weight up to 10.9 mM; at this concentration, the rat ingested a daily dose of 350–590 μmol/kg of the nitrile. The animal exposed to 21.9 mM acetone cyanohydrin in the drinking water gradually lost almost 10% of its body weight, which it later recovered, with an estimated daily intake of 590–820 μmol/kg of acetone cyanohydrin. One rat given 43.72 mM of the nitrile in drinking water presented a gradual decline in body weight which reached 20% with no further progression; after stabilization of its body weight, this rat drank 40–50% of the volume drunk by control rats, and ingested a daily dose in the range of 1200–1800 μmol/kg of acetone cyanohydrin. This intake was maintained up to 30 days with no evidence of neurological effects. We next explored the toxicity of acetone cyanohydrin as released from 2ML1 osmotic minipumps. In young male adult rats, infusion of 16.5 μmol/kg/h (n = 2), 20.8 μmol/kg/h (n = 1), or 25.5 μmol/kg/h (n = 1) for 7 days caused no evidence of toxicity, while death occurred within 3 days of exposure to 34.1 μmol/kg/h (n=1) and within 24 h of exposure to 36.4 μmol/kg/h. One animal given 26.7 μmol/kg/h was judged to have slightly abnormal posture and gait at day 3 of exposure, and one animal exposed to 30 μmol/kg/h showed tachypnea and muscular hypotonia at day 1, but both showed full recovery afterwards. Using a 24 h delivery schedule with the 2001D pumps, one animal also died before this time of exposure to 35 μmol/kg/h of acetone cyanohydrin. The effect of minipump exposure to ACH was modified when the rats had been previously exposed to ACH via drinking water: adult male rats exposed to 20 mM of acetone cyanohydrin in drinking water for 4 weeks showed no overt effects after pump exposure to doses of 23.5 or 35.3 μmol/kg/h for 24 h and mild transient effects in behavior (decreased activity, slightly abnormal postures) after 58.8 or 82.3 μmol/kg/h (n = 1 per dose). One animal in which a pump was implanted delivering 117 μmol/kg/h of acetone cyanohydrin died after 24 h. In females exposed to 10 mM of acetone cyanohydrin in the drinking water for 3 weeks, followed by 20 mM for 4 weeks, no toxic effects appeared at a pump exposure of 94 μmol/kh/h for 24 h (n = 3), while the same dose was lethal in 2 out of 2 animals not previously exposed to the nitrile through drinking water. In young (4 week-old at the start of the study) females exposed to 10 mM of acetone cyanohydrin in the drinking water for 2 weeks, followed by 20 mM for 4–6 weeks, and finally fasted for 24 h before implantation of the osmotic pumps, lethality occurred by 24 h of exposure to 118 or 129 μmol/kg/h (n = 1/each), but not after exposure to 108 (n = 1) or 100 (n = 2) μmol/kg/h. The effect on acetone cyanohydrin toxicity of malnutrition was then explored in chronically exposed young female rats (2 weeks at 10 + 6 weeks at 20 mM) by substituting the food pellets with cassava starch for a number of days before minipump implantation. Whereas 2 animals on standard diet pellets survived 110 μmol/kg/h, cassava starch feeding for seven days resulted in increased lethality, with 3 out of 3 animals dying at 110 μmol/kg/h, 2 out of 2 dying at 100 μmol/kg/h, 1 out of 1 dying at 90 μmol/kg/h, and 1 out of 3 dying at 70 μmol/kg/h. When cassava starch feeding was introduced for the last three weeks
292
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
before implantation, mortality was recorded in one animal with final exposure to 60 μmol/kg/h, but 3 out of 4 animals survived after 24 h of exposure to 50 μmol/kg/h. These observations set the conditions selected for experiment use. 3.2. Effects of acetone cyanohydrin in exposure conditions mimicking those associated with konzo Female rats exposed from 4 weeks of age to 10 mM of acetone cyanohydrin in drinking water for two weeks followed by 20 mM for 7 weeks, and fed cassava starch for the last three of these weeks, did not show any overt neurological deficit. Subsequent exposure of these animals to subcutaneous osmotic pump delivery of 50 μmol/kg/h for 24 h, initiated after a 24 h period of food deprivation, caused behavioral effects indicative of acute toxicity, which were more evident during the last hours of exposure. These effects included immobility, decreased response to stimuli and abnormal walking postures with heightened hips; these abnormalities were present at notably different intensities in the different rats under acetone cyanohydrin infusion, and were not observed in animals in which pumps were implanted filled with control vehicle. No tremors, seizures and labored breathing were observed and no deaths occurred. One day after pump removal, 4 of the 6 acetone cyanohydrin rats were estimated to show a slightly abnormal posture, with more than normal extension of the hindpaws. One of the two rats with an apparently more abnormal behavior was sacrificed at day 2. The other rat of this pair was the only one out of five with an apparent abnormal walking posture at day 6. This second animal was processed for histology at day 7 along with a third acetone cyanohydrin rat and a saline minipump rat. Quantitative behavioral analysis showed no group differences between acetone cyanohydrin and saline rats at days 1 or 6 after minipump exposure (data not shown). In the open field, neither the treatment nor the treatment-by-day interaction factors were significant for either horizontal activity or rearing counts (all p's N 0.2). Similarly, gait analysis showed no effects of treatment on stepping patterns. Fore and hind limb stride length and stride width showed no treatment effect (all p's N 0.14) while significant treatment by day interactions for hind limb stride length (F(2,8) = 4.488, p = 0.049) and stride width (F(82,8) = 4.68, p = 0.045) were not followed by significant treatment effects at days 1 or 6 after exposure (all p's N 0.12). Among the rats exposed to acetone cyanohydrin, the one killed at day 2 after minipump removal, and one of the two killed at day 7 showed a positive fluoro-jade labeling in discrete brain areas, while the other rat (and the saline rat) showed no label in any brain region. In both labeled rats, the degeneration stain revealed neuronal profiles in a bilaterally symmetrical distribution. Some brain areas were labeled in both rats, while others were found only in the rat studied at day 7. Thus, in both rats, a consistent fluoro-jade staining occurred in discrete thalamic nuclei, namely the paraventricular and the ventral reuniens nuclei (Fig. 2A). In a few sections, an area lateral to the reuniens nucleus was also labeled, perhaps in the zona incerta or adjacent nuclei. In the rat studied at day 7, numerous labeled neurons were found across the whole dorsal endopiriform nucleus, spanning up to more than 10 mm in the rostro-caudal axis, with some involvement of adjacent areas of the lateral entorhinal cortex (Fig. 2B). 3.3. Effects of KCN One out of 5 young female rats exposed to 50 μmol/kg/h for 24 h of KCN died, while the other four animals showed no fluoro-jade B staining 7 days after osmotic pump exposure. After exposure to KCN at 60 μmol/kg/h, two out of three animals died, and the third one showed no fluoro-jade B staining at day 7 after dosing.
Fig. 2. Neuronal degeneration in the CNS of rats exposed to acetone cyanohydrin by minipump exposure (50 μmol/kg/h for 24 h). Fluoro-jade B stain of coronal vibratome sections (50 μm) of brain from a rat killed at 7 days after exposure. 3V = third ventricle. DEn = dorsal endopiriform nucleus. LEnt = lateral entorhinal cortex. PVA = paraventricular thalamic nucleus. rf = rhinal fissure. VRe = ventral reuniens nucleus.
4. Discussion The association of konzo with the consumption of insufficiently processed cassava has been firmly established [3], but the causative neurotoxic agent remains to be identified. The present study was carried out as a preliminary evaluation of the neurotoxicity of acetone cyanohydrin, the nitrile metabolite of linamarin, which is abundantly present in cassava flours associated with konzo. The results showed selective neuronal degeneration in a number of brain areas of some, but not all, of the rats exposed to acetone cyanohydrin in exposure conditions designed to mimic those associated with human konzo. The possibility that nitriles (organic compounds containing the cyano group) are the cause of konzo has received little consideration. Increasing evidence indicates that a number of nitriles cause a wide variety of neurotoxic effects, such as the neuronal degeneration caused in selected areas of the rat brain by trans-crotononitrile and 2,4-hexadienenitrile [7,37]. Although the main targets in the rat are
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
the inferior olive and the piriform cortex, degenerating neurons have also been observed in cortical areas [7], so it may be that a similar toxic action causes motor cortex damage in humans and accounts for the clinical signs in konzo. Both trans-crotononitrile and 2,4hexadienenitrile are α,β-unsaturated nitriles. Metabolism of these nitriles has been hypothesized to proceed by P450 mediated epoxidation of the double bound, followed by opening of the epoxide ring to form a cyanohydrin (α-hydroxy-nitrile), subsequently decomposing to release cyanide [38]. Therefore, we hypothesized that the α-hydroxy-nitrile end has a neurotoxic potential responsible for both human konzo and the neurotoxic effects of trans-crotononitrile and 2,4-hexadienenitrile in the rat, and focused on the most abundant cyanohydrin in cassava, acetone cyanohydrin, as a candidate causative agent for the disease. On initial evaluation, lethality occurred as expected at small acute doses, but drinking water exposure resulted in ingestion of large daily doses, up to several times the lethal dose given in a single bolus, with no apparent adverse neurological effects. Continuous exposure for one or seven days through osmotic minipump infusion also resulted either in death or in no overt evidence of neurological effects in survivors. When drinking water and minipump infusion exposure models were combined, rats previously exposed to acetone cyanohydrin via drinking water survived higher minipump doses than rats with no previous exposure. A possible explanation is that chronic acetone cyanohydrin exposure increases the activity of cyanide detoxification systems, allowing the animals to cope with higher cyanide amounts resulting from final high dose cyanohydrin exposure. We hypothesized that this increased survival might be important in generating a window of doses causing neurotoxicity but not death. Other conditions associated with development or increased risk of konzo are young age, female sex and malnutrition. To include these variables in the model, we chronically exposed female rats to acetone cyanohydrin through drinking water starting shortly after weaning, and included a period of cassava starch (tapioca) feeding. Successive introduction of these conditions required adjustments in the maximal sublethal doses that could be administered by minipump infusion as the candidate final precipitating event. It was particularly striking that, after 3 weeks on tapioca, lethal effects were recorded at doses half of those that were lethal in rats fed standard diet pellets throughout the whole period. This supports the widely diffused hypothesis [2] that cassava toxicity may depend on the nutritional status of the subject. Based on the preliminary observations, we designed an experiment to test the hypothesis that minipump exposure to sublethal acetone cyanohydrin doses induces neurotoxic effects in female rats “pre-conditioned” with chronic exposure to the compound, malnutrition, and final acute fasting to induce high body acetone concentrations. Under these exposure conditions, we observed overt signs of toxicity in the rats exposed to high doses of ACH by minipump infusion, and obtained proof of ongoing neuronal degeneration in some, but not all, of these animals, while no deaths were recorded. The animal showing the worst condition was sacrificed and processed for histology at day 2 to limit its suffering, but it was not certain that it would have died without our intervention. This is in contrast with the effects of minipump exposure to equal or slightly higher molar doses of KCN, which resulted in either death or complete absence of neuronal degeneration as assessed by fluoro-jade B staining. These results thus support our initial hypothesis that the particular exposure conditions included in the model would allow for the appearance of neurotoxic effects in the absence of subsequent mortality. However, subsequent experiments in the mouse have however not supported the hypothesis that high blood acetone concentrations play a key role in this toxicity. Fasted CYP2E1-null mice, which are known to accumulate higher acetone concentrations [8], showed similar susceptibility to acetone cyanohydrin acute toxicity to wild type animals (unpublished data).
293
The final exposure to acetone cyanohydrin through minipump infusion resulted in a number of changes in behavior that were compatible with either general illness or motor effects. However, a marked recovery occurred in most animals after the 24 h exposure period, and in fact no consistent motor effects could be demonstrated by quantitative assessment at 1 and 6 days after exposure. Although the methods used were simple, and perhaps more refined methods could have revealed group differences, no major alterations in motor control occurred. Konzo has been hypothesized to involve damage of the cortico-spinal tract [43], which differs substantially between humans and rats [22], so the apparent lack of a measurable motor effect in the acetone cyanohydrin rats would not constitute by itself definite proof of the lack of relevance of this rat model to human konzo. Damage to the cortico-spinal system is hypothesized in konzo [43], but no pathological data are available and it may be the case that other motor-related areas are involved. Species-dependent differences in lesion sites may also appear even though a common mechanism of action acts across species. Therefore, we did not search for lesions in pre-selected regions, but instead used fluoro-jade B staining to screen for degenerating neurons in the whole of the central nervous system. The effectiveness of this stain in revealing neuronal degeneration, and the simplicity of its use, allows for a complete evaluation of the system and the identification of neuronal targets without the need for a previous hypothesis of where the damage may be [7,35,36]. The data revealed the presence of degenerating neurons in 2 out of 3 rats examined. The lesions recorded were bilaterally symmetrical and coincided in the two affected animals, although one rat had degenerating neurons in more areas than the other; these findings support the conclusion that the lesions observed were the expression of a selective pattern of neuronal degeneration. However, the pattern observed was not the one anticipated in an animal model of konzo, which would be expected to involve the motor cortex and/or other motor-related areas. Nor did the observed pattern match the one presented by the nitriles with CNS toxicity, trans-crotononitrile and hexadienenitrile, which mainly involve the inferior olive and the piriform cortex [7,37]. The roles of the paraventricular and reuniens nuclei, apparently the main acetone cyanohydrin targets, are not precisely defined, but most of the evidence relates these structures to limbic-cognitive or visceral/autonomic functions [28,48]. Calculations made on the basis of data in the literature [3,10,44] give an estimation of maximal acetone cyanohydrin doses in konzo cases in the range of 10 to 100 μmol/kg/day, which were effectively exceeded by our final 50 μmol/kg/h dosing schedule. This makes it unlikely that the mismatch between the neurotoxic effects recorded in the present model and those expected in a useful animal model of konzo is due to a failure to attain large enough doses. Degeneration of the paraventricular and reuniens nuclei and the dorsal endopiriform nucleus has been observed in rats administered single intraseptal injections of kainic acid as an animal model of status epilepticus [33]. In that model, evidence for degeneration of the dorsal endopiriform nucleus was observed at 7 days after exposure, while the thalamic nuclei were affected after one day. As we observed all these areas in the affected animal at 7 days, but at day 2 the dorsal endopiriform nucleus was spared, it is tempting to speculate that similar mechanisms may be involved in these models. It is thus possible that our exposure model caused lesions through kainate-like convulsant activity. However, the status epilepticus induced by intraseptal kainic injection typically causes hippocampal lesions [33], and the hippocampus was spared in the acetone cyanohydrin rats. And although our assessment was not extensive enough to rule out the presence of convulsive activity, we did not observe status epilepticus in our rats during handling and behavioral testing. The value of the present observations in the development of an animal model of konzo remains uncertain. The data gathered do not provide a useful animal model for konzo, and thus do not support the notion that acetone cyanohydrin is the causal agent, although they do
294
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
not conclusively disprove the hypothesis either. Our observation, for the first time, of selective neuronal degeneration in some animals exposed to acetone cyanohydrin supports several of the starting assumptions, the main one being that acetone cyanohydrin neurotoxicity may, under particular exposure conditions, differ from that elicited by simple acute or chronic cyanide exposure. While this would call for additional research, a major difficulty is posed by the fact that sublethal doses, associated with overt systemic toxicity, appeared to be necessary for the neurotoxic effects, while konzo typically appears as a neurotoxic disease involving little or no systemic toxicity. One recent observation that challenges the hypothesis further is the finding that, at least for the sensory toxicity of allylnitrile, the corresponding cyanohydrin is unlikely to be the bioactivated toxic metabolite of the parent nitrile [6]. Conflict of interest We have no conflicting interest to declare. Acknowledgements We thank our students Fabian Márquez and Eduardo Balbuena for their contribution to preliminary experiments. This work was supported by grants BFI2003-01606 and BFU2006-00343/BFI from the Spanish Ministry of Science and Innovation/EU FEDER and 2005SGR00022 from the Generalitat of Catalonia.
References [1] A. Akintonwa, O. Tunwashe, A. Onifade, Fatal and non-fatal acute poisoning attributed to cassava-based meal, Acta Hortic. 375 (1994) 285–288. [2] Anonymous, Konzo — a distinct type of upper motoneuron disease, Week. Epidem. Rec. WHO 71 (1996) 225–228. [3] J.P. Banea-Mayambu, T. Tylleskär, N. Gitebo, N. Matadi, M. Gebre-Medhin, H. Rosling, Geographical and seasonal association between linamarin and cyanide exposure from cassava and the upper motor neurone disease konzo in former Zaire, Trop. Med. Int. Health. 2 (1997) 1143–1151. [4] R.S. Bitner, A. Kanthasamy, G.E. Isom, G.K.W. Yim, Seizures and selective CA-1 hippocampal lesions induced by an excitotoxic cyanide metabolite, 2-iminothiazolidine-4-carboxylic acid, Neurotoxicology 16 (1995) 115–122. [5] P. Boadas-Vaello, E. Jover, N. Díez-Padrisa, J.M. Bayona, J. Llorens, Differential role of CYP2E1-mediated metabolism in the lethal and vestibulotoxic effects of ciscrotononitrile in the mouse, Toxicol. Appl. Pharmacol. 225 (2007) 310–317. [6] P. Boadas-Vaello, E. Jover, S. Saldaña-Ruíz, C. Soler-Martín, C. Chabbert, J.M. Bayona, J. Llorens, Allylnitrile metabolism by CYP2E1 and other CYPs leads to distinct lethal and vestibulotoxic effects in the mouse, Toxicol. Sci. 107 (2009) 461–472. [7] P. Boadas-Vaello, J. Riera, J. Llorens, Behavioral and pathological effects in the rat define two groups of neurotoxic nitriles, Toxicol. Sci. 88 (2005) 456–466. [8] F.Y. Bondoc, Z. Bao, W.Y. Hu, F.J. Gonzalez, Y. Wang, C.S. Yang, J.Y. Hong, Acetone catabolism by cytochrome P450 2E1: studies with CYP2E1-null mice, Biochem. Pharmacol. 58 (1999) 461–463. [9] J.L. Borowitz, G.E. Isom, D.V. Nakles, Human toxicology of cyanide, in: D.A. Dzombak, R.S. Ghosh, G.M. Wong-Chong (Eds.), Cyanide in Water and Soil: Chemistry, Risk and Management, CRC Press-Taylor and Francis Group, Boca Raton, 2006, pp. 237–250. [10] L. Carlsson, N. Mlingi, A. Juma, G. Ronquist, H. Rosling, Metabolic fates in humans of linamarin in cassava flour ingested as stiff porridge, Food Chem. Toxicol. 37 (1999) 307–312. [11] L. Chiwona-Karltun, J. Mkumbira, J. Saka, M. Bovin, N.M. Mahungu, H. Rosling, The importance of being bitter — a qualitative study on cassava cultivar preference in Malawi, Ecol. Food Nutr. 37 (1998) 219–245. [12] L. Chiwona-Karltun, T. Tylleskär, J. Mkumbira, M. Gebre-Medhin, H. Rosling, Low dietary cyanogen exposure from frequent consumption of potentially toxic cassava in Malawi, Int. J. Food Sci. Nutr. 51 (2000) 33–43. [13] S.M. Chou, H.A. Hartmann, Axonal lesions and waltzing syndrome after IDPN administration in rats. With a concept — axostasis, Acta Neuropathol. 3 (1964) 428–450. [14] J. Cliff, P. Lundqvist, J. Mårtensson, H. Rosling, B. Sörbo, Association of high cyanide and low sulphur intake in cassava-induced spastic paraparesis, Lancet 8466 (1985) 1211–1213. [15] J. Cliff, D. Nicala, F. Saute, R. Givragy, G. Azambuja, A. Taela, L. Chavane, J. Howarth, Konzo associated with war in Mozambique, Trop. Med. Int. Health. 2 (1997) 1068–1074. [16] Committee on Acute Exposure Guideline Levels, Acetone cyanohydrin, National Research Council Committee on Acute Exposure Guideline Levels, Acute Exposure Guidelines for Selected Airbone Chemicals, vol 7, National Academies Press, Washington, DC, 2009, pp. 13–39.
[17] E.E. Conn, Cyanogenesis — a personal perspectiva, Acta Hortic. 375 (1994) 31–43. [18] R.D. Cooke, An enzymatic assay for the total cyanide content of cassava (Manihot esculenta Crantz), J. Sci. Food. Agric. 29 (1978) 345–352. [19] J.R. Froines, E.M. Postlethwait, E.J. LaFuente, W.C.V. Liu, In vivo and in vitro release of cyanide from neurotoxic aminonitriles, J. Toxicol. Environ. Health 16 (1985) 449–460. [20] W.P. Howlett, G.R. Brubaker, N. Mlingi, H. Rosling, Konzo, an epidemic upper motor neuron disease studied in Tanzania, Brain 113 (1990) 223–235. [21] F.R. Johannsen, G.J. Levinskas, Relationships between toxicity and structure of aliphatic nitriles, Fundam. Appl. Toxicol. 7 (1986) 690–697. [22] R.N. Lemon, Descending pathways in motor control, Annu. Rev. Neurosci. 31 (2008) 195–218. [23] J. Llorens, D. Demêmes, A. Sans, The behavioral syndrome caused by 3, 3′iminodipropionitrile and related nitriles in the rat is associated with degeneration of the vestibular sensory hair cells, Toxicol. Appl. Pharmacol. 123 (1993) 199–210. [24] Ministry of Health, Mozambique, Mantakassa: an epidemic of spastic paraparesis associated with chronic cyanide intoxication in a cassava staple area of Mozambique. 1. Epidemiology and clinical and laboratory findings in patients, Bull. World Hlth. Organ. 62 (1984) 477–484. [25] N. Mlingi, N.H. Poulter, H. Rosling, An outbreak of acute intoxications from consumption of insufficiently processed cassava in Tanzania, Nutr. Res. 12 (1992) 677–687. [26] M. Orfila, Traité des poisons tirés des regnes minéral, végétal et animal, ou toxicologie générale, considérée sous les rapports de la physiologie, de la pathologie et de la médicine légale, Third Ed.Crochard et Gabon Editeurs, Paris, 1826. [27] A.J. Parker, K.A. Clarke, Gait topography in rat locomotion, Physiol. Behav. 48 (1990) 41–47. [28] M.P. Parsons, S. Li, G.J. Kirouac, Functional and anatomical connection between the paraventricular nucleus of the thalamus and dopamine fibers of the nucleus accumbens, J. Comp. Neurol. 500 (2007) 1050–1063. [29] G. Paxinos, P. Carrive, H. Wang, P-Y. Wang, Chemoarchitectonic atlas of the rat brainstem, Academic Press, San Diego, 1999. [30] G. Paxinos, L. Kus, K.W.S. Ashwell, C. Watson, Chemoarchitectonic Atlas of the Rat Forebrain, Academic Press, San Diego, 1999. [31] A. Pestronk, N, N′-dimethylaminopropionitrile, in: P.S. Spencer, H.H. Schaumburg, A.C. Ludolph (Eds.), Experimental and clinical neurotoxicology, Second Ed., Oxford University Press, New York, 2000, pp. 494–502. [32] D.J. Philbrick, D.C. Hill, J.C. Alexander, Physiological and biochemical changes associated with linamarin administration to rats, Toxicol. Appl. Pharmacol. 42 (1977) 539–551. [33] M. Revuelta, A. Castaño, J.L. Venero, A. Machado, J. Cano, Long-lasting induction of brain-derived neurotrophic factor is restricted to resistant cell populations in an animal model of status epilepticus, Neuroscience 103 (2001) 955–969. [34] N.L. Rosenberg, J.A. Myers, W.R. Martin, Cyanide-induced parkinsonism: clinical, MRI, and 6-fluorodopa PET studies, Neurology 39 (1989) 142–144. [35] L.C. Schmued, K.J. Hopkins, Fluoro-Jade B: a high affinity fluorescent marker for the localization of neuronal degeneration, Brain Res. 874 (2000) 123–130. [36] L.C. Schmued, K.J. Hopkins, Fluoro-Jade: novel fluorochromes for detecting toxicant-induced neuronal degeneration, Toxicol. Pathol. 28 (2000) 91–99. [37] A. Seoane, R. Apps, E. Balbuena, L. Herrero, J. Llorens, Differential effects of transcrotononitrile and 3-acetylpyridine on inferior olive integrity and behavioural performance in the rat, Eur. J. Neurosci. 22 (2005) 880–894. [38] E.H. Silver, S.H. Kuttab, T. Hasan, M. Hassan, Structural considerations in the metabolism of nitriles to cyanide in vivo, Drug Metab. Dispos. 10 (1982) 495–498. [39] P.S. Spencer, Food toxins, AMPA receptors, and motor neuron diseases, Drug Metab. Rev. 31 (1999) 561–587. [40] I. Swenne, U.J. Eriksson, R. Christoffersson, B. Kågedal, P. Lundquist, L. Nilsson, T. Tylleskär, H. Rosling, Cyanide detoxification in rats exposed to acetonitrile and fed a low protein diet, Fundam. Appl. Toxicol. 32 (1996) 66–71. [41] J. Tor-Agbidye, V.S. Palmer, M.R. Lasarev, A.M. Craig, L.L. Blythe, M.I. Sabri, P.S. Spencer, Bioactivation of cyanide to cyanate in sulfur amino acid deficiency: relevance to neurological disease in humans subsisting on cassava, Toxicol. Sci. 50 (1999) 228–235. [42] J. Tor-Agbidye, V.S. Palmer, P.S. Spencer, A.M. Craig, L.L. Blythe, M.I. Sabri, Sodium cyanate alters glutathione homeostasis in rodent brain: relationship to neurodegenerative diseases in protein-deficient malnourished populations in Africa, Brain Res. 820 (1999) 12–19. [43] D. Tshala-Katumbay, K.E. Eeg-Olofsson, T. Kazadi-Kayembe, T. Tylleskär, P. Fällmar, Analysis of motor pathway involvement in konzo using transcranial electrical and magnetic stimulation, Muscle Nerve. 25 (2002) 230–235. [44] T. Tylleskär, M. Banea, N. Bikangi, R.D. Cooke, N.H. Poulter, H. Rosling, Cassava cyanogens and konzo, an upper motoneuron disease found in Africa, Lancet 339 (1992) 208–211. [45] T. Tylleskär, M. Banea, N. Bikangi, G. Nahimana, L.A. Persson, H. Rosling, Dietary determinants of a non-progressive spastic paraparesis (Konzo): a case-referent study in a high incidence area of Zaire, Int. J. Epidemiol. 24 (1995) 949–956. [46] T. Tylleskär, W.P. Howlett, H.T. Rwiza, S.M. Aquilonius, E. Stålberg, B. Lindén, A. Mandahl, H.C. Larsen, G.R. Brubaker, H. Rosling, Konzo: a distinct disease entity with selective upper motor neuron damage, J. Neurol. Neurosurg. Psychiatry. 56 (1993) 638–643. [47] T. Tylleskär, F.D. Légué, S. Peterson, E. Kpizingui, P. Stecker, Konzo in the Central African Republic, Neurology 44 (1994) 959–961. [48] R.P. Vertes, Differential projections of the infralimbic and prelimbic cortex in the rat, Synapse 51 (2004) 32–58. [49] M. Zagrobelny, M. Soren, B.L. Moller, Cyanogenesis in plants and arthropods, Phytochemistry 69 (2008) 1457–1468.
Contents lists available at ScienceDirect
Neurotoxicology and Teratology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / n e u t e r a
Brief communication
The targets of acetone cyanohydrin neurotoxicity in the rat are not the ones expected in an animal model of konzo☆ Carla Soler-Martín 1, Judith Riera 1, Ana Seoane, Blanca Cutillas, Santiago Ambrosio, Pere Boadas-Vaello, Jordi Llorens ⁎ Departament de Ciències Fisiològiques II, Universitat de Barcelona, 08907 Hospitalet de Llobregat, Spain
a r t i c l e
i n f o
Article history: Received 13 October 2009 Received in revised form 12 November 2009 Accepted 12 November 2009 Available online 20 November 2009 Keywords: Konzo Cyanide Manihot esculenta Nitriles Acetone cyanohydrin
a b s t r a c t Konzo is a neurotoxic motor disease caused by excess consumption of insufficiently processed cassava. Cassava contains the cyanogenic glucoside linamarin, but konzo does not present the known pathological effects of cyanide. We hypothesized that the aglycone of linamarin, acetone cyanohydrin, may be the cause of konzo. This nitrile rapidly decomposes into cyanide and acetone, but the particular exposure and nutrition conditions involved in the emergence of konzo may favor its stabilization and subsequent acute neurotoxicity. A number of preliminary observations were used to design an experiment to test this hypothesis. In the experiment, young female Long–Evans rats were given 10 mM acetone cyanohydrin in drinking water for 2 weeks, and then 20 mM for 6 weeks. Nutrition deficits associated with konzo were modeled by providing tapioca (cassava starch) as food for the last 3 of these weeks. After this period, rats were fasted for 24 h in order to increase endogenous acetone synthesis, and then exposed to 0 (control group) or 50 μmol/kg-h of acetone cyanohydrin for 24 h (treated group) through subcutaneous osmotic minipump infusion (n = 6/group). Motor activity and gait were evaluated before exposure (pre-test), and 1 and 6 days after exposure. Brains (n = 4) were stained for neuronal degeneration by fluoro-jade B. Rats exposed to 50 μmol/kg-h of acetone cyanohydrin showed acute signs of toxicity, but no persistent motor deficits. Two animals showed fluoro-jade staining in discrete thalamic nuclei, including the paraventricular and the ventral reuniens nuclei; one also exhibited labeling of the dorsal endopiriform nucleus. Similar effects were not elicited by equimolar KCN exposure. Therefore, acetone cyanohydrin may cause selective neuronal degeneration in the rat, but the affected areas are not those expected in an animal model of konzo. © 2009 Elsevier Inc. All rights reserved.
1. Introduction The tuberous root of cassava (Manihot esculenta) is the main staple food in many tropical and sub-tropical regions of the world. Cassava contains nitrile derivatives of glucose, mostly linamarin (2-(beta-Dglucopyranosyloxy)-2-methylpropanenitrile), and smaller amounts of lotaustralin. These and other similar nitriles are known as cyanogenic glucosides, because they release cyanide upon metabolism [17,49]. Disruption of the cassava cells puts linamarin in contact with linamarase, a selective beta-glucosidase present in the tissue, which metabolizes the glucoside into glucose and the aglycone, acetone cyanohydrin (CAS # 7586-5). This nitrile then decomposes spontaneously or enzymatically into acetone and HCN; therefore, chewing cassava tubers that contain high concentrations of the glucosides will lead to cyanide intoxication. ☆ Parts of the present work were presented at the 6th Forum of European Neuroscience (Geneva, Switzerland, July 2008). ⁎ Corresponding author. Departament de Ciències Fisiològiques II. Universitat de Barcelona. Feixa Llarga s/n. 08907 Hospitalet de Llobregat. Spain. Tel.: + 34 93 402 4277; fax: + 34 93 402 4268. E-mail address: [email protected] (J. Llorens). 1 Contributed equally to this work. 0892-0362/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.ntt.2009.11.001
There are many varieties, or cultivars, of cassava, which all contain different concentrations of glucosides. The cultivars that are most frequently used as a staple food tend to be the most toxic, because this feature protects the crop against pests and theft [11]. Because of their cyanogenic potential, they are appropriate for consumption only after thorough processing. Several procedures are available for processing, and safe flour can be obtained from highly toxic tubers [12]. However, adequate processing may require 3 to 5 days, and the correct procedures may be ignored when socio-economic circumstances make immediate access to this staple food the overriding consideration [3,15,44,45]. Dietary use of cassava has been associated with konzo, a neurotoxic disease affecting populations in rural areas of Africa. Clinically, konzo is characterized by the abrupt onset of an isolated and symmetric spastic paraparesis which is permanent but non-progressive. Spastic gait stands out as the chief sign of the disease [2,46]. Prevalence is higher in women and children than in adult males. Epidemic outbreaks of konzo are associated with periods of agro-ecological crisis, including drought and war, but endemic occurrence has also been reported. Invariably, the disease is associated with consumption of insufficiently processed, highly toxic cassava as the main (or the only) food for a period of several weeks [3,15,20,24,44,45,47]. One striking feature of the disease is that it appears
290
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
in an abrupt form, within days or even hours, strongly suggesting that the degeneration of a particular set of neurons is occurring in an acute time frame, despite the fact that a previous high intake of toxic cassava products for several weeks seems to be required for konzo development. The causative agent(s) and the pathogenic mechanisms of konzo are unidentified. Insufficiently processed cassava flour contains significant amounts of linamarin, acetone cyanohydrin, and cyanide [44], and the association of exposure to cyanide and linamarin and the acute stages of konzo is well established [3,44]. However, konzo is invariably linked to a chronic pattern of cassava intake; it is a chronic disease with effects that differ markedly from those of acute cassava poisoning [1,25]. What is more, none of the known effects of acute or chronic cyanide exposure in either humans or animals, such as convulsions and delayed Parkinsonism [9,26,34] match with the singular features of konzo [2]. Linamarin is considered to have low intrinsic toxicity, besides its cyanogenic potential [10,32] and the possibility that it may be the ultimate toxic agent remains largely unexplored. One major reason for this is the lack of an affordable source of gram quantities of linamarin for animal experimentation. So far, two main non-exclusive hypotheses have been proposed. The first is that a cyanide metabolite may be the causative agent of konzo; these metabolites include cyanate [41,42], thiocyanate [39] and 2-iminothiazolidine-4-carboxylic acid [4]. The second is that deficient intake of sulphur aminoacids may compromise cyanide metabolism to thiocyanate by rhodanese [14]. The attempts to explore these hypotheses [40–42] have not provided an animal model of konzo. The aim of this study was to explore an alternative hypothesis, namely that acetone cyanohydrin is the causative agent for konzo. This nitrile is chemically unstable in aqueous neutral or alkali solutions [18], but cassava flours associated with konzo contain significant amounts of it, due to their low pH as a consequence of lactic acid fermentation [44]. The available toxicological information for acetone cyanohydrin is scarce, but generally cyanide-like effects are reported [16,21], and exposure to this compound has been assumed to be equivalent to cyanide exposure. However, this evidence does not rule out the possibility that acetone cyanohydrin has an intrinsic neurotoxic potential. In fact, its chemical instability offers a possible explanation of the acute characteristics of the disease in association with chronic exposure: Only under particular circumstances would target organ concentrations of acetone cyanohydrin reach toxic levels; in most circumstances no cyanohydrin accumulation would occur, and its breakdown would instead lead to subtoxic cyanide exposure, as revealed by high thiocyanate serum and urine concentrations [3,44]. The hypothesized neurotoxic potential of acetone cyanohydrin would also be congruent with the fact that a number of small alkyl nitriles do cause a variety of toxic effects in selected populations of neurons and sensory cells [7,13,23,31], which may not depend on cyanide release [5,6,19]. To explore the hypothesis that acetone cyanohydrin is the causative agent in konzo, we studied the neurotoxic effects of this nitrile in the rat. Here we report a number of preliminary observations and an initial experiment evaluating its effects in an exposure model including several conditions associated with konzo such as chronic oral exposure and poor nutrition, which may be important for the expression of acetone cyanohydrin neurotoxicity and for survival after the associated cyanide exposure. Other conditions included final exposure to high doses, and fasting, which is known to increase the circulating concentrations of acetone. Acetone is one of the products of acetone cyanohydrin breakdown, and the presence of high acetone concentrations may increase the nitrile's half life.
cula, CA, USA). Other chemicals were of analytical grade as obtained from common commercial sources. 2.2. Animals The care and use of animals were in accordance with the Law 5/ 1995 and Act 214/1997 of the Autonomous Community (Generalitat) of Catalonia, and were approved by the Ethics Committee on Animal Experimentation of the University of Barcelona. Male and female Long–Evans rats (CERJ, Le-Genest-Saint-Isle, France) of different ages were used for preliminary studies. For final experiments, females of three weeks of age on receipt were used. The rats were housed two to four per cage in standard Macrolon cages (280 × 520 × 145 mm) with wood shavings as bedding at 22 ± 2 ºC. At least 7 days were provided for acclimation before experimentation. The rats were maintained on a 12:12 L:D cycle (0700:1900 h) and given standard diet pellets (A04, U.A.R., France) ad libitum except when otherwise indicated. For histology, rats were anesthetized with 400 mg kg− 1 chloral hydrate and transcardially perfused with 50 ml of heparinized saline followed by 400 ml of 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). 2.3. Dosing and experimental design We initially explored the toxicity of acetone cyanohydrin in a variety of exposure paradigms, using small numbers of animals, 1 to 3 in each group, as detailed throughout the Results section. For dosing, acetone cyanohydrin was dissolved in acidified saline (pH 3.0–3.5, with HCl) immediately before use. Subchronic exposure through acidified drinking water was also evaluated. Finally, acetone cyanohydrin exposure was performed using osmotic pumps, with the aim of modeling the continuous exposure to acetone cyanohydrin that may result from the release of this nitrile from ingested cassava flour. The final experiment used 12 females of the Long–Evans strain, aged 35–40 days (114–123 g) at the beginning of the study. The experimental design is illustrated in Fig. 1. As we hypothesized that konzo is caused by acute high dose acetone cyanohydrin exposure in subjects chronically exposed to the same cyanogenic nitrile, we used two groups of chronically exposed animals finally receiving (“treated”) or not receiving (“control”) the final dose. Thus, all rats were exposed to 10 mM acetone cyanohydrin in drinking water for 14 days, followed by 20 mM acetone cyanohydrin for 42 days. For the last 21 days in this period, the rats were deprived of standard food and instead given cassava starch (tapioca, Riera Marsa brand, Nabisco Iberia, Barcelona, Spain), with the aim of modeling the unbalanced diet that is associated with konzo development. The rats were then starved for 24 h in order to induce an increase in the concentration of acetone in the body. Osmotic pumps were subsequently implanted in the animals, delivering 0 (n = 6) or 50 (n = 6) μmol/kg/h of acetone cyanohydrin for 24 h. At the end of this period, the pumps were removed and the animals were returned to free access to standard food pellets. The behavior of the animals was assessed at days −8 (pre-test), 1 and 6 after the end of the pump exposure period. On each
2. Methods 2.1. Chemicals and reagents Acetone cyanohydrin (99%) was obtained from Aldrich Química (Alcobendas, Spain), KCN (N98%) from Fluka Chemika (Buchs, Switzerland), and Fluoro-Jade B from Chemicon International (Teme-
Fig. 1. Design of the experiment evaluating the neurotoxic effects of acetone cyanohydrin in the rat in an exposure model mimicking the conditions associated with human konzo. For further details, see the methods section. ACH = acetone cyanohydrin.
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
of these days, we assessed locomotor and rearing activities in the open field and gait performance. One rat exposed to subcutaneous acetone cyanohydrin was killed at day 2 on ethical grounds; its brain was included in the histological assessment. Two other treated animals and one control were killed for histology assessment at day 7. As a control experiment, we assessed the central nervous system effects of 24 h minipump exposure to KCN, at 50 μmol/kg/h (n = 5) and 60 μmol/kg/h (n = 3). Survivors (4 and 1, respectively) were assessed for fluoro-jade B staining at 7 days after exposure. 2.4. Osmotic pumps Osmotic pumps (ALZET) were from ALZA Co. (Palo Alto, CA, USA). Models used were number 2001D, with a nominal release rate of 8 μl/h for 24 h, and 2ML1, with a nominal release rate of 10 μl/h for 7 days. The pumps were filled and used according to the manufacturer's instructions. The concentrations of acetone cyanohydrin were adjusted to deliver the desired dose considering the body weight of each animal, although on some occasions, an initial fixed concentration was used and the exact acetone cyanohydrin dose calculated afterwards. Once filled, the osmotic pumps were placed in saline at 37 ºC for 3 h (model 2001D) or 24 h (model 2ML1) before implantation. The animals were anesthetized with isoflurane, and the pumps implanted subcutaneously in the interscapular region. The incision was sutured using tissue adhesive (Vetbond). At the end of the exposure period, the rats were briefly anesthetized for pump removal through the reopening of the same incision used for placement. 2.5. Open field Open field behavior was assessed in a white wood 1 × 1-m arena divided into 20 × 20-cm squares by black lines, enclosed with 50-cm high side walls and illuminated with a 100 W light bulb placed 70 cm above the floor. In 5 min sessions, the number of rears and the number of square crossings were counted [23]. 2.6. Gait topography analysis Stepping movements were evaluated following the method of Parker and Clarke [27], as previously described [37]. Briefly, after marking the rat hind and forefeet with ink (red and black ink respectively), the animal walks across chart paper on an elevated pathway leaving a permanent record of its footprints. Stride length and stride width were measured. 2.7. Histology Brain and spinal cord tissues were removed from the perfusionfixed animals and stored in the same fixative at 4 °C for subsequent staining. To examine all brain regions and also the spinal cord, the whole brain and one slice sample from each the cervical and the lumbar regions of the spinal cord were cut in transverse sections (50 µm) using a Leica VT1000M vibrating blade microtome. Every third section was dried onto a microscopy slide for staining with fluoro-jade B to identify degenerating neurons in the central nervous system [35,36]. Brain nuclei containing degenerating neurons were identified by comparison of fluoro-jade B stained sections with the appearance of the corresponding structures in the normal brain, according to the atlases by Paxinos et al. [29,30]. 2.8. Statistics Open field and gait data were tested with repeated measures MANOVA — Wilks' criterion — with day as the within-subject factor, using the SPSS 12.0.1 for Windows program package. The α level was set at 0.05.
291
3. Results 3.1. Preliminary observations Two adult male rats administered (ip) 59 μmol/kg of acetone cyanohydrin showed acute effects, including hyporeflexia, prostration or recumbent posture with muscular hypotonia, depressed or superficial and accelerated ventilation, and clonic movements of the limbs. These alternating effects started within 1–2 min of administration, and lasted 30–60 min. Similar effects were observed with oral doses of 235 μmol/kg (n = 2), while one rat which was administered a higher ip dose (88 μmol/kg) died through convulsions. Adult male rats exposed to acetone cyanohydrin in drinking water (n = 7, one animal per concentration from 0.68 to 43.72 mM, with doubling-up increments) showed no effect or only a small transient loss of body weight up to 10.9 mM; at this concentration, the rat ingested a daily dose of 350–590 μmol/kg of the nitrile. The animal exposed to 21.9 mM acetone cyanohydrin in the drinking water gradually lost almost 10% of its body weight, which it later recovered, with an estimated daily intake of 590–820 μmol/kg of acetone cyanohydrin. One rat given 43.72 mM of the nitrile in drinking water presented a gradual decline in body weight which reached 20% with no further progression; after stabilization of its body weight, this rat drank 40–50% of the volume drunk by control rats, and ingested a daily dose in the range of 1200–1800 μmol/kg of acetone cyanohydrin. This intake was maintained up to 30 days with no evidence of neurological effects. We next explored the toxicity of acetone cyanohydrin as released from 2ML1 osmotic minipumps. In young male adult rats, infusion of 16.5 μmol/kg/h (n = 2), 20.8 μmol/kg/h (n = 1), or 25.5 μmol/kg/h (n = 1) for 7 days caused no evidence of toxicity, while death occurred within 3 days of exposure to 34.1 μmol/kg/h (n=1) and within 24 h of exposure to 36.4 μmol/kg/h. One animal given 26.7 μmol/kg/h was judged to have slightly abnormal posture and gait at day 3 of exposure, and one animal exposed to 30 μmol/kg/h showed tachypnea and muscular hypotonia at day 1, but both showed full recovery afterwards. Using a 24 h delivery schedule with the 2001D pumps, one animal also died before this time of exposure to 35 μmol/kg/h of acetone cyanohydrin. The effect of minipump exposure to ACH was modified when the rats had been previously exposed to ACH via drinking water: adult male rats exposed to 20 mM of acetone cyanohydrin in drinking water for 4 weeks showed no overt effects after pump exposure to doses of 23.5 or 35.3 μmol/kg/h for 24 h and mild transient effects in behavior (decreased activity, slightly abnormal postures) after 58.8 or 82.3 μmol/kg/h (n = 1 per dose). One animal in which a pump was implanted delivering 117 μmol/kg/h of acetone cyanohydrin died after 24 h. In females exposed to 10 mM of acetone cyanohydrin in the drinking water for 3 weeks, followed by 20 mM for 4 weeks, no toxic effects appeared at a pump exposure of 94 μmol/kh/h for 24 h (n = 3), while the same dose was lethal in 2 out of 2 animals not previously exposed to the nitrile through drinking water. In young (4 week-old at the start of the study) females exposed to 10 mM of acetone cyanohydrin in the drinking water for 2 weeks, followed by 20 mM for 4–6 weeks, and finally fasted for 24 h before implantation of the osmotic pumps, lethality occurred by 24 h of exposure to 118 or 129 μmol/kg/h (n = 1/each), but not after exposure to 108 (n = 1) or 100 (n = 2) μmol/kg/h. The effect on acetone cyanohydrin toxicity of malnutrition was then explored in chronically exposed young female rats (2 weeks at 10 + 6 weeks at 20 mM) by substituting the food pellets with cassava starch for a number of days before minipump implantation. Whereas 2 animals on standard diet pellets survived 110 μmol/kg/h, cassava starch feeding for seven days resulted in increased lethality, with 3 out of 3 animals dying at 110 μmol/kg/h, 2 out of 2 dying at 100 μmol/kg/h, 1 out of 1 dying at 90 μmol/kg/h, and 1 out of 3 dying at 70 μmol/kg/h. When cassava starch feeding was introduced for the last three weeks
292
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
before implantation, mortality was recorded in one animal with final exposure to 60 μmol/kg/h, but 3 out of 4 animals survived after 24 h of exposure to 50 μmol/kg/h. These observations set the conditions selected for experiment use. 3.2. Effects of acetone cyanohydrin in exposure conditions mimicking those associated with konzo Female rats exposed from 4 weeks of age to 10 mM of acetone cyanohydrin in drinking water for two weeks followed by 20 mM for 7 weeks, and fed cassava starch for the last three of these weeks, did not show any overt neurological deficit. Subsequent exposure of these animals to subcutaneous osmotic pump delivery of 50 μmol/kg/h for 24 h, initiated after a 24 h period of food deprivation, caused behavioral effects indicative of acute toxicity, which were more evident during the last hours of exposure. These effects included immobility, decreased response to stimuli and abnormal walking postures with heightened hips; these abnormalities were present at notably different intensities in the different rats under acetone cyanohydrin infusion, and were not observed in animals in which pumps were implanted filled with control vehicle. No tremors, seizures and labored breathing were observed and no deaths occurred. One day after pump removal, 4 of the 6 acetone cyanohydrin rats were estimated to show a slightly abnormal posture, with more than normal extension of the hindpaws. One of the two rats with an apparently more abnormal behavior was sacrificed at day 2. The other rat of this pair was the only one out of five with an apparent abnormal walking posture at day 6. This second animal was processed for histology at day 7 along with a third acetone cyanohydrin rat and a saline minipump rat. Quantitative behavioral analysis showed no group differences between acetone cyanohydrin and saline rats at days 1 or 6 after minipump exposure (data not shown). In the open field, neither the treatment nor the treatment-by-day interaction factors were significant for either horizontal activity or rearing counts (all p's N 0.2). Similarly, gait analysis showed no effects of treatment on stepping patterns. Fore and hind limb stride length and stride width showed no treatment effect (all p's N 0.14) while significant treatment by day interactions for hind limb stride length (F(2,8) = 4.488, p = 0.049) and stride width (F(82,8) = 4.68, p = 0.045) were not followed by significant treatment effects at days 1 or 6 after exposure (all p's N 0.12). Among the rats exposed to acetone cyanohydrin, the one killed at day 2 after minipump removal, and one of the two killed at day 7 showed a positive fluoro-jade labeling in discrete brain areas, while the other rat (and the saline rat) showed no label in any brain region. In both labeled rats, the degeneration stain revealed neuronal profiles in a bilaterally symmetrical distribution. Some brain areas were labeled in both rats, while others were found only in the rat studied at day 7. Thus, in both rats, a consistent fluoro-jade staining occurred in discrete thalamic nuclei, namely the paraventricular and the ventral reuniens nuclei (Fig. 2A). In a few sections, an area lateral to the reuniens nucleus was also labeled, perhaps in the zona incerta or adjacent nuclei. In the rat studied at day 7, numerous labeled neurons were found across the whole dorsal endopiriform nucleus, spanning up to more than 10 mm in the rostro-caudal axis, with some involvement of adjacent areas of the lateral entorhinal cortex (Fig. 2B). 3.3. Effects of KCN One out of 5 young female rats exposed to 50 μmol/kg/h for 24 h of KCN died, while the other four animals showed no fluoro-jade B staining 7 days after osmotic pump exposure. After exposure to KCN at 60 μmol/kg/h, two out of three animals died, and the third one showed no fluoro-jade B staining at day 7 after dosing.
Fig. 2. Neuronal degeneration in the CNS of rats exposed to acetone cyanohydrin by minipump exposure (50 μmol/kg/h for 24 h). Fluoro-jade B stain of coronal vibratome sections (50 μm) of brain from a rat killed at 7 days after exposure. 3V = third ventricle. DEn = dorsal endopiriform nucleus. LEnt = lateral entorhinal cortex. PVA = paraventricular thalamic nucleus. rf = rhinal fissure. VRe = ventral reuniens nucleus.
4. Discussion The association of konzo with the consumption of insufficiently processed cassava has been firmly established [3], but the causative neurotoxic agent remains to be identified. The present study was carried out as a preliminary evaluation of the neurotoxicity of acetone cyanohydrin, the nitrile metabolite of linamarin, which is abundantly present in cassava flours associated with konzo. The results showed selective neuronal degeneration in a number of brain areas of some, but not all, of the rats exposed to acetone cyanohydrin in exposure conditions designed to mimic those associated with human konzo. The possibility that nitriles (organic compounds containing the cyano group) are the cause of konzo has received little consideration. Increasing evidence indicates that a number of nitriles cause a wide variety of neurotoxic effects, such as the neuronal degeneration caused in selected areas of the rat brain by trans-crotononitrile and 2,4-hexadienenitrile [7,37]. Although the main targets in the rat are
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
the inferior olive and the piriform cortex, degenerating neurons have also been observed in cortical areas [7], so it may be that a similar toxic action causes motor cortex damage in humans and accounts for the clinical signs in konzo. Both trans-crotononitrile and 2,4hexadienenitrile are α,β-unsaturated nitriles. Metabolism of these nitriles has been hypothesized to proceed by P450 mediated epoxidation of the double bound, followed by opening of the epoxide ring to form a cyanohydrin (α-hydroxy-nitrile), subsequently decomposing to release cyanide [38]. Therefore, we hypothesized that the α-hydroxy-nitrile end has a neurotoxic potential responsible for both human konzo and the neurotoxic effects of trans-crotononitrile and 2,4-hexadienenitrile in the rat, and focused on the most abundant cyanohydrin in cassava, acetone cyanohydrin, as a candidate causative agent for the disease. On initial evaluation, lethality occurred as expected at small acute doses, but drinking water exposure resulted in ingestion of large daily doses, up to several times the lethal dose given in a single bolus, with no apparent adverse neurological effects. Continuous exposure for one or seven days through osmotic minipump infusion also resulted either in death or in no overt evidence of neurological effects in survivors. When drinking water and minipump infusion exposure models were combined, rats previously exposed to acetone cyanohydrin via drinking water survived higher minipump doses than rats with no previous exposure. A possible explanation is that chronic acetone cyanohydrin exposure increases the activity of cyanide detoxification systems, allowing the animals to cope with higher cyanide amounts resulting from final high dose cyanohydrin exposure. We hypothesized that this increased survival might be important in generating a window of doses causing neurotoxicity but not death. Other conditions associated with development or increased risk of konzo are young age, female sex and malnutrition. To include these variables in the model, we chronically exposed female rats to acetone cyanohydrin through drinking water starting shortly after weaning, and included a period of cassava starch (tapioca) feeding. Successive introduction of these conditions required adjustments in the maximal sublethal doses that could be administered by minipump infusion as the candidate final precipitating event. It was particularly striking that, after 3 weeks on tapioca, lethal effects were recorded at doses half of those that were lethal in rats fed standard diet pellets throughout the whole period. This supports the widely diffused hypothesis [2] that cassava toxicity may depend on the nutritional status of the subject. Based on the preliminary observations, we designed an experiment to test the hypothesis that minipump exposure to sublethal acetone cyanohydrin doses induces neurotoxic effects in female rats “pre-conditioned” with chronic exposure to the compound, malnutrition, and final acute fasting to induce high body acetone concentrations. Under these exposure conditions, we observed overt signs of toxicity in the rats exposed to high doses of ACH by minipump infusion, and obtained proof of ongoing neuronal degeneration in some, but not all, of these animals, while no deaths were recorded. The animal showing the worst condition was sacrificed and processed for histology at day 2 to limit its suffering, but it was not certain that it would have died without our intervention. This is in contrast with the effects of minipump exposure to equal or slightly higher molar doses of KCN, which resulted in either death or complete absence of neuronal degeneration as assessed by fluoro-jade B staining. These results thus support our initial hypothesis that the particular exposure conditions included in the model would allow for the appearance of neurotoxic effects in the absence of subsequent mortality. However, subsequent experiments in the mouse have however not supported the hypothesis that high blood acetone concentrations play a key role in this toxicity. Fasted CYP2E1-null mice, which are known to accumulate higher acetone concentrations [8], showed similar susceptibility to acetone cyanohydrin acute toxicity to wild type animals (unpublished data).
293
The final exposure to acetone cyanohydrin through minipump infusion resulted in a number of changes in behavior that were compatible with either general illness or motor effects. However, a marked recovery occurred in most animals after the 24 h exposure period, and in fact no consistent motor effects could be demonstrated by quantitative assessment at 1 and 6 days after exposure. Although the methods used were simple, and perhaps more refined methods could have revealed group differences, no major alterations in motor control occurred. Konzo has been hypothesized to involve damage of the cortico-spinal tract [43], which differs substantially between humans and rats [22], so the apparent lack of a measurable motor effect in the acetone cyanohydrin rats would not constitute by itself definite proof of the lack of relevance of this rat model to human konzo. Damage to the cortico-spinal system is hypothesized in konzo [43], but no pathological data are available and it may be the case that other motor-related areas are involved. Species-dependent differences in lesion sites may also appear even though a common mechanism of action acts across species. Therefore, we did not search for lesions in pre-selected regions, but instead used fluoro-jade B staining to screen for degenerating neurons in the whole of the central nervous system. The effectiveness of this stain in revealing neuronal degeneration, and the simplicity of its use, allows for a complete evaluation of the system and the identification of neuronal targets without the need for a previous hypothesis of where the damage may be [7,35,36]. The data revealed the presence of degenerating neurons in 2 out of 3 rats examined. The lesions recorded were bilaterally symmetrical and coincided in the two affected animals, although one rat had degenerating neurons in more areas than the other; these findings support the conclusion that the lesions observed were the expression of a selective pattern of neuronal degeneration. However, the pattern observed was not the one anticipated in an animal model of konzo, which would be expected to involve the motor cortex and/or other motor-related areas. Nor did the observed pattern match the one presented by the nitriles with CNS toxicity, trans-crotononitrile and hexadienenitrile, which mainly involve the inferior olive and the piriform cortex [7,37]. The roles of the paraventricular and reuniens nuclei, apparently the main acetone cyanohydrin targets, are not precisely defined, but most of the evidence relates these structures to limbic-cognitive or visceral/autonomic functions [28,48]. Calculations made on the basis of data in the literature [3,10,44] give an estimation of maximal acetone cyanohydrin doses in konzo cases in the range of 10 to 100 μmol/kg/day, which were effectively exceeded by our final 50 μmol/kg/h dosing schedule. This makes it unlikely that the mismatch between the neurotoxic effects recorded in the present model and those expected in a useful animal model of konzo is due to a failure to attain large enough doses. Degeneration of the paraventricular and reuniens nuclei and the dorsal endopiriform nucleus has been observed in rats administered single intraseptal injections of kainic acid as an animal model of status epilepticus [33]. In that model, evidence for degeneration of the dorsal endopiriform nucleus was observed at 7 days after exposure, while the thalamic nuclei were affected after one day. As we observed all these areas in the affected animal at 7 days, but at day 2 the dorsal endopiriform nucleus was spared, it is tempting to speculate that similar mechanisms may be involved in these models. It is thus possible that our exposure model caused lesions through kainate-like convulsant activity. However, the status epilepticus induced by intraseptal kainic injection typically causes hippocampal lesions [33], and the hippocampus was spared in the acetone cyanohydrin rats. And although our assessment was not extensive enough to rule out the presence of convulsive activity, we did not observe status epilepticus in our rats during handling and behavioral testing. The value of the present observations in the development of an animal model of konzo remains uncertain. The data gathered do not provide a useful animal model for konzo, and thus do not support the notion that acetone cyanohydrin is the causal agent, although they do
294
C. Soler-Martín et al. / Neurotoxicology and Teratology 32 (2010) 289–294
not conclusively disprove the hypothesis either. Our observation, for the first time, of selective neuronal degeneration in some animals exposed to acetone cyanohydrin supports several of the starting assumptions, the main one being that acetone cyanohydrin neurotoxicity may, under particular exposure conditions, differ from that elicited by simple acute or chronic cyanide exposure. While this would call for additional research, a major difficulty is posed by the fact that sublethal doses, associated with overt systemic toxicity, appeared to be necessary for the neurotoxic effects, while konzo typically appears as a neurotoxic disease involving little or no systemic toxicity. One recent observation that challenges the hypothesis further is the finding that, at least for the sensory toxicity of allylnitrile, the corresponding cyanohydrin is unlikely to be the bioactivated toxic metabolite of the parent nitrile [6]. Conflict of interest We have no conflicting interest to declare. Acknowledgements We thank our students Fabian Márquez and Eduardo Balbuena for their contribution to preliminary experiments. This work was supported by grants BFI2003-01606 and BFU2006-00343/BFI from the Spanish Ministry of Science and Innovation/EU FEDER and 2005SGR00022 from the Generalitat of Catalonia.
References [1] A. Akintonwa, O. Tunwashe, A. Onifade, Fatal and non-fatal acute poisoning attributed to cassava-based meal, Acta Hortic. 375 (1994) 285–288. [2] Anonymous, Konzo — a distinct type of upper motoneuron disease, Week. Epidem. Rec. WHO 71 (1996) 225–228. [3] J.P. Banea-Mayambu, T. Tylleskär, N. Gitebo, N. Matadi, M. Gebre-Medhin, H. Rosling, Geographical and seasonal association between linamarin and cyanide exposure from cassava and the upper motor neurone disease konzo in former Zaire, Trop. Med. Int. Health. 2 (1997) 1143–1151. [4] R.S. Bitner, A. Kanthasamy, G.E. Isom, G.K.W. Yim, Seizures and selective CA-1 hippocampal lesions induced by an excitotoxic cyanide metabolite, 2-iminothiazolidine-4-carboxylic acid, Neurotoxicology 16 (1995) 115–122. [5] P. Boadas-Vaello, E. Jover, N. Díez-Padrisa, J.M. Bayona, J. Llorens, Differential role of CYP2E1-mediated metabolism in the lethal and vestibulotoxic effects of ciscrotononitrile in the mouse, Toxicol. Appl. Pharmacol. 225 (2007) 310–317. [6] P. Boadas-Vaello, E. Jover, S. Saldaña-Ruíz, C. Soler-Martín, C. Chabbert, J.M. Bayona, J. Llorens, Allylnitrile metabolism by CYP2E1 and other CYPs leads to distinct lethal and vestibulotoxic effects in the mouse, Toxicol. Sci. 107 (2009) 461–472. [7] P. Boadas-Vaello, J. Riera, J. Llorens, Behavioral and pathological effects in the rat define two groups of neurotoxic nitriles, Toxicol. Sci. 88 (2005) 456–466. [8] F.Y. Bondoc, Z. Bao, W.Y. Hu, F.J. Gonzalez, Y. Wang, C.S. Yang, J.Y. Hong, Acetone catabolism by cytochrome P450 2E1: studies with CYP2E1-null mice, Biochem. Pharmacol. 58 (1999) 461–463. [9] J.L. Borowitz, G.E. Isom, D.V. Nakles, Human toxicology of cyanide, in: D.A. Dzombak, R.S. Ghosh, G.M. Wong-Chong (Eds.), Cyanide in Water and Soil: Chemistry, Risk and Management, CRC Press-Taylor and Francis Group, Boca Raton, 2006, pp. 237–250. [10] L. Carlsson, N. Mlingi, A. Juma, G. Ronquist, H. Rosling, Metabolic fates in humans of linamarin in cassava flour ingested as stiff porridge, Food Chem. Toxicol. 37 (1999) 307–312. [11] L. Chiwona-Karltun, J. Mkumbira, J. Saka, M. Bovin, N.M. Mahungu, H. Rosling, The importance of being bitter — a qualitative study on cassava cultivar preference in Malawi, Ecol. Food Nutr. 37 (1998) 219–245. [12] L. Chiwona-Karltun, T. Tylleskär, J. Mkumbira, M. Gebre-Medhin, H. Rosling, Low dietary cyanogen exposure from frequent consumption of potentially toxic cassava in Malawi, Int. J. Food Sci. Nutr. 51 (2000) 33–43. [13] S.M. Chou, H.A. Hartmann, Axonal lesions and waltzing syndrome after IDPN administration in rats. With a concept — axostasis, Acta Neuropathol. 3 (1964) 428–450. [14] J. Cliff, P. Lundqvist, J. Mårtensson, H. Rosling, B. Sörbo, Association of high cyanide and low sulphur intake in cassava-induced spastic paraparesis, Lancet 8466 (1985) 1211–1213. [15] J. Cliff, D. Nicala, F. Saute, R. Givragy, G. Azambuja, A. Taela, L. Chavane, J. Howarth, Konzo associated with war in Mozambique, Trop. Med. Int. Health. 2 (1997) 1068–1074. [16] Committee on Acute Exposure Guideline Levels, Acetone cyanohydrin, National Research Council Committee on Acute Exposure Guideline Levels, Acute Exposure Guidelines for Selected Airbone Chemicals, vol 7, National Academies Press, Washington, DC, 2009, pp. 13–39.
[17] E.E. Conn, Cyanogenesis — a personal perspectiva, Acta Hortic. 375 (1994) 31–43. [18] R.D. Cooke, An enzymatic assay for the total cyanide content of cassava (Manihot esculenta Crantz), J. Sci. Food. Agric. 29 (1978) 345–352. [19] J.R. Froines, E.M. Postlethwait, E.J. LaFuente, W.C.V. Liu, In vivo and in vitro release of cyanide from neurotoxic aminonitriles, J. Toxicol. Environ. Health 16 (1985) 449–460. [20] W.P. Howlett, G.R. Brubaker, N. Mlingi, H. Rosling, Konzo, an epidemic upper motor neuron disease studied in Tanzania, Brain 113 (1990) 223–235. [21] F.R. Johannsen, G.J. Levinskas, Relationships between toxicity and structure of aliphatic nitriles, Fundam. Appl. Toxicol. 7 (1986) 690–697. [22] R.N. Lemon, Descending pathways in motor control, Annu. Rev. Neurosci. 31 (2008) 195–218. [23] J. Llorens, D. Demêmes, A. Sans, The behavioral syndrome caused by 3, 3′iminodipropionitrile and related nitriles in the rat is associated with degeneration of the vestibular sensory hair cells, Toxicol. Appl. Pharmacol. 123 (1993) 199–210. [24] Ministry of Health, Mozambique, Mantakassa: an epidemic of spastic paraparesis associated with chronic cyanide intoxication in a cassava staple area of Mozambique. 1. Epidemiology and clinical and laboratory findings in patients, Bull. World Hlth. Organ. 62 (1984) 477–484. [25] N. Mlingi, N.H. Poulter, H. Rosling, An outbreak of acute intoxications from consumption of insufficiently processed cassava in Tanzania, Nutr. Res. 12 (1992) 677–687. [26] M. Orfila, Traité des poisons tirés des regnes minéral, végétal et animal, ou toxicologie générale, considérée sous les rapports de la physiologie, de la pathologie et de la médicine légale, Third Ed.Crochard et Gabon Editeurs, Paris, 1826. [27] A.J. Parker, K.A. Clarke, Gait topography in rat locomotion, Physiol. Behav. 48 (1990) 41–47. [28] M.P. Parsons, S. Li, G.J. Kirouac, Functional and anatomical connection between the paraventricular nucleus of the thalamus and dopamine fibers of the nucleus accumbens, J. Comp. Neurol. 500 (2007) 1050–1063. [29] G. Paxinos, P. Carrive, H. Wang, P-Y. Wang, Chemoarchitectonic atlas of the rat brainstem, Academic Press, San Diego, 1999. [30] G. Paxinos, L. Kus, K.W.S. Ashwell, C. Watson, Chemoarchitectonic Atlas of the Rat Forebrain, Academic Press, San Diego, 1999. [31] A. Pestronk, N, N′-dimethylaminopropionitrile, in: P.S. Spencer, H.H. Schaumburg, A.C. Ludolph (Eds.), Experimental and clinical neurotoxicology, Second Ed., Oxford University Press, New York, 2000, pp. 494–502. [32] D.J. Philbrick, D.C. Hill, J.C. Alexander, Physiological and biochemical changes associated with linamarin administration to rats, Toxicol. Appl. Pharmacol. 42 (1977) 539–551. [33] M. Revuelta, A. Castaño, J.L. Venero, A. Machado, J. Cano, Long-lasting induction of brain-derived neurotrophic factor is restricted to resistant cell populations in an animal model of status epilepticus, Neuroscience 103 (2001) 955–969. [34] N.L. Rosenberg, J.A. Myers, W.R. Martin, Cyanide-induced parkinsonism: clinical, MRI, and 6-fluorodopa PET studies, Neurology 39 (1989) 142–144. [35] L.C. Schmued, K.J. Hopkins, Fluoro-Jade B: a high affinity fluorescent marker for the localization of neuronal degeneration, Brain Res. 874 (2000) 123–130. [36] L.C. Schmued, K.J. Hopkins, Fluoro-Jade: novel fluorochromes for detecting toxicant-induced neuronal degeneration, Toxicol. Pathol. 28 (2000) 91–99. [37] A. Seoane, R. Apps, E. Balbuena, L. Herrero, J. Llorens, Differential effects of transcrotononitrile and 3-acetylpyridine on inferior olive integrity and behavioural performance in the rat, Eur. J. Neurosci. 22 (2005) 880–894. [38] E.H. Silver, S.H. Kuttab, T. Hasan, M. Hassan, Structural considerations in the metabolism of nitriles to cyanide in vivo, Drug Metab. Dispos. 10 (1982) 495–498. [39] P.S. Spencer, Food toxins, AMPA receptors, and motor neuron diseases, Drug Metab. Rev. 31 (1999) 561–587. [40] I. Swenne, U.J. Eriksson, R. Christoffersson, B. Kågedal, P. Lundquist, L. Nilsson, T. Tylleskär, H. Rosling, Cyanide detoxification in rats exposed to acetonitrile and fed a low protein diet, Fundam. Appl. Toxicol. 32 (1996) 66–71. [41] J. Tor-Agbidye, V.S. Palmer, M.R. Lasarev, A.M. Craig, L.L. Blythe, M.I. Sabri, P.S. Spencer, Bioactivation of cyanide to cyanate in sulfur amino acid deficiency: relevance to neurological disease in humans subsisting on cassava, Toxicol. Sci. 50 (1999) 228–235. [42] J. Tor-Agbidye, V.S. Palmer, P.S. Spencer, A.M. Craig, L.L. Blythe, M.I. Sabri, Sodium cyanate alters glutathione homeostasis in rodent brain: relationship to neurodegenerative diseases in protein-deficient malnourished populations in Africa, Brain Res. 820 (1999) 12–19. [43] D. Tshala-Katumbay, K.E. Eeg-Olofsson, T. Kazadi-Kayembe, T. Tylleskär, P. Fällmar, Analysis of motor pathway involvement in konzo using transcranial electrical and magnetic stimulation, Muscle Nerve. 25 (2002) 230–235. [44] T. Tylleskär, M. Banea, N. Bikangi, R.D. Cooke, N.H. Poulter, H. Rosling, Cassava cyanogens and konzo, an upper motoneuron disease found in Africa, Lancet 339 (1992) 208–211. [45] T. Tylleskär, M. Banea, N. Bikangi, G. Nahimana, L.A. Persson, H. Rosling, Dietary determinants of a non-progressive spastic paraparesis (Konzo): a case-referent study in a high incidence area of Zaire, Int. J. Epidemiol. 24 (1995) 949–956. [46] T. Tylleskär, W.P. Howlett, H.T. Rwiza, S.M. Aquilonius, E. Stålberg, B. Lindén, A. Mandahl, H.C. Larsen, G.R. Brubaker, H. Rosling, Konzo: a distinct disease entity with selective upper motor neuron damage, J. Neurol. Neurosurg. Psychiatry. 56 (1993) 638–643. [47] T. Tylleskär, F.D. Légué, S. Peterson, E. Kpizingui, P. Stecker, Konzo in the Central African Republic, Neurology 44 (1994) 959–961. [48] R.P. Vertes, Differential projections of the infralimbic and prelimbic cortex in the rat, Synapse 51 (2004) 32–58. [49] M. Zagrobelny, M. Soren, B.L. Moller, Cyanogenesis in plants and arthropods, Phytochemistry 69 (2008) 1457–1468.