Bioactivation of cyanide to cyanate in sulfur amino acid deficiency ...

Download PDF
17 downloads 99906 Views 127KB Size Report
Comment
Aug 1, 1999 - Abstract. Neurological disorders have been reported from parts of Africa with protein-deficient populations and attributed to cyanide (CN-) ...
50, 228 –235 (1999) Copyright © 1999 by the Society of Toxicology

TOXICOLOGICAL SCIENCES

Bioactivation of Cyanide to Cyanate in Sulfur Amino Acid Deficiency: Relevance to Neurological Disease in Humans Subsisting on Cassava John Tor-Agbidye,* Valerie S. Palmer,† Michael R. Lasarev,* A. Morrie Craig,‡ Linda L. Blythe,‡ Mohammad I. Sabri,* and Peter S. Spencer* ,1 *Center for Research on Occupational and Environmental Toxicology, Oregon Health Sciences University, Portland, Oregon; †Third World Medical Research Foundation, Portland, Oregon; and ‡College of Veterinary Medicine, Oregon State University, Corvallis, Oregon Received March 30, 1998; accepted October 16, 1998

Neurological disorders have been reported from parts of Africa with protein-deficient populations and attributed to cyanide (CN –) exposure from prolonged dietary use of cassava, a cyanophoric plant. Cyanide is normally metabolized to thiocyanate (SCN –) by the sulfur-dependent enzyme rhodanese. However, in proteindeficient subjects where sulfur amino acids (SAA) are low, CN – may conceivably be converted to cyanate (OCN –), which is known to cause neurodegenerative disease in humans and animals. This study investigates the fate of potassium cyanide administered orally to rats maintained for up to 4 weeks on either a balanced diet (BD) or a diet lacking the SAAs, L-cystine and L-methionine. In both groups, there was a time-dependent increase in plasma cyanate, with exponential OCN – increases in SAA-deficient rats. A strongly positive linear relationship between blood CN – and plasma OCN – concentrations was observed in these animals. These data are consistent with the hypothesis that cyanate is an important mediator of chronic cyanide neurotoxicity during protein-calorie deficiency. The potential role of thiocyanate in cassava-associated konzo is discussed in relationship to the etiology of the comparable pattern of motor-system disease (spastic paraparesis) seen in lathyrism. Key Words: cassava; cyanide; cyanate; metabolism; sulfur amino acid deficiency; sulfate; thiocyanate; neurodegenerative disease.

An estimated half-billion people in tropical and sub-tropical climes consume the processed roots and leaves of the cyanophoric plant cassava (Manihot esculenta) (Rosling, 1996). In certain parts of Africa, cassava is the main source of food for populations prone to protein-calorie deficiency. The combination of dietary cyanide and low protein intake has been implicated in outbreaks of neurological disease in these populations (Roman et al., 1985). One form of cassava-related disease is a slowly developing ataxic myeloneuropathy originally described in Nigeria (Osuntokun, 1968, 1973, 1981); the other is a sub-acute disease manifest principally by spastic paraparesis 1

To whom all correspondence should be addressed at the Center for Research on Occupational and Environmental Toxicology, L606, Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Road, Portland, Oregon 97201. Fax: (503) 494-4278. E-mail: [email protected].

(konzo) (Cliff et al., 1985, Howlett et al., 1990, Rosling, 1996; Tylleska¨r et al., 1992, 1994). The development of these syndromes is hypothesized to depend on (a) the amount and duration of exposure to dietary cyanide, and (b) the ability of the body to detoxify cyanide, a function that may vary with nutritional status. Cassava-associated neurologic disease has been reported throughout southern Africa (except South Africa) and parts of central and western Africa (Rosling and Tylleska¨r, in press). Free cyanide must be sequestered and metabolized to avoid inhibition of cytochrome c oxidase, blockade of mitochondrial electron transport and consequent energy failure. Following an acute exposure, cyanide is reportedly first trapped by methemoglobin in the form of cyanomethemoglobin (Schultz, 1984). Cyanide is converted to thiocyanate (SCN – ), a reaction that requires sulfane sulfur as a ratelimiting cofactor for the enzyme rhodanese (Lundquist, 1992). The concentration of sulfane sulfur is dependent on the availability of sulfur amino acids (SAA) from dietary protein (Cliff et al., 1985). Even in protein malnutrition, available sulfur is preferentially utilized for cyanide detoxication (Swenne et al., 1996). Cyanide also may be sequestered by albumin and metabolized to 2-aminothiazoline-4carboxylic acid (ATC) (Bitner et al., 1995, 1997; Lundquist et al., 1995a) or to cyanate (OCN – ) which (Swenne et al., 1996), in turn, is converted by the cysteine-containing enzyme cyanase [E.C. 3.5.5.3] ammonia and bicarbonate (Schultz, 1949). Several investigators have reported that prolonged cyanate treatment induces peripheral neuropathy in humans and rodents, and spastic paraparesis in macaques (Ohnishi et al., 1975; Shaw et al., 1974; Tellez 1979; Tellez-Nagel et al., 1977). That dietary CN – must be converted to OCN – for neurological disease to develop in cassava-dependent populations is therefore a plausible proposal. The present study employs a defined model of SAA deficiency (Tor-Agbidye et al., 1998) to test the hypothesis that conversion of CN – to OCN – is favored in this abnormal nutritional state. Whereas neurological disease develops in humans and animals after months of exposure to cassava or

228

229

BIOACTIVATION OF CYANIDE TO CYANATE

treatment with cyanate, respectively, the present study focuses on the early metabolic events that may trigger nervous system degeneration.

TABLE 1 Amino Acid Compositions of Rat Chow, the Balanced Diet and the Sulfur Amino Acid (SAA)-Free Diet Amino acid

Rat chow

Balanced diet

SAA-free diet

L-Alanine L-Arginine L-Asparagine L-Aspartic acid L-Cystine L-Glutamic acid Glycine L-Histidine L-Isoleucine L-Leucine L-Lysine L-Methionine L-Phenylalanine L-Proline L-Serine Taurine L-Threonine L-Tryptophan L-Tyrosine L-Valine

1.44% 1.38% – 2.83% 0.32% 4.54% 1.20% 0.55% 1.18% 1.70% 1.42% 0.43% 1.03% 1.55% 1.21% 0.02% 0.91% 0.29% 0.68% 1.21%

0.35% 1.21% 0.6% 0.35% 0.35% 4.0% 2.33% 0.45% 0.82% 1.11% 1.80% 0.82% 0.75% 0.35% 0.35% – 0.82% 0.18% 0.5% 0.82%

0.35% 1.21% 0.6% 0.35% – 4.0% 2.33% 0.45% 0.82% 1.11% 1.80% – 0.75% 0.35% 0.35% – 0.82% 0.18% 0.5% 0.82%

MATERIALS AND METHODS Chemicals. Analytical-grade perchloric acid, sodium hydroxide, acid silver sulfate, barbituric acid, sulfuric acid and potassium cyanide; barium chloride, polyethylene glycol 600, sodium sulfate, sodium hypochlorite (Sigma, St. Louis, MO), potassium thiocyanate (Baker, Phillipsburg, MJ), and sodium perchlorate (EM Science, Gibbstown) were used. Experimental design. The experiment was designed to assess the effect and possible interaction between diet (SAA-free and BD) and time (1, 2, 3, and 4 weeks) in the presence of orally administered potassium cyanide solution. Four rats were randomly allocated to each of the 8 possible treatment combinations (a 2 3 4 factorial experiment in a completely randomized design with replication). Analysis of variance was used on all responses as a first step, and if time could be modeled as a continuous covariate without disrupting the quality of their fit, multiple regression was then applied. All analyses were done using S-Plus 3.3 (Math Soft, Inc., Seattle, WA). Diets and treatment. Sprague-Dawley rats (250 –300g) were obtained from Bantin and Kingman, Inc. Seattle, WA. Female animals were selected because cassava-related neurological disease primarily impacts women of childbearing age (Rosling, 1996). Rats were housed in standard cages in a temperature-controlled environment on a 12h/12h light-dark cycle. Rats were allowed to acclimate (one rat per cage); after 7 days of acclimatization on standard rat chow, animals were randomly selected, placed individually in metabolic cages overnight, and 24-h urine was collected to determine baseline values for the concentration of thiocyanate and inorganic sulfate. Blood was collected intracardially (n, 4) and separated into 2 portions; one for determination of baseline values for cyanide and the second portion was centrifuged at 2500 3 g for 10 min at 4°C, the plasma collected, and baseline concentrations of cyanate determined. Rats were randomly divided into 2 groups (4 rats per week for up to 4 weeks, 16 per group), kept individually in regular cages and fed 1 of 2 closely matched experimental diets (Harlan Teklad, Madison, WI. The composition of the sulfur amino acid (SAA)-free diet is given in Table 1; the balanced diet (BD) additionally contained L-cystine (0.35%) and L-methionine (0.82%). One hundred mL of double-distilled water containing potassium cyanide adjusted to a target daily dose of 50 mg/kg body weight was administered to all rats ad libitum; every 2 days, consumption was measured and a fresh solution of cyanide provided. Potassium cyanide solution was stable for up to 48 h. Rats were maintained for 1-, 2-, 3-, or 4-week periods. The health of the rats was observed daily and their body weights recorded weekly. At the end of each week, 8 rats (4 from each group) were transferred from their home cages into individual metabolic cages. Urine samples were collected from each rat at a uniform time (10:00 A.M.) and maintained at –20°C until they were analyzed for inorganic sulfate and thiocyanate. Prior to the selective termination of animals at week 1 (n, 8), week 2 (n, 8), week 3 (n, 8), and week 4 (n, 8), blood samples were collected by cardiac puncture and divided into 2 parts: one part was treated with acidified silver sulfate and frozen at –20°C for cyanide analysis; the other was centrifuged at 2500 3 g for 10 min and the plasma separated and frozen at – 84°C for cyanate analysis. Urinary inorganic sulfate. Urinary inorganic sulfate was determined spectrophotometrically by the method of Lundquist et al. (1980). The method is based on a turbidity estimation of insoluble barium sulfate formed from the reaction between soluble barium chloride and urinary sulfate. Turbidity is measured spectrophotometrically at 600 nm. Aliquots of urine were diluted to 3 mL with double-distilled water; 1 mL HCl, 0.5 mol/L, and 1 mL barium polyethylene glycol were added to the solution. A reagent blank was prepared in which urine was replaced with an equal volume of double-distilled water. All assays were performed in duplicate. The amount of inorganic sulfate in the

samples was read against a standard curve with known concentrations of inorganic sulfate. Data were expressed as mmol inorganic sulfate/mL. Urinary thiocyanate. Urinary thiocyanate levels were determined spectrophotometrically by the improved method of Lundquist et al. (1995b). The method determines the absorbance of blue color formed from the reaction between the color reagent (a mixture of isonicotinic acid and 1,3-dimethylbarbituric acid) and urinary thiocyanate. The intensity of the color is measured at 607 nm. Briefly, aliquots of urine (500 mL) diluted with 5.0 mL of 1M NaOH were applied to columns (2.5 3 0.7 cm) of Amberlyst A-21. Columns were washed 3 times with 5 mL of double-distilled water before eluting thiocyanate by the addition of 8 mL of 1 M sodium perchlorate. Aliquots of 4 mL of eluate were acidified with 0.2 mL of 0.35M acetic acid, chlorinated for 2 min with 0.1 mL of 50 mM sodium hypochlorite, and 0.6 mL of the color reagent added to the mixture. A reagent blank was prepared in which urine was replaced with double-distilled water. Assays were performed in duplicate. Thiocyanate concentrations in the samples were read against a standard curve with known concentrations of potassium thiocyanate. Data were expressed as nmol thiocyanate/mL. Blood cyanide. Blood cyanide concentrations were estimated spectrophotometrically by the method of Lundquist et al. (1985). This method is based on the Konig reaction, which produces a chromophore from cyanide. This method employs a simple aeration apparatus consisting of 2 side-arm tubes (150 3 25mm and 150 3 20mm fitted with a 1-hole rubber stopper carrying a 7-mm glass inlet tube) connected by a short length of rubber latex tubing. Aliquots of blood (1.0 mL) treated with 10 mL of 20-mM acid silver sulfate were placed into the large side-arm tubes along with 5 drops of anti-foam agent. Four mL of 0.1 M NaOH was placed in the smaller side-arm tube and 5 mL of 11 M H 2 SO 4 added to the sample. The liberated HCN was transferred to a sodium hydroxide solution by passing nitrogen (purified by passage through a gas washing bottle containing 0.1 M NaOH) at a flow rate of 0.5 L/min for 2 h. Acetic acid, 0.4 mL 2 M, was added to the NaOH solution and mixed briefly before adding 0.1 mL 50 mM sodium hypochlorite. The solution was mixed again, 0.5 mL of barbituric acidpyridine reagent added after 1 min, and absorbancy at 580 nm determined 5–15 min later. A blank (10 mL acid silver sulfate reagent) was prepared,

230

TOR-AGBIDYE ET AL.

RESULTS

Feed Intake Rats were fed 1 of 2 experimental diets for 1, 2, 3, or 4 weeks. Rats on the BD ate an average of 108 g/week (95% confidence interval from 95.6 to 120.3 g) while those animals on the SAA-free diet consumed an average of 44.7 g less per week [t (degrees of freedom 28) 5 –5.26 2-sided p value , 0.001] (Fig. 2). At week 4, cumulative feed intake in rats on BD was 506.8 6 13.7 while SAA-deficient rats consumed only 300.0 6 20.5 g of diet. There was general weakness in animals on the SAA-free diet, and rats in this group excreted a brownish-yellow material (porphyrins) around their necks and were hyper-responsive to noise. KCN Intake

FIG. 1. A typical chromatograph of plasma from rats maintained on the SAA-free diet for 3 weeks. One mL of plasma was derivatized to form compound Q and subjected to HPLC as described in Materials and Methods. The retention time (15.03 min) of the peak is comparable to that for authentic compound Q. Other small peaks were not identified and may represent plasma contaminants. Plasma cyanate concentrations were determined from a standard curve prepared as described in Materials and Methods.

and cyanide concentrations in the samples were read against a standard curve with known concentrations of potassium cyanide. Cyanide concentrations were expressed in mmol/L. Determination of cyanate. Plasma cyanate concentrations were determined by the method of Lundquist et al. (1993). Cyanate was converted to compound Q (2,4[1H, 3H]-quinazolinedione) and quantified by reversephase, high-performance liquid chromatography (HPLC). The authenticity of compound Q was verified spectrofluorometrically by excitation and emission wavelengths, retention times on HPLC (15.03 min.) (Fig. 1), and by gas chromatography mass spectrometry (GC-MS) (92, 119 and 162 mass ions). To derivatize cyanate to Compound Q, one mL of plasma was diluted with 0.5 mL of 0.2M sodium phosphate buffer (pH 7.4) and the pH adjusted to 4.7 with 4M acetic acid. This mixture was reacted with excess 40 mM 2-aminobenzoic acid (pH adjusted to 4.7 with a few drops of sodium acetate) and incubated for 10 min at 40°C. The reaction was stopped by placing the solution on ice. Cyclization of the intermediate product was performed by addition of 6.0 M sulfuric acid and heating at 100°C for 2 min, followed by cooling on ice. Compound Q was extracted thrice with ethyl acetate; the organic phases were pooled and dried under nitrogen at 40°C. The dried product was reconstituted with 2 mL of HPLC-grade water, acidified with 60 ml 5.0 M hydrochloric acid, and transferred onto PRS columns to remove excess 2-aminobenzoic acid. The column was washed 23 with 2-mL portions of HPLC-grade water, 0.5 mL 0.1 M sodium phosphate buffer (pH 7.4) was added, pH was adjusted to 6 – 8 with 3 M sodium hydroxide, and the combined effluent applied to a Sepak C-18 column. The column was washed with 3 mL HPLC-grade water, compound Q was eluted with 3 mL methanol, dried at 60°C under nitrogen, dissolved in methanol:water (1:2) and subjected to HPLC. Plasma cyanate concentrations in the samples were calculated from a standard curve prepared with known concentrations of potassium cyanate. Cyanate concentrations were expressed in nmol/L.

Cumulative intake of KCN-laced water in animals on BD or SAA-free diet was similar at day 25 (Fig. 3). Total KCN intake was calculated at 9, 15, 23, and 25 days after the initiation of the experiment and can be modeled as a quadratic function of the elapsed time. For rats on BD, the quadratic relationship is concave down (1-sided p value, 0.0088) although the maximum value for the function is not observed until day 25 of the experiment. Rats on an SAA-free diet show a weak quadratic relationship that is concave up (1-sided p value, 0.109). At day 9, there was no difference in mean amount of KCN intake accumulated up to that point (2-sided p value, 0.681). At day 15, the group of rats on the balanced diet was estimated to be drinking a cumulative average of 110 mL more than animals on SAA-free diet. By day 23, the difference had lessened to only 77.4 mL. At termination of the experiment (i.e., day 25), there was no significant difference between the two groups of animals (2-sided p value, 0.235).

FIG. 2. Cumulative food intake in rats fed the balanced diet (BD) or the SAA-free diet. At time point 0, rats were transferred from standard rat chow to 1 of 2 customized diets: BD or SAA-free diet. Measurements of food intake for the 4 rats per group were taken approximately 1, 2, 3, and 4 weeks past time point 0.

231

BIOACTIVATION OF CYANIDE TO CYANATE

TABLE 2 Total Urine Output (mL/24 h) Weeks on diet

Balanced diet

SAA-free diet

1 2 3 4

6.860.8 5.260.3 6.860.5 5.560.3

2.560.6 2.360.3 2.860.5 3.560.3

Note. Rats were fed rat chow, a balanced diet, or a sulfur amino acid (SAA)-free diet. Urine was collected individually from 8 rats at a uniform time weekly, as described in Materials and Methods. Values are means 6 SEM.

FIG. 3. Cumulative potassium cyanide intake in rats fed the BD or the SAA-free diet. The concentration of KCN in drinking water was adjusted for the average body weight of rats at each time point: Week 1, n 5 32; Week 2, n 5 24; Week 3, n 5 16; and Week 4, n 5 8. Rats on either diet show a cumulative cyanide intake of the SAA-free diet that is basically linear (1-sided p value, 0.109, test of quadratic term).

3.31 mL greater urine output than rats fed an SAA-free diet (95% confidence interval from 2.63 to 3.99 mL). Urine output was not affected by duration of treatment [F 5 1.87 on 3 and 24 degrees of freedom (df), p value, 0.162] but was highly influenced by the diet regimen (F 5 101.5 on 1 and 24 df, p value ,, 0.001).

Body Weight

Urinary Inorganic Sulfate

A significant difference in mean rates of weight gain was found between rats on the 2 diets after accounting for baseline body weight [t(27) 5 8.49 2-sided p value , 0.001) (Fig. 4). Rats on the BD experienced an average increase of 11.8 g/week (95% confidence interval from 7.28 to 16.37 g), while animals on the SAA-free diet lost body weight at a mean rate of 14.3 g/week (95% confidence interval from 9.9 to 18.7 g).

Inorganic sulfate in urine samples was measured in animals (on BD or SAA-free diets) receiving KCN in drinking water for 1, 2, 3, or 4 weeks. For control animals fed chow, the baseline value at week 0 (n, 8) for urinary sulfate was 330 6 15 mmol inorganic sulfate/mL urine (Fig. 5). Urinary sulfate in rats fed BD had more than doubled by Week 1 and remained at this level up to Week 4. By contrast, urinary sulfate in SAA-free animals had dropped precipitously by Week 1 and continued to approach zero through Week 1. Analysis of urinary sulfate showed that rats on the BD maintained a constant median value of 752 mmol inorganic sulfate/mL over the 4 weeks. At the same time, those animals on SAA-free diet

Urine Output Total urine output was greater at each time point in animals fed the BD (Table 2). Generally, rats on the BD had a mean of

FIG. 4. Body weight changes in rats fed the BD or the SAA-free diet. Rats were maintained on experimental diets for 1-, 2-, 3- or 4-week periods. [The number of rats therefore declined weekly from 32 (Week 0) to 8 (Week 4)]. Positive and negative values reflect increased and decreased body weights relative to baseline values at time 0. Body weights of rats on the BD and the SAA-free diets differ significantly (2-sided p value , 0.001).

FIG. 5. Urinary inorganic sulfate in rats fed the BD or the SAA-free diet. Rats received potassium cyanide for 1, 2, 3, or 4 weeks in drinking water. Urinary inorganic sulfate concentrations were determined as described in Matrials and Methods. Controls were fed rat chow.

232

TOR-AGBIDYE ET AL.

TABLE 3 Blood Cyanide (mmol/L)

FIG. 6. Thiocyanate concentrations were determined as described in the Materials and Methods. Mean thiocyanate concentrations were comparable in both groups at Weeks 1, 2, 3 and 4. Control animals were maintained on rat chow.

exhibited a 48% decrease in median urinary sulfate for each additional week (95% CI is from 38.2 to 61.2% decrease). Urinary Thiocyanate The baseline mean value (n, 8) for urinary thiocyanate was 4.0 6 0.8 nmol/thiocyanate/mL urine. Both groups displayed a marked increase in mean thiocyanate excretion coincident with the introduction of potassium cyanide in drinking water (Fig. 6). Average urinary thiocyanate concentrations, which were approximately 5 times baseline values, were comparable in both groups at Weeks 2, 3, and 4. Urinary thiocyanate concentrations followed a similar temporal pattern in both groups of rats. Analysis showed that urinary thiocyanate did not differ between the two diet groups, and there was no statistical difference between thiocyanate levels and cumulative potassium cyanide intake.

Weeks on diet

Balanced diet

SAA-free diet

2 3 4

6.06 2.0 38.0612.0 60.0621.0

8.06 2.0 a 40.06 5.0 b 81.0611.0 c

Note. Blood was collected intracardially (n, 8) from rats on BD or SAA-free diet (n, 8) at Weeks 1, 2, 3, and 4 for determination of cyanide concentrations. Values for Week 1 are missing because the sample for this period was lost during assay procedures. Blood cyanide concentrations were determined as described in Materials and Methods. Controls were fed rat chow and had blood cyanide levels of 2.8 –3.3 mmol/L (from Tor-Abigdye et al., 1998). Values are means 6 SEM. a p value , 0.038; bp value , 0.001; cp value , 0.001.

2 for the SAA-free group were estimated to be 44% higher compared to the balanced diet group (95% confidence interval for this multiplicative effect was from 3.7 to 10 percent). Overall, median plasma cyanate levels for the BD group increased 1.79 times for each additional week the rats remained on the treatment (95% confidence interval for this multiplicative effect was from 1.48 to 2.16). Similarly, the median plasma cyanate levels for the SAA-free group increased 2.80 times with each passing week (95% confidence interval was 2.31 to 3.39 times). Median cyanate levels for those rats on rat chow (5.0 6 0.8 nmol/L) did not differ significantly from either treatment group (at Week 1), 4.88 6 1.5 and 4.91 6 1.0 nmol/l for BD and SAA-free, respectively. However, both treatment groups had plasma cyanate levels higher than control values (chow diet) at

Blood Cyanide Blood cyanide levels (mmol/L) in both groups showed a definite increase with time (Table 3). Levels of blood cyanide increased an average of 29.4 mmol/L per week (95% CI from 13.6 to 45.2 mmol/L). This rate of increase did not differ for the two diet types (2-sided p value, 0.565) nor did the initial (i.e., Week 2) levels of blood cyanide differ for the 2 diet groups (2-sided p value, 0.506). Plasma Cyanate Plasma cyanate levels exhibited a non-linear trend with respect to time for both treatment groups (Fig. 7). Although cyanate levels in the SAA-deficient rats were significantly higher than those on BD at weeks 3 and 4, there was no difference in median levels at week one [t (28) 5 – 0.324, 2-sided p value, .75]. Median levels of plasma cyanate at Week

FIG. 7. Plasma cyanate in rats fed chow, BD, or SAA-free diet. Median cyanate levels for those rats on chow diet (control) did not differ significantly from either treatment (BD or SAA-free) group at Week one (F 5 0.79 on 31 df with two sided p-values 50.44 and 0.72, respectively). A significant difference between the 2 treatment groups was seen at weeks 2, 3, and 4 (2-sided p value, 0.038, , 0.001, and ,, 0.001, respectively). Plasma cyanate concentrations were determined as described in Materials and Methods. Control rats were fed rat chow.

BIOACTIVATION OF CYANIDE TO CYANATE

FIG. 8. Concentration-response relationships of blood cyanide and plasma cyanate in rats fed the BD or the SAA-free diet. Plasma cyanate and blood cyanide concentrations in rats on BD (n 5 16) and SAA-free diet (n 5 16) were analyzed by linear regression analysis. There is a strong linear relationship between blood cyanide and plasma cyanate concentrations for rats on the SAA-free regimen (r 5 0.927, 2-sided p value , 0.001). A non-significant linear relationship is evident for rats on BD (r 5 0.166, 2-sided p value, 0.61).

Weeks 2, 3, and 4. At Week 2, plasma cyanate concentrations increased in the 2 groups of rats to 16.82 6 5.4 and 19.23 6 2.47 nmol/l in rats on BD and SAA-free diet, respectively. By 3 weeks of treatment, plasma concentrations had risen in both groups of rats, with approximately 2-fold higher concentrations in rats fed the SAA-free diet (2-sided p value , 0.001). At 4 weeks, plasma cyanate concentrations were approximately 4-fold higher in the SAA-free group (2-sided p value , 0.001). Correlation analysis revealed a strong positive linear relationship between blood cyanide and plasma cyanate concentrations (r, 0.927, 2-sided p value , 0.001) for those rats placed on the SAA-free regimen (Fig. 8). For those animals on the BD, there was a weak, insignificant linear relationship between blood cyanide and plasma cyanate (r, 0.166, 2-sided p value, 0.61). DISCUSSION

The early and dramatic rise in urinary thiocyanate, a phenomenon equally apparent in rats fed either the BD or SAAfree diet, demonstrates that enzymatic conversion of cyanide to thiocyanate is a major pathway of cyanide metabolism in the rodent, as in humans (Lundquist, 1992); this conversion is dependent on the availability of sulfur. The marked loss of body weight in cyanide-treated rats fed the SAA-free diet, a process concurrent with the dramatic rise in urinary thiocyanate, suggests the mobilization of endogenous sulfur, principally from the breakdown of body proteins. Rats fed the BD were able to increase their body weight and presumably metabolize cyanide to thiocyanate. The early dramatic rise in their urinary sulfate excretion may in part reflect a higher concen-

233

tration of dietary methionine in the balanced diet relative to that in the standard chow on which animals were acclimated. The marked decline of mean urinary sulfate concentrations in rats fed the SAA-free diet suggests the presence of a sulfur shortage and the possibility that mobilization of endogenous sulfur had begun. Sulfur shortage in rats fed the SAA-free diet was already evident at the end of Week 1. However, even at Week 3, when urinary sulfate approached zero, thiocyanate levels were maintained at 5 times baseline values. Taken together, these observations strongly suggest that CN –/SCN – conversion is a priority metabolic pathway in the rat. This is consistent with observations in protein-deficient rats treated orally with acetonitrile (Sweene et al., 1996). While cyanomethemoglobin is a rapid buffer system in vitro (Lundquist et al., 1985), the present experiments suggest this mechanism comes into play only after the CN –/SCN – conversion pathway has been saturated. The steady increase of total blood cyanide in both groups of rats suggests that the methemoglobin/cyanomethemoglobin conversion is an effective mechanism for the development of cyanide tolerance in rodents. The present study apparently was terminated before the limit of cyanide tolerance was reached in either group of animals. The point at which this occurs would likely correspond to the sudden appearance of free blood cyanide (Lundquist et al., 1985) and consequent effects on the central nervous system. In humans, free cyanide can be converted to cyanate (Lundquist et al., 1993, 1995) which, in the form of sodium or potassium salts, is known to induce neurodegenerative disease in humans, non-human primates, and rodents (Ohnishi et al., 1975; Shaw et al., 1974; Tellez-Nagel et al., 1977). Repeated administration of the anti-sickling drug sodium cyanate to humans, primates, and rodents variably results in damage to basal ganglia, cerebral cortex, spinal cord, and peripheral nerves (Shaw et al., 1974, Tellez, 1979; Tellez-Nagel et al., 1977). Plasma cyanate increased in both groups of rats, but animals fed the SAA-free diet showed markedly higher levels of this neurotoxic moiety. Indeed, plasma cyanate levels appeared to increase exponentially and doubtless would have reached neurotoxic levels (unknown) if the experiment had been prolonged. Cassava leaves and roots harbor the cyanide-releasing glucosides linamarin and lotaustralin. Research conducted in Nigeria examined the fate of pure linamarin (30g/100g body weight) administered in food to 4 groups of female and male Wistar rats: (a) animals deficient in dietary vitamin B 12, (b) rats with sufficient vitamin B 12, (c) malnourished rats, and (d) rats fed a nourishing diet (Umoh et al., 1986). Estimates were made of cyanide and thiocyanate in urine and feces at 0, 24, 48, and 72 h periods after linamarin administration. After 24 and 48 h, the malnourished rats excreted higher concentrations of linamarin metabolites in urine relative to those from wellnourished rats. There were no detectable levels of linamarin in fecal or blood samples drawn at 72 h. In agreement with the present findings, the Nigerian study shows that protein-mal-

234

TOR-AGBIDYE ET AL.

nourished rats retain their ability to convert linamarin-generated cyanide to thiocyanate. Moreover, high levels of urinary thiocyanate were found in both malnourished and well-nourished rats, as in the present study. These findings support the view that thiocyanate excretion is a reasonable quantitative measure of cyanide exposure (Lundquist et al., 1995). The present study shows that once a steady cyanide intake is established, this relationship holds true for at least several weeks of oral exposure to potassium cyanide. Whereas our data demonstrate temporarily increasing blood cyanide concentrations in both animal groups given drinking water containing potassium cyanide, the Nigerian study was unable to detect cyanide levels in blood obtained 72 h after the administration of a single dose of linamarin. This is consistent with the view that cyanide is rapidly converted in both well-nourished and malnourished rats to thiocyanate and not in the first instance to cyanomethemoglobin. The culpable agent responsible for neuronal and/or axonal degeneration in cassava-dependent populations is not established. Two patterns of neurological disease are recognized: an acute onset spastic paraparesis (konzo) and a slowly evolving ataxic myeloneuropathy. The possible etiological agents include cyanide, ATC, thiocyanate, and cyanate.

ated neuronal activity (i.e., glutamate excitotoxicity) associated with the markedly increased levels of thiocyanate selectively found in this type of cassava-related disease (see Tylleska¨r, 1994). These considerations demand a reassessment of the neurotoxic potential of cyanide and its major metabolite, thiocyanate (Spencer, 1999). In summary, we have demonstrated a significant increase in the plasma cyanate concentrations of SAA-deficient rats treated orally with potassium cyanide for 4 weeks. This finding supports the hypothesis that, in SAA-deficient states (as in protein-calorie malnutrition), dietary cyanide exposure results in higher cyanate concentrations, a likely cause of a certain type of neurodegenerative disease in poorly nourished, cassava-dependent African populations. The potential role of thiocyanate in konzo, by analogy with the molecular pathogenesis of lathyrism, merits investigation. ACKNOWLEDGMENTS The technical expertise and assistance of Juan Mun˜iz with GC-MS analysis is very much appreciated. This work was supported by NIH grant NS 19611 and, in part, by the Dupont Foundation.

REFERENCES ●

Cyanide is unlikely to be responsible since the outcome of acute sub-lethal cyanide intoxication is parkinsonism, with changes in basal ganglia, cerebellum, and cerebral cortex (Carella et al., 1988; Rosenberg et al., 1989). ● ATC, a minor neurotoxic component of cyanide metabolism, has not been investigated for systemic toxicity; intracerebroventricular injection in rats induces seizures and hippocampal CA1 damage, neither of which are known to occur in cassava-related neurological disease. ● Cyanate is a plausible etiological candidate because rodents, primates and humans develop motor-system disorders that variably involve both the central and peripheral nervous systems (Ohnishi et al., 1975; Shaw et al., 1974; Tellez 1979; Tellez-Nagel et al., 1977). These neurological conditions appear to be more closely related to cassava-associated ataxic myeloneuropathy than to konzo; cyanate may therefore be a significant etiological factor for the ataxic myeloneuropathy. ● Thiocyanate is generally considered a major, innocuous detoxication product of cyanide. However, recent experimental evidence shows that thiocyanate increases glutamate binding to the a -amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) receptor and potentiates AMPA-mediated responses; there is no effect on N-methyl-D-aspartate receptor-mediated responses (Cha et al., 1992; Hall et al., 1993; Shahi and Baudry, 1992). AMPA receptors are also selectively activated by b-Noxalylamino-L-alanine, the culpable agent in lathyrism which, like konzo, is featured by the abrupt onset of central motor system disease in the form of spastic paraparesis (Abegaz et al., 1994; Spencer et al., 1986). Conceivably, therefore, upper motor neuron disease in konzo may result from enhancement of AMPA-medi-

Abegaz, B. M., Haimanot, R. T., Palmer, V. S., and Spencer, P. S. (1994). Nutrition, Neurotoxins, and Lathyrism: The ODAP Challenge. Third World Medical Research Foundation, New York. Bitner, R. S., Kanthasamy, A., Isom, G. E., and Yim, G. K. (1995). Seizures and selective CA-1 hippocampal lesions induced by an excitotoxic cyanide metabolite, 2-iminothiazolidine-4-carboxylic acid. Neurotoxicology 16, 115–122. Bitner, R. S., Yim, K., and Isom G. E. (1997). 2-Iminothiazolidine-4-carboxylic acid produces hippocampal CA1 lesions independent of seizure excitation and glutamate receptor activation. Neurotoxicology 18, 191–200. Cha, J. H., Makowiec, R. L., Penney, J. B., and Young, A. B. (1992). Multiple states of rat brain (RS)-alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors as revealed by quantitative autoradiography. Mol. Pharmacol. 41, 832– 838 Carella F., Grassi, M. P., Savoiardo, M., Contri, P., Rapuzzi, B., and Mangoni, A. (1988). Dystonic-parkinsonian syndrome after cyanide poisoning: Clinical and MRI findings. J. Neurol. Neurosurg. Psychiatry 51,1345– 8 Cliff, J., Lundquist, P., Martenson, J., Rosling, H., and So¨rbo B. (1985). Association of high cyanide and low sulfur intake in cassava-induced spastic paraparesis. Lancet 2, 1211–1213. Hall, R. A., Massicotte, G., Kessler, M., Baudry, M., and Lynch, G. (1993). Thiocyanate equally increases affinity for two DL-alpha-amino-3-hydroxy5-methylisoxazolepropionic acid (AMPA) receptor states. Mol Pharmacol. 43, 459 – 464. Howlett, W., Brubaker, G., Mlingi, N., and Rosling, H. (1990). Konzo an epidemic upper motor neuron disease studied in Tanzania. Brain 113, 223–235. Lundquist P. (1992). Determination of cyanide and thiocyanate in humans: Ph.D. thesis. Faculty of Medicine. Linko¨ping: Linko¨ping University. Lundquist, P., Backman-Gullers, B., and Kagedal, B. (1993). Fluorometric determination of cyanate in plasma by conversion to 2,4(1H,3H)-quinazolinedione and separation by high-performance liquid chromatography. Anal.. Biochem. 211, 23–27.

BIOACTIVATION OF CYANIDE TO CYANATE Lundquist, P., Kagedal, B., and Nilsson, L. (1995a). Analysis of the cyanide metabolite 2-aminothiazoline-4-carboxylic acid in urine by high-performance liquid chromatography. Anal.. Biochem. 228, 27–34. Lundquist P., Kagedal, B., and Nilsson, L. (1995b). An improved method for determination of thiocyanate in plasma and urine. Eur. J. Clin. Chem. Clin. Biochem. 33, 343–349. Lundquist, P., Mårtensson, J., and So¨rbo, B. (1980). Turbidimetry of inorganic sulfate, ester sulfate, and total sulfur in urine. Clin. Chem. 26, 1178 –1181. Lundquist, P., Rosling, H., and So¨rbo, B. (1985). Determination of cyanide in whole blood, erythrocytes, and plasma. Clin. Chem. 31, 591–595. Ohnishi, A., Peterson, C. M., and Dyck, P. J. (1975). Axonal degeneration in sodium cyanate-induced neuropathy. Arch. Neurol. 32, 530 –534. Osuntokun, B. O. (1968). An ataxic neuropathy in Nigeria: a clinical, biochemical and electrophysiological study. Brain 91, 215–248. Osuntokun, B. O. (1973). Neurological disorders in Nigeria. In Tropical Neurology (J.D. Spillane, Ed.), p. 161. Oxford University Press, London. Osuntokun, B. O. (1981). Cassava diet, chronic cyanide intoxication and neuropathy in Nigerian Africans. World Rev. Nutr. Diet 36, 141–173. Roma´n, G. C., Spencer, P. S., and Schoenberg, B. S. (1985). Tropical myelopathies: The hidden endemias. Neurology 35, 1158 –1170. Rosenberg, N. L., Myers, J. A., and Martin, W. R. (1989). Cyanide-induced parkinsonism: clinical, MRI, and 6-fluorodopa PET studies. Neurology 39, 142–144. Rosling, H. (1996). Molecular anthropology of cassava cyanogenesis. In The Impact of Plant Molecular Genetics (B.W.S. Soral, Ed.), pp. 315. Birhauser, Boston. Rosling, H., and Tylleska¨r, T. (1999). Cassava. In Experimental and Clinical Neurotoxicology (P. S. Spencer and H.H. Schaumburg, Eds.), pp. 338 –343, Oxford University Press, New York. Schultz, F. (1949). Cyanate. Experientia 5, 133–171. Schultz, V. (1984). Clinical pharmacokinetics of nitroprusside, cyanide, thiosulfate and thiocyanate. Clin. Pharmacokin. 9, 239 –251. Shahi, K., and Baudry, M. (1992) Increasing binding affinity of agonists to

235

glutamate receptors increases synaptic responses at glutamatergic synapses. Proc. Natl. Acad. Sci. U S A 89, 6881– 6885. Shaw, C. M., Papayannopolou, T., and Stamatoyannopoulos, G. (1974). Neuropathology of cyanate toxicity in Rhesus monkeys. Pharmacology 12, 166 –176. Spencer P. S. (1999). Food toxins, AMPA receptors and motor neuron diseases. Drug Metab. Rev. 31, 561–587. Spencer, P. S., Roy, D. N, Ludolph, A. C., Hugon, J., Divivedi, M. P., and Schaumburg, H. H. (1986). Lathyrism: Evidence for the role of the neuroexcitatory amino acid BOAA. Lancet 2, 1066 –1067. Swenne, I., Eriksson, U. J., Christoffersson, R., Kagedal, B., Lundquist, P., Nilsson, L., Tylleska¨r, T., and Rosling, H. (1996). Cyanide detoxification in rats exposed to acetonitrile and fed a low protein diet. Fund. Appl. Toxicol. 31, 66 –71. Tellez, I., Johnson, D., Nagel, R. L., and Cerami, A. (1979). Neurotoxicology of sodium cyanate: New pathological and ultrastructural observations in Macaca nemestrina. Acta Neuropathol. 47, 75–79. Tellez-Nagel, I., Korthals, J. K., Vlassara, H. V., and Cerami, A. (1977). An ultrastructural study of chronic sodium cyanate-induced neuropathy. J. Neuropathol. Exp. Neurol. 36, 351–363. Tor-Agbidye, J., Palmer, V. S., Sabri, M. I., Craig, A. M., Blythe, L. L., and Spencer, P. S. (1998). Dietary deficiency of cystine and methionine in rats alters thiol homeostasis required for cyanide detoxification. J. Toxicol. Environ. Health 55, 583–95. Tylleska¨r, T. (1994) The association between cassava and the paralytic disease Konzo. Acta Horticulturae 375, 331–339. Tylleska¨r, T., Banea, M., Bikangi, N., Cooke, R., Poulter, N., and Rosling, H. (1992). Cassava cyanogens and Konzo, an upper motor neuron disease found in Africa. Lancet. 339, 208 –211. Tylleska¨r, T., Franc¸ois, D., Legue, R. M., Petersson, S. Kpizingui, E., and Stecker, P. (1994). Konzo in the Central African Republic. Neurology 44, 959 –961. Umoh, I. B., Maduagwu, E. M., and Amole, A. A. (1986). Fate of ingested linamarin in malnourished rats. Food Chem. 20, 1–9.