TOXICOLOGY AND APPLIED PHARMACOLOGY 83, 10 1- 107 ( 1986)

Encapsulation of Thiosulfate:Cyanide Sulfurtransferase by Mouse Erythrocytes

PETER LEUNG, LEE E. RAY, CHRISTIAN SANDER, JON LEONG WAY, DIANE M. SYLVESTER, AND JAMES L. WAY

Department of Medical Pharmacology and Toxicology, Texas A & M University, College of Medicine. College Station, Texas 77843-1114

Received September 7, I985; accepted October 15, 1985

Encapsulation of Thiosulfate:Cyanide Sulfurtransferase by Mouse Erythrocytes. LEUNG, P.,

RAY, L. E., SANDER, C., LEONG WAY, J., SYLVESTER, D. M., AND WAY, J. L. (1986). Toxicol. Appl. Pharmacol. 83, 101-107. Murine carrier erythrocytes, prepared by hypotonic dialysis, were employed in the encapsulation of several compounds including [‘4C]sucrose, [3H]inulin, and bovine thiosulfate:cyanide sulfurtransferase (rhodanese), a mitochondrial enzyme which converts cyanide to thiocyanate. Approximately 30% of the added [r4C]sucrose, [‘H]inulin, and rhodanese was encapsulated by predialyzed erythrocytes, and a decrease in the mean corpuscular volume and mean corpuscular hemoglobin was observed. In the encapsulation of rhodanese a recovery of 95% of the erythrocytes was achieved and an 85% equilibrium was established. The addition of potassium cyanide (50 rn@ to intact, rhodanese-loaded erythrocytes containing sodium thio- sulfate resulted in its metabolism to thiocyanate. These results establish the potential use of eryth- rocytes as biodegradable drug carrier in drug antagonism. 0 1986 Academic &ss. I~C.

The toxic effects of cyanide have been attrib- uted to the inhibition of cytochrome oxidase, the terminal oxidase of the mitochondria re- spiratory pathway (Keilin, 1929; Warburg, 193 I), resulting in the production of a histo- toxic anoxia. Sodium thiosulfate, one of the antidotes employed in treating cyanide tox- icity, is a sulfur donor for thiosulfate:cyanide sulfurtransferase (EC 2.8.1.1), which is also known by its trivial name of rhodanese, a mi- tochondrial enzyme responsible for biotrans- forming cyanide to thiocyanate (Lang, 1933; Himwich and Saunders, 1948). This enzyme has a turnover number of 20,000 min-’ (Sorbo, 1953; Westley, 198 1); therefore, under ideal in vitro conditions, 1 mol of rhodanese can detoxify 20,000 mol of cyanide per min- ute. Since the intracellular distribution of rho- danese is mitochondrial, the efficacy of en- dogenous rhodanese to detoxify cyanide is limited even though it is present in large

amounts, since sodium thiosulfate does not readily penetrate cell membranes (Way, 1984). Therefore, sodium thiosulfate would neither distribute to the sites of cyanide localization nor to sites of rhodanese distribution. This would partially explain why sodium thiosulfate does not attain its theoretical potential against the lethal effects of cyanide (Way, 1984).

These studies were conducted in an attempt to circumvent some of these adverse disposi- tion factors by placing the enzyme and sulfur donor in close proximity to each other in a protective environment which is still readily accessible by the toxicant, hydrogen cyanide, for detoxification. The approach, as depicted in Fig. 1, is to encapsulate rhodanese with so- dium thiosulfate in an erythrocyte, which would function as a biodegradable drug car- rier. The carrier erythrocyte is highly perme- able to cyanide under physiological conditions because the pK, of cyanide is 9.2, and also it

101 0041-008X/86 $3.00 Copyright 0 1986 by Academic Press, Inc. All rights of repmduction in any form reserved.

102 LEUNG ET AL.

FIG. I. Erythrocyte encapsulation of rhodanese (Rh) and sodium thiosulfate.

would provide a protective environment for the detoxifying enzyme and its substrate. Al- though encapsulation of exogenous enzymes or drugs in intact erythrocytes has been re- ported to serve as a potential vehicle for the delivery of therapeutic agents (Ihler et al., 1973; Deloach et al., 1980b; Fiddler et al.,

1980; Pitt et al., 1983) these studies represent the first attempt to use carrier erythrocytes en- capsulated with enzymes and substrates in an attempt to antagonize an exogenous chemical toxicant.

METHODS

Animals. Balb/c male mice (Charles River Breeding Laboratories, Inc., Wilmington, Mass.) weighing between 18 and 30 g, were housed in temperature (22-24”C)- and light-controlled rooms and were maintained on Rat Chow 5012 (Ralston Purina Co., St. Louis, MO.) and water ad libitum.

Chemicals and isotopes. Carbon IClabeled sucrose (673 mCi/mmol) was purchased from New England Nuclear Corporation (Boston, Mass.). Tritiated inulin (1.84 Ci/ mmol) was obtained from Amersham Corporation (Ar- lington Heights, Ill.). Cibacron blue agarose gel is a product of BioRad (Richmond, Calif.). All other chemicals em- ployed were of the highest purity commercially available.

Isolation andpurification of rhodanese. All isolation and purification procedures were conducted at 2-5°C as de- scribed by Sorb0 (1953) and Westley (1981) with modi- fications for preparative procedures. Fresh bovine livers (approximately 14 kg) were sliced I in. thick, frozen, and thawed several times. After refreezing, the liver was divided into three ahquots, homogenized with I.5 volume of cold

10 mM sodium thiosulfate, and centrifuged (27,578g for 60 min) to remove cellular debris. Supematant fluids were combined and subjected to successive ammonium sulfate (0.4 M), pH 3.8, and ammonium sulfate (1.4 M) fraction- ation procedures, and the precipitate was removed by cen- trifugation (27,578g for 60 min). Subsequently, ammo- nium sulfate was added to the supematant fraction to bring the final concentration to 2.0 M and the preparation was allowed to stand overnight. The enzyme was collected by centrifugation (25,578g for 30 min), dialyzed, and sub- jected to column chromatography with Cibacron blue agarose gel (10 cm X 5 cm’). The sample was applied and washed with 10 mM sodium phosphate buffer (pH, 7.2) eluted with 10 mM sodium phosphate buffer containing 75 mM sodium thiosulfate (pH, 7.2), and concentrated in an Amicon stirred cell with a Diaflo PM10 filter. The concentrated sample was then subjected to a series of pH- ammonium sulfate fractionation and precipitated over- night (Westley, 198 I). White crystalline enzyme was ob- tained by recrystallization in 1.8 M ammonium sulfate. The purification steps resulted in an approximate 600- fold increase in purity with about 9-10s recovery. It is this highly purified enzyme which was employed in our erythrocyte encapsulation studies. One unit of rhodanese is defined as that amount of enzyme which catalyzed the production of 1 rmol of thiocyanate per min (Westley, 1981).

Erythrocyte encapsulation. Whole blood was obtained from adult male Balb/c mice under anesthesia (diethyl ether) by cardiac puncture with heparinized syringes. After collection, plasma and white cells were removed by cen- trifugation and the pelleted red blood cells were resus- pended and washed three times with an isotonic phosphate- buffered solution ( 144 mM NaCl, 2 mM MgCl2 * 6HzO, 5 mM dextrose, pH, 7.4, 304 mmol/kg). The ratio between buffer and red blood cells was maintained lo:1 (v:v). Hy- potonic dialysis of the prewashed erythrocytes then was conducted by the method of Deloach et al. (I 980b) with minor modifications. Dialysis of red blood cells (RBCs) was conducted for 30 min at 2°C in hypotonic phosphate- buffered saline (4 mM MgC&. 6H@, 5 mM dextrose, pH, 7.3,45 mmol/kg). The predialyzed erythrocytes were then mixed for 30 min at room temperature (25°C) with either [?]sucrose, [‘H]inulin, or rhodanese (30-40 units, 65 U/ mg). Subsequently, the erythrocytes were resealed and an- nealed by incubating for 30 min at 37°C in a hypertonic phosphate buffered saline (450 mM NaCI, 10 mM Na2 HP04/NaH2P04, pH, 7.3, 852 mmol/kg). The cells were then washed three times with an isotonic phosphate-buff- ered solution. Hematologic determinations including red blood cell counts, mean corpuscular volume (MCV) in femtoliters (fl = 1 X lo-‘* ml), and hematocrit (Hct) were performed with a Coulter counter in conjunction with a MCV/Hct/RBC computer (Curtis Matheson Scientific, Inc., Houston, Tex.). Osmolality (mmol/kg) was aster- tained with a vapor pressure osmometer (Wescor, Inc., Logan, Utah). The mean corpuscular hemoglobin content

ENCAPSULATION OF RHODANESE BY RBC 103

(MCHB) in picograms was calculated by dividing the weight of hemoglobin by the number ofthe red blood cells in a unit volume of blood. Radioactivity in blood samples was measured by counting aliquots in Bray’s scintillation fluid in a LS 7500 liquid-scintillation counter (Beckman Instruments, Irvine, Calif.). The percentage of rhodanese or [‘%]sucrose encapsulated was calculated from the amount encapsulated divided by the amount added times 100. Percentage of equilibrium was obtained from the concentration of [‘4C]sucrose or rhodanese inside the erythrocytes divided by the concentration of [?Z]sucrose or rhodanese outside the erythrocytes during the mixing phase. The percentage of cells recovered was calculated from actual cell counts performed with a Coulter counter.

Rhodanese activity in loaded red blood cells was deter- mined by suspending an aliquot ( 10 ~1) of loaded red blood cells in an isotonic-buffered solution containing 50 mM

KCN and 50 mM Na2S203 * 5H20 and the formation of thiocyanate at 37°C was measured spectrophotometrically (Westley. 1981) at periodic time intervals. All experiments were performed at least two to three independent times and the results were expressed as mean f 1 SE.

4OOr a

10 20 30 40

HYPOTONIC DIALYSIS (minutes)

g

5 -I 3

RESULTS

To ascertain that the encapsulation tech- nique was consistent with other laboratories, [‘4C]sucrose was employed as a controlled marker in the Balb/c mouse erythrocytes. In an attempt to establish optimal conditions in preparing murine carrier erythrocytes, washed erythrocytes were dialyzed against a hypotonic saline solution, and its osmolality (Fig. 2a) and percent encapsulation of [‘4C]sucrose (Fig. 2b) were determined. A reduction in the osmo- lality of the erythrocytes occurred quite rapidly (Fig. 2a) and 30 min of dialysis was suffi- cient to attain maximum encapsulation of [‘4C]sucrose (Fig. 2b). Since dialysis in excess of 30 min resulted in a decrease in the per- centage of encapsulation, a 30-min dialysis time interval was employed for all subsequent experiments. Approximately 30% of the [14C]sucrose and [3H]inulin can be encapsu- lated. Radioactive [3H]inulin was employed as a larger molecular weight marker, since rhodanese has a molecular weight of approx- imately 33,000.

HYPOTONIC DIALYSIS (minutes)

FIG. 2. (a) Osmolality of the erythrocytes during hy- potonic dialysis. Erythrocytes were dialyzed against a hy- potonic saline solution. (b) Relationship of duration of dialysis and percentage of encapsulation. Aliquots of erythrocytes were removed during hypotonic dialysis and equilibrated with [‘4C]sucrose.

rocytes with control-treated erythrocytes (Ta- ble 1). After encapsulation, the MCV of the sucrose-loaded erythrocytes was reduced by 14% (Table 1) when compared to control- treated erythrocytes (40 fl). Similarly, the rho- danese-loaded erythrocytes had a smaller MCV (Table 2) when compared to control- treated erythrocytes. With a reduction in the MCV, there was an associated decrease in he- moglobin. Approximately 55% of the hemo- globin was retained after encapsulation (Table 1). The cell recovery and equilibrium in the encapsulation procedure were 89 and 87% re- spectively. In the controlled studies where the

During the preparation of carrier erythro- cytes, several hematological indices were monitored to compare sucrose-loaded eryth-

104 LEUNG ET AL.

TABLE I

ENCAFWLATION OF [ “C]S~~~~~~ IN MOUSE ERYTHROCYTES

Erythrocytes

Dialyzed Control

Encapsulation (%) 28.25 f 3.88 0.69 +- 0.06

Equilibration (%) 86.69 iz 5.76 4.93 k 0.38 Cell (%) recovery 89.00 _+ 2.08 96.03 t_ 0.79

MCV (fl) 34.33 2 0.08 40.13 f 0.33 MCHB (pg) 13.99 +- 1.20 25.50 + 0.24

Nole. Percentage of equilibrium was obtained from the concentration of [‘*C]sucrose inside the erythrocytes di- vided by the concentration of [‘4C]sucrose outside the erythrocytes during the mixing phase,

erythrocytes were dialyzed against isotonic sa- line solution, less than 1% of the [ “C]sucrose was encapsulated but the cell recovery was al- most complete (Table 1). The low percentage of encapsulation suggests that most of the [‘4C]sucrose was still extracellular.

Other studies employing rhodanese dem- onstrate that approximately 30% of the rho- danese was encapsulated and changes in the MCV, MCHB, and cell recovery after encap- sulation were similar to those obtained with

“C-sucrose. It should be noted that in control studies, 4.6% of the rhodanese was “encap- sulated” by erythrocytes which were predi- alyzed against isotonic saline. This amount was slightly higher than expected. Again, there was no difference in the percentage of cells recovered in either the dialyzed or control cells.

To evaluate the potential ability of the en- capsulated enzyme to metabolize cyanide, in vitro studies employing erythrocytes loaded with rhodanese and thiosulfate were con- ducted. Subsequent to the addition of potas- sium cyanide to the above loaded erythrocytes, the metabolism of cyanide was quite evident as demonstrated by the formation of thiocy- anate (Fig. 3).

DISCUSSION

These studies have conceptual significance to toxicology since they probably represent the first attempt to employ erythrocyte as a drug carrier mechanism in the antagonism of ex- ogenous chemical toxicants. The application of erythrocytes as a carrier is relatively recent: it was only reported in 1979 by Ihler that with hypotonic dialysis, the erythrocytes become swollen with the development of numerous

TABLE 2

ENCAPSULATION OF THIOSULFATE:~YANIDE SULFXJR- TRANSFERASE IN MOUSE ERYTHROCYTES

2 20

Erythrocytes

Dialyzed Control t

6 Encapsulation (W) 29.63 f 0.12 4.65 + 0.26 G Equilibrium (%) 84.90 k 3.95 4.71 H k 0.05

Cell (W) recovery 94.47 f 2.30 99.03 + 1.61 5 MCV (fl) 35.52 f 0.06 40.13 f 0.33 k MCHB W 11.86 Zk 0.15 18.85 k 0.13 :

15- I /’ ,.

,

lo F4.‘.

,/ -*

5- ! , , I

I I I 0

I 5 10 15 20

Note. Percentage of equilibrium was obtained from the concentration of thiosulfate:cyanide sulfurtransferase in- TME (MINUTES)

side the erythrocytes divided by the concentration of thio- FIG. 3. Thiocyanate formation in erythrocytes contain- sulfate:cyanide sulfurtransferase outside the erythrocytes ing rhodanese (8.25 f 0.23 U/ml erythrocyte) and sodium during the mixing phase. thiosulfate following the addition of potassium cyanide.

ENCAPSULATION OF RHODANESE BY RBC 105

transient openings (Seeman, 1967; Baker, 1967) of sufficient size to permit macromol- ecules as large as hemoglobin to escape (Baker, 1967). Since hemoglobin is much larger than either sucrose, inulin, or rhodanese, the dif- ference in their molecular weights (within limits) should not significantly alter their abil- ity to diffuse into the cell through the transient openings which formed during hypotonic di- alysis. In this regard, their ability to achieve equilibrium would not be affected, and this is reflected in a similar extent of encapsulation for all these three molecules. This then pro- vides the basis to incorporate various sub- stances into red blood cells, since equilibrium can be readily established, resulting in the dif- fusion of intracellular contents out of the cell and large extracellular molecules into the cell. In addition, the lack of any basis for demon- strable proteolytic activity on encapsulated proteins within the red blood cells (Ihler, 1983) provides a basis for using erythrocytes as a carrier for enzyme therapy. Furthermore, erythrocytes employed as drug carriers are stable and remain in circulation from 10 to 36 days depending on the animal species (De- loach, 1983; Deloach et al., 1980a, 1981; Hubbard ef al., 1980). Also, leakage constitutes a very small percentage of the injected dose of erythrocyte-encapsulated drug (Deloach and Barton, 1982).

Thiosulfate, as a sulfur donor, should be a more effective antagonist against the lethal ef- fect of cyanide because of the high content and turnover number of rhodanese. The in- ability of sodium thiosulfate to achieve its maximal therapeutic potential is attributed to its limited penetration of cell membranes; therefore, it does not distribute to sites of rho- danese localization (Way, 1984). In contrast, cyanide exists predominately in the diffusible, unionized form of hydrogen cyanide since the pK, is 9.2. Therefore it can readily penetrate cell membranes and freely distribute through- out the various biological compartments. By the use of erythrocytes as a drug carrier, it is possible to incorporate rhodanese and sodium

thiosulfate into a single compartment where cyanide could readily penetrate the cell mem- brane in the form of hydrogen cyanide. More- over, consideration of the vascular compart- ment as the site of cyanide detoxication would be consistent with the pharmacokinetic studies previously described where detoxification oc- curs predominately in the central compart- ment, which has a volume of distribution equal to that of blood (Sylvester et al., 1983). The potential use of this drug carrier mecha- nism to detoxify cyanide has important im- plications because there is a high content of bovine rhodanese, its turnover number is 20,000, and this reaction is essentially irre- versible. Therefore, the encapsulation of the cyanide antagonist provides a unique oppor- tunity to investigate drug antagonism under optimal disposition conditions. If one is to ex- trapolate the potential of these studies to pri- mates, it would provide a mechanism for a cell to circulate for a few months, and its ac- tivity to detoxify cyanide, in this case, is latent unless cyanide is present. Under the presence of cyanide, it now becomes a very potent mechanism to rapidly convert cyanide to the much less toxic metabolite, thiocyanate.

The initial studies were conducted in mice since the protective effect against the lethality of cyanide was being assessed. From a general concensus, [‘4C]sucrose was employed as a standard marker for comparing preparations of carrier erythrocytes from different labora- tories. In this regard, our encapsulation tech- niques apparently have been established be- cause the various encapsulations and hema- tological parameters are consistent with reports by other laboratories. Also, tritiated inulin, a larger molecular weight standard, was employed in our studies, since rhodanese and low molecular weight substrate, sodium thio- sulfate, were encapsulated. The only discrep- ancy between the [‘4C]sucrose and the rho- danese incorporation lies in the control studies conducted with isotonic saline. Under these circumstances, the rhodanese was membrane bound in higher amounts to the cell than su-

106 LEUNG ET AL.

crose, as was evident in controlled isotonic predialyzed studies. Rhodanese was bound about seven times higher to red blood cell membranes than sucrose. This binding of rhodanese to red blood cells probably does not represent encapsulation, but nonspecific binding, as occurs with many proteins, to the red blood cell membrane. With regard to du- ration of dialysis, control studies indicate that dialysis time should not exceed 30 min in the mouse since the amount of encapsulation rapidly decreases thereafter. These results are consistent with those reported by DeLoach and co-workers ( 1980a) and have been attrib- uted to irreversible lyses or rupturing of the erythrocytes after prolonged dialysis. The he- matological studies indicate that the MCV and MCHB decrease after red blood cell encap- sulation. The mechanism underlying this de- crease in MCV is not clear. However, it would seem reasonable that the loss of cellular con- tent during hypotonic dialysis may account for the reduction of the mean corpuscular vol- ume, as well as hemoglobin concentration.

Once the enzymes and substrates have been encapsulated into red blood cells, it is impor- tant to assess the functional capacity of these encapsulated erythrocytes in performing nor- mal metabolic function as well as specialized toxicologic function. These encapsulated cells appear to retain the anticipated capability of catalyzing the biotransformation of cyanide. Although the external, membrane-bound en- zyme is active in metabolizing cyanide, it only represents a minor fraction (- 5%) of the rho- danese which is present within the erythro- cytes, and therefore would not contribute sig- nificantly to the metabolism of cyanide par- ticularly since the sulfur donor is localized intracellularly. This then represents a potential vehicle for the delivery of therapeutic agents and would protect them from premature deg- radation, inactivation, or excretion with sub- sequent loss of pharmacologic activity. A sus- tained Ievei of the detoxifying enzyme and substrate is of considerable potential since on a theoretical basis, this unique drug carrier

erythrocytic mechanism provides the basis to antagonize the lethal effects of cyanide in a magnitude that far exceeds any methods pres- ently known.

ACKNOWLEDGMENTS

This study was supported by research funds from Texas A & M University, NIEHS and USAMRDC. The authors are indebted to Dr. John R. Deloach for invaluable dis- cussion and advice and the use of his equipment. We also thank Mr. Clayton J. Meyers and Mrs. Kate Andrews for technical assistance and Mrs. Chris Wilson, Mrs. Frances F. Hunley, and Ms. Lynn Baxter for the preparation of this manuscript.

REFERENCES

BAKER, R. F. (1967). Entry of fenitin into human red cells during hypotonic haemolysis. Nature (London) 215, 424-425.

DELOACH, J. R. (1983). In vlvo survival of “C-sucrose- loaded porcine carrier erythrocytes. Amer. J. Vet. Res. 44, 1159-1161.

DELOACH, J. R., AND BARTON, C. ( 1982). Circulating car- rier erythrocytes: slow-release vehicle for an antileukemic drug, cytosine arabinoside. Amer. J. Vet. Rex 43,22 IO- 2212.

DELOACH, J. R., BARTON, C., AND CULLER. K. ( 198 I ). Preparation of resealed carrier erythroytes and in vivo survival in dogs. Amer. J. Vet. Res. 42, 667-669.

DELOACH, J. R., CULLER. K., AND HARRIS, R. L. (1980a). In vivo survival of sucrose-loaded bovine erythrocytes. J. Appi. Biochem. 2, 177-182.

DELOACH. J. R., HARRIS. R. L.. AND IHLER, G. M. ( I980b). An erythrocyte encapsulation dialyzer used in preparing large quantities of erythrocyte ghosts and en- capsulation of pesticide in erythrocyte ghosts. Anal. Biochem. 102,220-227.

RDDLER, M. B., HUDSON, D. S.. WHITE, J. G., AND DES- NICK, R. J. (1980). Enzyme therapy XIV. Comparison of methods for enzyme entrapment in human eryth- rocytes. J. Lab. Clin. Med. 96, 307-317.

HIMWICH, W. A., AND SAUNDERS, J. P. ( 1948). Enzymatic conversion of cyanide to thiocyanate. Amer. J. Physiol. 153,348-354.

HUBBARD, A. R., SPRANDEL, U.. AND CHALMERS, R. A. ( 1980). Dependence of mouse erythrocyte “ghost” survival in vivo on cellular ATP concentration. Biochem. Sot. Trans. 8, 579.

IHLER, G. M. ( 1979). Potential use of erythrocytes as car-

ENCAPSULATION OF RHODANESE BY RBC 107

riers for enzymes and drugs. In Drug Carriers in Biology (G. Gregoriadis, ed.), pp. 129-I 53. Academic Press, New York.

IHLER. G. M. (1983). Erythrocyte carriers. Pharmacol. Ther. 20, 151-169.

IHLER, G. M., CLEW, R. H., ANDSCHNURE, F. W. (1973). Enzyme loading of erythrocytes. Proc. Natl. Acad. Sci. l’S.4 70, 2663-2666.

aILIN. D. (1929). Cytochrome and respiratory enzymes. Proc. R. Sot. London Ser. B. 104,206-252.

LANG, K. (1933). Die rhodanbildung in tierk&per. Biochem Z. 259, 243-256.

PITT. E., JOHNSON, C. M., AND LEWIS, D. A. (1983). En- capsulation of drugs in intact erythrocytes: An intra- venous delivery system. Biochem. Pharmacol. 32,3359- 3368.

SEEMAN, P. (1967). Transient holes in the erythrocyte membrane during hypotonic hemolysis and stable holes

in the membrane after lysis by saponin and lysolecithin. .I. Cell. Biol. 32, 55-70.

SORBS, B. H. (1953). Crystalline Rhodanese: I. Purification and physiochemical examination. Arfa Chem. Stand. 7, 1129-l 136.

SYLVESTER, D. M.. HAYTON, W. L., MORGAN, R. L., AND WAY, J. L. (1983). Effects ofthiosulfate on cyanide phannacokinetics in dogs. To.xicol .4ppi. Pharmacol. 69,265-27 1.

WARBURG. 0. (1931). Uber nicht-hemmung der zellat- mung durch blautiure. Biochem Z. 231,493-491.

WAY, J. L. (1984). Mechanism of cyanide intoxication and its antagonism. Annu. Rev. Pharmacol. Toxicol. 24, 451-481.

WESTLEY, J. (1981). Thiosulfate: cyanide sulfurtransfemse (Rhodanese). In Methods in Enzymology (S. P. Colowick and N. 0. Nathan. eds.), Vol. 77. pp. 285-291. Aca- demic Press, New York.