CHAPTER 2 0
Sulfhydryl Agents: Arsenicals
R. M. Johnstone
I. Introduction 99 II. Classification of the Arsenicals 100
A. Pentavalent Arsenicals 100 B. Trivalent Arsenicals 100 III. Reactions of Arsenicals with Simple Thiols 101
IV. Enzymes as Thiol Compounds 102 A. Effects of Trivalent Arsenicals on Enzyme Systems 102
B. Effects of Pentavalent Arsenicals on Enzyme Systems 106 C. Comparison of the Reactivities of Different Arsenicals with
Various Enzymes 106 D. Reversal by Monothiols of Enzyme Inhibitions Caused by
Arsenicals 107 E. Reactivation with Dithiols 108
V. Inhibition of Lipoic Acid-Dependent Reactions 110 VI. Factors Involved in the Stability of Thioarsenites and Their A p -
plication to the Inhibition of Enzyme Activities I l l VII. Effects of Arsenicals on Physiological Activity 112 VIII. Some Effects of Arsenicals on Tumors 112
IX. Structure of Arsenicals in Relation to Parasiticidal Activity . . . . 113
References 115
I. INTRODUCTION
In recent years considerable work has been carried out on the active groups of proteins to determine which groups are responsible for enzymic activity (1). One of the first groups recognized, which was shown to be essential for enzyme activity, was the sulfhydryl group. The use of arseni- cals, and particularly the organic arsenicals, has yielded much infor- mation on the degrees of reactivity of the sulfhydryl group in various proteins and to what extent the presence of the sulfhydryl group is essen- tial for enzymic activity. As well as being useful in enzyme research, the
99
100 R. Μ . J O H N S T O N E
organic arsenicals have proved to be valuable tools in chemotherapy. In fact, it was from the use of organic arsenicals in chemotherapy that much of the interest in the mode of action of the arsenicals has been derived.
II. CLASSIFICATION OF THE ARSENICALS
In general, the arsenicals which have been studied may be divided into five types:
(a) Pentavalent organic arsenicals, R — A s 03H2. (b) Trivalent arseno compounds, R — As = As — R.
(c) Trivalent arsenoso compounds, R — AsO.
(d) Inorganic arsenite. Inorganic arsenite has also been extensively used, especially in enzyme studies. However, inorganic arsenite is rela
tively inert compared with the trivalent organic arsenicals.
(e) Inorganic arsenate. Arsenate does not appear to act as a sulfhydryl reagent, but has been widely used in place of inorganic phosphate.
A. Pentavalent Arsenicals
It is now generally accepted that the active form of the organic arseni
cals is the arsenoso derivative. The work of Ehrlich (2, 3) suggested that the pentavalent form of arsenic must be reduced to the trivalent to obtain parasiticidal activity. Hawking et al. (4) later showed that the pentava
lent arsenical, tryparsamide, is reduced in the body to a highly trypano
cidal trivalent arsenical. Although the trivalent arsenoso compounds were at first considered too toxic for use, Tatum and Cooper (5) demonstrated that Mapharsen (2-amino-4-arsenosophenol) can be used successfully in place of the less toxic pentavalent compounds.
B. Trivalent Arsenicals
1. A R S E N O C O M P O U N D S
Considerable confusion existed for some time as to whether the arseno compounds are active as such. Voegtlin and Smith {6) showed that with arseno compounds there is a latent period before trypanosomes disappear from the blood, whereas with the arsenoso compounds there is an im
mediate effect. Other workers claimed that arseno compounds are active without further modification (7, 8). Eagle (9) resolved the discrepancy
20. S U L F H Y D R Y L A G E N T S : A R S E N I C A L S 101 by showing that certain arseno compounds, the arsphenamines, are con- siderably less active against Treponema pallidum if precautions are taken to dissolve and test the arseno derivative under anaerobic conditions.
Under aerobic conditions the arseno compounds are readily oxidized, and their activity is therefore due to the formation of the arsenoso derivatives.
The trivalent arsenoso compounds are therefore the active form of this class of compounds.
2. A R S E N O S O C O M P O U N D S
In the majority of the compounds studied, both in enzyme research and in chemotherapy, the R is benzenoid. Many of the arsenical war gases, however, have aliphatic side chains (10). The aliphatic and polycyclic arsenicals are generally less active than the benzenoid derivatives.
III. REACTIONS OF ARSENICALS WITH SIMPLE THIOLS
The evidence accumulated so far indicates that the arsenicals owe their toxic effects to their ability to combine with thiol groups. Ehrlich (2, 8) originally suggested that arsenicals exert their effects by combining with sulfhydryl groups. Later, Voegtlin and associates (11) postulated that the toxic action of arsenoso compounds toward trypanosomes is due to a combination with thiol groups in the protoplasm (they suggested gluta- thione), leading to an inhibition of respiration. Barber (12) demonstrated that arsenosobenzenes may combine with organic sulfhydryl compounds such as thiolacetic and thioglycolic acid to form a thioarsenite according t o E q . (1).
SR /
C6H6AsO + 2RSH -> C6H5As + H20 (1)
\
SR
It was also shown (12) that thiol compounds in sufficient concentration will reduce a phenylarsonic acid to the arsenosobenzene. Gough and King
(13) concluded from an examination of the properties of the arylthio- arsenites that these compounds are slowly hydrolyzed to yield highly toxic arsenoxides at ordinary temperatures and in neutral solution. The dis- sociation of a thioarsenite may be controlled by pH, a higher pH leading to dissociation of the thioarsenite (14).
It thus appears that the reaction of a trivalent arsenoso compound
1 0 2 R. Μ . J O H N S T O N E
with simple thiol compounds is a reversible one and that, as suggested by- Gordon and Quastel (15), in neutral solution an equilibrium can exist between an arsenoso and a thiol compound, as shown in Eq. ( 2 ) .
SR /
C6H5-AsO + 2RSH ^± C6H5As + H20 (2)
\ SR
IV. ENZYMES AS THIOL COMPOUNDS
The knowledge that trivalent arsenicals can combine with thiol com
pounds did not immediately lead to a realization that the toxic effect of the arsenicals may be due to their ability to combine with thiol groups present in essential enzymes. Prior to 1 9 2 0 , it was known that the nitro- prusside test for sulfhydryl groups is given by a variety of tissues. The work of Hopkins and Dixon (16) drew attention to two types of thiol compounds, "soluble/' like glutathione, and "fixed," meaning associated with cellular protein. It was not until later that thiol groups were shown to be an integral part of certain enzymes and that the enzymic activity often depends on the presence of these groups.
Sumner and Poland (17) in 1 9 3 3 first demonstrated the presence of sulfhydryl groups in crystalline urease. In the next few years, it was shown that several hydrolytic enzymes contain sulfhydryl groups (18-20) and that the enzymic activity depends on these groups. The presence of sulfhydryl groups in oxidizing enzymes was demonstrated in 1 9 3 8 by Hopkins and Morgan (21), by Morgan et al. (22) in succinoxidase, and by Rapkine (23) in phosphoglyceraldehyde dehydrogenase. Thus, it be
came apparent that a number of widely different enzymes contain thiol groups as integral parts of their active centers.
A. Effects of Arsenicals on Enzyme Systems
1. P Y R U V A T E
a. Pyruvate Metabolism. It is now well established that the oxida
tion of pyruvate in many types of cells, animal as well as bacterial, is markedly inhibited by organic trivalent arsenicals and inorganic arsenite.
As early as 1 9 1 1 , it was shown (24) that biological oxidations are in
hibited by arsenite. Subsequent observations confirmed these findings
20. S U L F H Y D R Y L AGENTS'. A R S E N I C A L S 103 (25,26), and it was shown (27) that keto acid oxidations are particularly sensitive to the action of arsenite. Peters and co-workers (28) found that pyruvate oxidation in brain is much more affected by the arsenical war gas, lewisite, than by inorganic arsenite at equivalent concentrations.
The work of Barron and Singer (29) and of Gordon and Quastel (15) showed that several organic arsenicals are powerful inhibitors of pyruvate metabolism. A number of trivalent arsenicals, derivatives of arsenoso- benzene, inhibit the pyruvic oxidase activities in brain, liver, and micro
organisms (15, 29). N o t only is the oxygen uptake with pyruvate as substrate inhibited, but the synthesis of a number of substances, such as acetoacetate, α-ketoglutarate, acetylmethylcarbinol, citrate and glycogen, is inhibited (80). The concentration of arsenical required to obtain complete inhibition of pyruvate metabolism varies from one arsenical to another but is usuallly considerably less than 1 m M (15, 29).
y-(p-Arsenosophenyl)butyrate at a concentration of 5 X 1 0- 6i l f causes complete inhibition of pyruvate utilization and citrate formation in rat heart sarcosomes (81).
The oxidation of α-ketoglutarate as well as other intermediates of the citric acid cycle, such as malate and succinate, is also inhibited by organic arsenicals (15, 29).
b. Pyruvate Decarboxylation and Alcohol Oxidation. An enzyme de
rived from two different sources does not necessarily respond to thiol reagents in the same way. Lutwak-Mann (32) showed that yeast alcohol dehydrogenase, but not liver alcohol dehydrogenase, is inhibited by iodo- acetate. These findings were confirmed by Barron and his colleagues using arsenicals as thiol reagents (29, 33). Stoppani and his colleagues (84-36) showed that yeast pyruvic decarboxylase is sensitive to the action of a number of organic arsenicals.
c. Aldehyde Dehydrogenase. N o t only the oxidation of keto acids, but the oxidation of aldehydes, such as glycoaldehyde, acetaldehyde (86a, b, c), succinic semialdehyde (36d), and xanthine (36e), is inhibited by inorganic arsenite.
2. S U C C I N A T E O X I D A T I O N
The succinoxidase system and succinic dehydrogenase have been ex
tensively studied with respect to the inhibitory activity of arsenicals.
Although Peters et al. (28) reported that lewisite does not affect succinic dehydrogenase activity, later work has shown that this system is appre
ciably inhibited by a variety of organic arsenicals, including lewisite.
Thus, Gordon and Quastel (15) showed that complete inhibition of suc
cinic dehydrogenase activity in rat liver slices is obtained with 1 0 ~3 Μ
1 0 4 R. Μ . J O H N S T O N E
2-amino-4-arsenosophenol. Similar results were obtained by Barron and Singer (29). Slater (37) compared the effects of p-arsenosoaniline on suc
cinic dehydrogenase and on the succinoxidase system and found that the inhibition of succinic dehydrogenase by p-arsenosoaniline can account for the observed inhibition of the succinoxidase system. A highly purified preparation of succinic dehydrogenase is also sensitive to the action of arsenicals (38).
In chick liver homogenates, succinic dehydrogenase may be inhibited by inorganic arsenite (38a). Malic and lactic acid dehydrogenases are also markedly inhibited by arsenite (38b).
3 . G L U C O S E U T I L I Z A T I O N
Van Heyningen (39) has suggested that hexokinase depends on sul
fhydryl groups for metabolic activity, and it was shown (40) that crystal
line yeast hexokinase is completely inhibited by 1 0 ~3 Μ lewisite. Gordon and Quastel (15) observed that glucose oxidation in brain tissue is in
hibited by trivalent organic arsenicals. Marshall (41) and Chen (42) suggested that in certain trypanosomes, hexokinase is inhibited by arsenosobenzene and 2-amino-4-arsenosophenol.
Glock and McLean (48) found that glucose-6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase in rat liver are inhibited by p-chloromercuribenzoate, whereas p-arsenosoaniline at 2 . 5 Χ 1 0 ~4 Μ has no effect. In most cases (29), where the arsenicals have been compared with p-chloromercuribenzoate,' both reagents inhibit metabolic activity, although p-chloromercuribenzoate is effective at lower concentrations.
Respiration and glycolysis are inhibited by arsenite in the Jenson sarcoma (48a, b), the former being considerably more affected than the latter.
4 . E F F E C T S OF A R S E N I C A L S O N A M I N O A C I D M E T A B O L I S M
Singer and Barron (44) in their systematic study of sulfhydryl enzymes in cell metabolism showed that the oxidation of D-alanine in rat kidney tissue is completely inhibited by (p-aminophenyl)dichloroarsine hydro
chloride at 5 X 1 0 ~5 M. Singer (45) showed similar inhibitions of pig- kidney D-amino acid oxidase with p-aminophenylarsine oxide. Using a variety of D-amino acids as substrate, Singer found that the extent of inhibition by the arsenical is independent of the amino acid used as sub
strate. The arsenical, y-(p-arsenosophenyl)butyrate, is also an effective inhibitor of D-alanine oxidation with a purified D-amino acid oxidase (45).
The DPN-linked oxidation of glutamic acid in liver is inhibited by
2 0 . S U L F H Y D R Y L A G E N T S : A R S E N I C A L S 1 0 5
trivalent arsenicals as well as the glutamate-pyruvate and glutamate- oxaloacetate transaminase activities (44)· In many cases where the organic arsenicals are effective inhibitors, inorganic arsenite or iodoacetate has little effect (15, 29, U)-
The enzymes of Clostridium sporogenes, which catalyze the anaerobic oxidation and reduction of amino acids (46), are highly sensitive to the action of organic arsenicals such as arsenosobenzene and 2-amino-4- arsenosophenol. The enzyme system in C. sporogenes activating molecular hydrogen is particularly sensitive to the action of arsenicals, being com
pletely inhibited at a concentration of 4 χ 1 0 ~5 Μ 2-amino-4-arsenoso- phenol. This concentration of 2-amino-4-arsenosophenol inhibits the reduction of amino acids by approximately 5 0 % (47).
The oxidation of monoamines, including adrenaline, is inhibited by arsenicals (44), in agreement with the observation of Friedenwald and Herrmann (48) that tyramine oxidase is a thiol enzyme. Diamine oxidase
(44), however, is not inhibited by arsenicals.
5. F A T T Y A C I D O X I D A T I O N
Many of the enzymes catalyzing the oxidation of fatty acids have been shown to be inhibited by arsenicals. Stearate, oleate, and β-hydroxy- butyrate oxidation in animal tissues, as well as acetate oxidation in yeast and bacteria, are inhibited by arsenicals (44)- β-Hydroxybutyrate oxida
tion is highly sensitive to organic arsenicals, being completely inhibited by 5 X 10~"4 Μ p-arsenosobenzoic acid (44) ·
The breakdown of acetoacetate in guinea pig kidney slices is nearly completely inhibited by inorganic arsenite (48a), and the formation of acetoacetate from fatty acids is largely prevented by atoxyl (48b).
6. H Y D R O L Y T I C E N Z Y M E S
A wide variety of hydrolytic enzymes have been shown to be inhibited by organic arsenicals. Indeed, the hydrolytic enzymes were the first in which the presence of free thiol groups was associated with enzyme activity (17-20). Rona and Gyorgy (49) found that arsenosobenzene, and arsenosomethane cause appreciable inhibition of urease activity, and Rona, Airila, and Lasnitski (50) found that maltase and a-methyl- glucosidase are also inhibited by arsenosomethane. Hellerman (18) had shown that urease is inhibited by thiol reagents, and these findings were confirmed (15) by showing that low concentrations of 2-amino-4- arsenosophenol and other arsenicals are capable of inhibiting urease activity.
106 R. Μ. JOHNSTONE Some of the hydrolytic enzymes are relatively insensitive to arsenicals.
High concentrations of 2-amino-4-arsenosophenol have to be used to in
hibit liver esterase activity {15). Gordon and Quastel (15) could obtain no inhibition of pancreatic lipase with arsenicals, whereas Singer and Barron (44) obtained a 50% inhibition of this enzyme with the same arsenicals.
The discrepancy in these results may perhaps be due to the different incubation periods used by the authors. The former authors (15) used a very short incubation period (15 minutes), whereas the latter (44) used an incubation period of 6 hours. Slater (87) has shown that the inhibition of succinoxidase activity by p-arsenosoaniline takes approximately 15 minutes to become established. Wheat germ lipase (45) is sensitive to low concentrations of p-arsenosoaniline. Singer (45) has further shown that with this enzyme preparation the extent of inhibition by the arsenical depends on the substrate used, oxidation of substrates of higher molecular weight being more inhibited than those of low molecular weight.
Acid phosphatase and liver arginase are inhibited by relatively high concentrations of lewisite (88).
Cholinesterase and choline dehydrogenase have both been shown to be thiol enzymes inhibited by arsenicals (15, 88, 51, 52).
Mounter and Whittaker (51) have compared the sensitivity of cholin
esterase to arsenicals with that of other thiol enzymes and have found it to be relatively insensitive.
B. Effects of Pentavalent Arsenicals on Enzyme Systems
The pentavalent arsenicals, unlike the trivalent ones, have little effect on enzymic activities (15, 47).
C. Comparison of t h e Reactivities of Different Arsenicals with Various Enzymes
The extent of inhibition of a given enzyme often depends on the type of arsenical used. This is clearly shown by the fact that arsenite is a much poorer inhibitor of enzyme activities than are organic trivalent arsenicals (15, 29). However, even among the organic arsenicals, there are many differences in their inhibitory effects on enzyme systems.
Yeast carboxylase, D-amino acid oxidase (kidney), and urease are not inhibited by lewisite (88) although all three enzymes are appreciably inhibited by derivatives of arsenosobenzene (15, 85, 44)- On the other hand, pyruvate oxidation in brain preparations is markedly inhibited by
20. SULFHYDRYL AGENTS: ARSENICALS 107 lewisite {28). Stoppani et al. (36) compared the effects of arsenoso- methane and 2-amino-4-arsenosophenol on purified yeast decarboxylase.
They showed that nearly 7 times as much arsenosomethane as 2-amino-4- arsenosophenol is required to obtain 50% inhibition of decarboxylase activity. The respiration of brain and kidney slices is far more sensitive to lewisite than to p-arsenosobenzoic acid (33), an inhibition of 66% of the respiration being obtained with the former and 10% with the latter at equimolar concentrations.
D. Reversal by Monothiols of Enzyme Inhibitions C a u s e d by Arsenicals Since the thiol group can be destroyed by a number of means, e.g., oxidation, alkylation, mercaptide formation, it is important to demon
strate that the loss of enzymic activity observed in the presence of arsenicals is due to the formation of a thioarsenite complex with enzyme thiol groups. It has been shown (13, 14) that the linkage of an arsenical with a thiol is a dissociable one; therefore, addition of excess thiol should bring about a reversal of the inhibitions and toxic reactions caused by arsenicals.
Voegtlin et al. (11, 53) have shown that an excess of thiol compounds, such as glutathione, may delay or prevent the action of arsenicals on trypanosomes. The inhibitions produced by a number of arsenicals and on a variety of enzymes can in every instance be largely prevented by the presence of excess thiol compound, such as glutathione and cysteine.
In most cases, a large excess (a hundredfold) is required for complete protection (15, 29, 44).
It is obvious, however, that protection against inhibition is not identical with reactivation. The experiments of Barron and Singer (29), Singer and Barron (44) > and Gordon and Quastel (15) showed that reactivations of enzyme inhibition do occur after the inhibition has set in; for example, cholinesterase may be completely reactivated by thiols even after a 2-hour exposure to arsenicals (15). In many cases, however, the extent of reacti
vation decreases after prolonged contact with the arsenical. Thus, Barron et al. (83) demonstrated that glutathione will reverse by 50% the lewisite inhibition of anaerobic glycolysis if added to the system 10 minutes after the addition of the arsenical. If the glutathione is added 30 minutes after the arsenical, only 20% reversal is observed.
The nature of the monothiol used also plays a role in determining the extent of reactivation. The succinoxidase system, inhibited by p-arsenoso- benzoic acid (5 X 10~~5 M), can be completely reactivated by Ι Ο- 3 Μ glutathione but only 25% reactivated with 1 0 ~2 Μ cysteine (29). Simi-
108 R. Μ. JOHNSTONE larly, Gordon and Quastel (15) and Barron and Singer (29) obtained reversals of pyruvate oxidation with glutathione after inhibition with arsenosobenzene derivatives, but Stocken et al. (54, 55) could not obtain a reversal of Mapharsen-inhibited pyruvate oxidation with 2-mercapto- ethanol. It is apparent, therefore, that the thiol used for reactivation greatly influences the extent of reactivation.
E. Reactivation with Dithiols
It is now well established that dithiols are more effective than mono- thiols in reversing the inhibitions caused by arsenicals. Curiously enough, it was the inability to demonstrate reversal with monothiols of inhibitions produced by lewisite (54, 55), which eventually led to the synthesis of British Anti-Lewisite (BAL).
To obtain information on the nature of the arsenical-protein complex, Stocken and Thompson (56) studied the effects of adding lewisite to the protein kerateine, which has a large number of thiol groups. Rosenthal (57) had shown that when trivalent arsenicals are combined with proteins, the arsenicals cannot be separated from the protein by ultrafiltration.
Stocken and Thompson (56) demonstrated that the addition of lewisite to kerateine causes a disappearance of the free thiol groups. Moreover, 75-90% of the arsenic bound to protein is combined in a molar ratio of one arsenic to two thiols. If kerateine is pretreated with mild oxidizing agents which destroy thiol groups, there is no combination with lewisite.
These data provide additional evidence that free thiol groups are required for combination with the arsenicals.
The ratio of arsenical to thiol in the kerateine-lewisite complex sug
gested to Stocken and Thompson (56) that lewisite combines with adjoin
ing thiols in this protein and that simple dithiols might reverse or prevent the toxic actions of the arsenicals more effectively than monothiols. The results of these experiments show that a number of low molecular weight dithiols form relative stable complexes with lewisite (55). Stocken and Thompson (55) measured the dissociation of the lewisite-thiol complex by estimating the time required for the reduction of porphyrindin with various thiol-lewisite complexes. The reaction proceeds according to Eqs.
(3) and (4).
The thioarsenites were used at equimolar concentrations and the porphyrindin at a concentration one-half that of the thioarsenites. Thio
arsenites from monothiols reduce porphyrindin much more quickly than do thioarsenites formed from dithiols (Table I ) .
Moreover, the dithiols, ethane-l,2-dithiol, 2,3-dimercaptopropanol
20. SULFHYDRYL AGENTS: ARSENICALS 109 R—S
\
As—CH=CHC1 + H20 τ± ClCH=CHAsO + 2RSH (3) /
R—S
2RSH + porphyrindin —> R—S—S—R + leueoporphyrindin (4) (BAL), 1,3-dimercaptopropanol, propane-1,3-di thiol, α,β-dimercaptopro-
pionic acid, are capable of reversing the inhibition by lewisite of pyruvate oxidation where simple monothiols are not effective. A number of dithiols such as β,β'-dimercaptodiethyl sulfide and pentane-l,5-dithiol do not
TABLE I
HYDROLYSIS RATES OF THIOARSENITES PREPARED FROM LEWISITE AND THIOLS (55)
Decolorization time
Thiol (minutes)
Cysteine 0.25
Thioacetic acid 0.50
Aminothiophenol 0.83
Glutathione 2.5
2-Mercaptoethanol 3 . 5
2,3-Dimercaptopropanol (BAL) >180 Ethane-l,2-dithiol >180 1,3-Dimercaptopropanol > 1 8 0 Propane-1,3-dithiol >180 0,/3'-Dimercaptodiethyl ether >180
overcome the lewisite-produced inhibition of pyruvate oxidation. These results prompted Stocken and Thompson (55) to conclude that a 5- or 6-membered ring between arsenical and dithiol is the most stable thio
arsenite complex and that this property of the dithiols can be used to advantage to liberate a free thiol from another less stable thioarsenite.
Whittaker (58), however, has shown that a number of dithiols are effec
tive in reversing lewisite inhibitions. The least effective in this group are butanedithiol and pentanedithiol. Whittaker (58) has concluded that these results are consistent with the hypothesis that a cyclic structure is formed between arsenicals and dithiols and that the ring need not be limited to five or six members. Eagle and Doak (59), however, suggest that there is insufficent evidence to support the hypothesis that cyclic thioarsenites are more stable than the straight chain compounds.
Nevertheless, there is no doubt that many dithiols are more effective
110 R. Μ. JOHNSTONE in reversing the inhibitory actions of arsenicals than are monothiols
(53-55). Stocken and Thompson (55) have shown that a concentration of British Anti-Lewisite three times that of arsenosobenzene is capable of reversing the arsenosobenzene inhibition of pyruvate oxidation. Con
siderable excess of monothiol (up to a hundredfold excess) is required to reverse the inhibitions caused by arsenosobenzene derivatives on the pyruvate oxidation system (15, 29).
Of all the dithiols prepared, BAL has proved to be the most effective and easiest to use for many practical reasons, such as solubility, relative lack of toxicity, and nonvolatility. The clinical use of BAL in cases of arsenical poisoning and poisoning by other heavy metals has been very extensive (59-61). Although originally prepared as an antidote to the war gas lewisite, it has found its greatest use in accidental arsenical poisoning.
V. INHIBITION OF LIPOIC ACID-DEPENDENT REACTIONS
Although there is much evidence to suggest that many of the inhibitions caused by the arsenicals are due to combination with the thiol groups of enzymes, recent evidence suggests that a site of action of arsenicals may also be the sulfhydryl groups of lipoic acid. The data already given show that α-keto acid oxidation is particularly sensitive to the action of triva
lent arsenicals. It is now well established that lipoic acid is a cofactor in α-keto acid oxidation (62). Reiss and Hellerman (81) and Reiss (68) have found that arsenicals inhibit all lipoic acid-dependent reactions.
Moreover, Sanadi et al. (64, 65) have shown that the biological oxida
tion of α-ketoglutarate by ferricyanide, which is lipoic acid independent, is insensitive to arsenite. The oxidation of α-ketoglutarate in the absence of an artificial hydrogen acceptor is well known to be arsenite sensitive The inhibition by y-(p-arsenosophenyl)butyrate of pyruvate utilization and citrate formation in rat heart sarcosomes can be reversed more effectively with dihydrolipoic acid than with BAL (31). These results have given support to the belief that lipoic may be a site of action of arsenicals.
It is interesting to note that there is some structural similarity between BAL and lipoic acid.
(29).
SH SH SH SH
CH2—CH—CH2OH CH2—CH2—CH—(CH2)4—COOH Dihydrolipoic acid BAL
20. SULFHYDRYL AGENTS '. ARSENICALS 111 It has also been suggested that the sulfhydryl group of coenzyme A may be a site of action of arsenicals.
VI. FACTORS INVOLVED IN THE STABILITY OF THIOARSENITES A N D THEIR APPLICATION TO THE
INHIBITION OF ENZYME ACTIVITIES
Thioarsenites formed from dithiols are relatively stable complexes and dissociate at far lower rates than do thioarsenites from monothiols {55).
This fact no doubt accounts for the greater ease with which dithiols reverse the inhibitions produced by arsenicals. However, thioarsenites formed from a given dithiol and a variety of arsenicals may have different rates of dissociation; for example, the Mapharsen-BAL complex dis- sociates to a greater extent than does the lewisite-BAL complex {66).
(A positive nitroprusside test is obtained with Mapharsen-BAL but not with lewisite-BAL.) The lewisite-BAL complex has less than one-fifth the toxicity of lewisite itself, whereas Mapharsen-BAL is at least as toxic as the free arsenical (66). However, other authors have reported a dimin- ished toxicity of the Mapharsen-BAL complex (67-69).
With a given arsenical, the dissociation rate of its monothioarsenites is greatly affected by the nature of the monothiol; for example, the rate of dissociation of the lewisite-cysteine complex is 10 times as great as that of the glutathione complex. Such factors no doubt account for the fact that glutathione is better able to reverse the effects of the arsenicals than is cysteine (15, 29). It no doubt accounts also for the observations that thioarsenites, such as arsenosobenzene thioglycolate (15) and other thio- arsenites (70-72) may be nearly as toxic as the parent compound
(see also reference 59).
The data described indicate that reactivation by a given thiol may depend on the dissociation rate of the thioarsenite formed. The reverse is probably also true; the extent of inhibition by a given arsenical may depend on the dissociation rate of the thioarsenite formed with cellular thiol compounds. Thus, appreciable differences exist in the extents of inhibition of various enzymes by a given arsenical (15, 29, 44, 4?)- In the pyruvic oxidase system, the thiol groups affected appear to be those of lipoic acid (31, 63). It is conceivable that the dissociation rate of the thioarsenites formed with lipoic acid may be appreciably lower than those formed from a protein thiol and an arsenical. The formation of cyclic thioarsenites has previously been proposed to explain the increased stability of dithiol arsenicals (54, 56, 58). The possibility of ring forma-
112 R. Μ. JOHNSTONE tion between enzyme thiol groups and arsenicals has not obtained support (87, 59), although ring formation between arsenicals and lipoic acid has obtained support (81, 63).
The nature of the arsenical may also influence the dissociation rate of the thioarsenite formed, be it monothiol or dithiol. Such an interpretation may explain the observations that different arsenicals inhibit the same enzymes to appreciably different extents; for example, urease, which is inhibited by 1.6 Χ 1 0 ~4 Μ Mapharsen (15), is not affected by Ι Ο -3 Μ lewisite (33). Yeast carboxylase is inhibited by (p-aminophenyl) dichloro- arsine hydrochloride (29) and Mapharsen (85) but not by lewisite (88).
Similarly, D-amino acid oxidase is inhibited by (3-amino-4-hydroxy- phenyl)dichloroarsine hydrochloride (44) and by p-arsenosoaniline (45) but not by lewisite (38). In many cases, enzymes inhibited by the arsenosobenzene derivatives are also inhibited by lewisite (15, 29, 38).
Such observations indicate that in studies of the inhibitory effects of organic arsenicals and the reversal of the inhibitions, the nature of the arsenical as well as the thiol compounds may influence the degree of inhibition and the extent of reversal.
VII. EFFECTS OF ARSENICALS O N PHYSIOLOGICAL ACTIVITY
There can be little doubt that the inhibitions of tissue respiration and glycolysis observed in presence of arsenicals are due to an inhibition of specific enzymes involved in these processes (15, 38, 73, 74) Mtiller has shown that mitosis in epithelial cells is disturbed by diphenylarsine (75).
Trivalent, but not pentavalent, arsenicals inhibit the growth of mam
malian cells cultured in vitro (76). Inorganic arsenite will abolish the effect of A C T H on steroid synthesis in adrenals (77). In plants, arsenite prevents the response to auxins (78). Virus reproduction is inhibited at a concentration of arsenite which has no effect on bacterial growth or virus adsorption (79).
These experiments illustrate the wide variety of physiological activities that may be affected by arsenicals in many types of cells.
VIII. SOME EFFECTS OF ARSENICALS O N TUMORS
Lacassagne et al. (80) have reported that the application of organic arsenicals to mouse skin does not initiate tumor growth. Beck (81) showed
20. SULFHYDRYL AGENTS'. ARSENICALS 113 that of 39 pentavalent arsenicals, aliphatic and aromatic, only sodium cacodylate produced gross histological damage to mouse sarcoma S 3 7 at maximum tolerated doses. Twenty-four trivalent arsenicals did produce gross histological damage to mouse tissue. With some arsenicals, e.g.
phenyldichloroarsine, the toxic response to the arsenical is decreased in tumor-bearing animals. With Mapharsen the toxicity of the arsenical is identical in normal and tumor-bearing mice (81). Further studies by Beck and Gillespie (82) on the toxicity of arsenicals on tumor-bearing mice showed that the decreased toxicity occurs only in animals bearing small tumors (82). Cortisone and A C T H protect the animal against sublethal doses of arsenicals (83).
Mapharsen has recently been shown to reduce the mitotic index of 6C3HED lymphosarcoma (84). A single injection kept the mitotic index at a low level over a prolonged period of time. The addition of 6-mercaptopurine appears to offset the effects of Mapharsen (84).
IX. STRUCTURE OF ARSENICALS IN RELATION TO PARASITIC I DAL ACTIVITY
Much of the work on the biological activity of the arsenosobenzenes had been done in relation to parasiticidal activity. Prior to the advent of penicillin, the arsenicals were extensively used in the treatment of trypanosomal infections. A large number of derivatives of arsenoso
benzene were synthesized in an attempt to enhance the parasiticidal activity of the arsenosobenzene without increasing the toxicity to the host.
A review of the structural changes of the arsenosobenzenes in relation to biological activity has appeared (59). An attempt will be made here to summarize how the changes in the structure of the arsenosobenzenes affect the parasiticidal activity.
Unsubstituted arsenosobenzene is highly toxic to every cell type and is too toxic to be used therapeutically. Substitution on the benzene ring by
—CI, — N 02, —OH, — C H3, — F and the position of these substituents do not appreciably alter the activity of arsenosobenzene and have no selec
tive action (59).
Acidic substituents, such as —COOH, — S 03H , and R—COOH, markedly depress both the treponemicidal and trypanocidal activities without decreasing the toxicity to the animal to the same extent (59).
Substitution by fatty acids to give the y - (p-arsenosophenyl) butyric acid, δ- (p-arsenosophenyl) valeric acid, and c-(p-arsenosophenyl)caproic acid enhances the parasiticidal activity relative to the toxicity to the host.
114 R. Μ. JOHNSTONE The first two compounds are actively trypanocidal against Trypanosoma equiperdum both in vivo and in vitro, although they have no sig
nificant effect against Treponema pallidum. Conversely, the caproic acid derivative has considerable treponemicidal activity but is almost inert against T. equiperdum. Evidently, highly specific factors are involved in producing the enhanced parasiticidal activity or diminished parasiticidal activity.
Eagle and Doak (59) have concluded that the nonionized forms of acid-substituted arsenosobenzenes are the active forms. With some of the acid-substituted arsenosobenzenes, a hundredfold difference in parasit
icidal activity is observed in the pH range 8.5-5.5.
Substitution of arsenosobenzenes with acid amide groups results in a remarkably uniform toxicity and parasiticidal activity. Sixteen com
pounds of this general structure are described by Eagle and Doak (59).
The activity of the acid amide-substituted arsenosobenzene lies midway between the free arsenosobenzene and the acid-substituted substance; for example, the in vitro parasiticidal activity of p-arsenosobenzamide is about half of that of the arsenosobenzene, while p-arsenosobenzoic acid has less than 1% the parasiticidal activity of arsenosobenzene.
An important property of the acid amide-substituted compound is a marked decrease in the toxicity to the host relative to the parasiticidal activity. The toxicity of p-arsenosobenzamide is 10% that of arsenoso
benzene, but as mentioned previously it has 50% of the parasiticidal activity of arsenosobenzene. The integrity of the amide group is essential for a favorable effect on toxicity, but the position of the substituent group on the benzene ring has little effect. The favorable therapeutic effect of amide-substituted arsenosobenzene has been used to advantage in chemo
therapy (71,72,85-87).
Esterification of the acid-substituted arsenosobenzene enhances both toxicity and parasiticidal activity (59).
Numerous other monosubstituted arsenosobenzenes have been syn
thesized. Few, however, have proved of great value in chemotherapy.
Melarsen oxide (p-[2,4-diaminotriazinyl-6]aminoarsenosobenzene) is an exception to this rule, being highly effective against trypanosomal infec
tions (88).
The effects of multiple substituents on the benzene ring have not been studied so extensively as those of the monosubstituents. With few excep
tions, the compounds studied are not much more useful than the parent compound, arsenosobenzene (59). Mapharsen, 2-amino-4-arsenosophenol, is one of the exceptions. Eagle et al. (89) tested the activity of a number of aminophenol-substituted arsenosobenzenes, R R ' C6H3A s 0 , where
20. SULFHYDRYL AGENTS'. ARSENICALS 115 R and R' are N H2 and OH, respectively: 3 - N H2, 4-OH (Mapharsen);
3-OH, 4 - N H2; 2 - N H2, 3-OH; 2-OH, 5 - N H2; 3-OH, 5 - N H2; and 2-OH, 3 - N H2. Only Mapharsen in this group possesses high parasiticidal activity and relatively low toxicity, and has been used successfully in chemo- therapy.
It is therefore apparent that substituents may greatly modify the bio- logical activities of an arsenical. Compounds have been obtained that have high parasiticidal activity combined with low toxicity to the animal.
While certain substituted arsenosobenzenes are active against one type of parasite, they may be considerably less active against another organ- ism. Clearly, in assessing the effects of substituents, specific factors must be considered, such as rate of penetration of the arsenical into a particular cell and its affinities for different thiol enzymes or thiol constituents of the cell.
REFERENCES
1. D. E. Koshland, Jr., Advances in Enzymol. 22, 45 (1960).
2. P. Ehrlich, Ber. 42, 17 (1909).
3. P. Ehrlich and A. Bertheim, Ber. 43, 917 (1910).
4. F. Hawking, T. J. Hennelly, and J. H. Quastel, J. Pharmacol Exptl Therap. 59, 157 (1937).
5. A. L. Tatum and G. A. Cooper, J. Pharmacol Exptl Therap. 50, 198 (1934).
6. C. Voegtlin and H. W. L. Smith, J. Pharmacol. Exptl Therap. 15, 475 (1920).
7. J. F. Shamberg, G. W. Raizess, and J. A. Kolmar, J. Am. Med. Assoc. 78, 402 (1922).
8. W. Yorke and F. Murgatroyd, Ann. Trop. Med. 24, 449 (1930).
9. H. Eagle, J. Pharmacol. Exptl. Therap. 66, 423 (1939).
10. M. J. Sartori, "The War Gases." Churchill, London, 1943.
11. C. Voegtlin, H. A. Dyer, and C. S. Leonard, U.S. Public Health Repts. 38, 1882 (1923).
12. H. J. Barber, J. Chem. Soc. p. 1020 (1929).
13. G. A. C. Gough and H. King, J. Chem. Soc. p. 669 (1930).
14. A. Cohen, H. King, and W. I. Strangeways, J. Chem. Soc. p. 3043 (1931).
15. J. J. Gordon and J. H. Quastel, Biochem. J. 42, 337 (1948).
16. F. G. Hopkins and M. Dixon, J. Biol. Chem. 54, 527 (1922).
17. J. B. Sumner and L. O. Poland, Proc. Soc. Exptl Biol Med. 30, 553 (1933).
18. L. Hellerman, Physiol Revs. 17, 454 (1937).
19. T. Bersin, Ergeb. Enzymforsch. 4, 68 (1935).
20. T. Bersin, Z. physiol. Chem. Hoppe-Seyler9 s 222,177 (1933).
21. F. G. Hopkins and E. J. Morgan, Biochem. J. 32, 611 (1938).
22. F. G. Hopkins, E. J. Morgan, and C. Lutwak-Mann, Biochem. J. 32, 1829 (1938).
23. L. Rapkine, Biochem. J. 32,1729 (1938).
24. M. Onada, Z. physiol. Chem. Hoppe-Seyler's 70, 433 (1911).
116 R. Μ. JOHNSTONE 25. A. Szent-Gyorgyi, Biochem. J. 24,1723 (1930).
26. C. Voegtlin, S. M. Rosenthal, and J. M. Johnson, U.S. Public Health Repts, 46, 339 (1931).
27. H. A. Krebs, Z. physiol. Chem. Hoppe-Seyler1s 217, 191 (1933).
28. R. A. Peters, L. A. Stocken, and R. H. S. Thompson, Nature 156, 616 (1945).
29. E. S. G. Barron and T. P. Singer, J. Biol. Chem. 157, 221 (1945).
30. E. S. G. Barron and T. P. Singer, Science 97, 356 (1943).
31. Ο. K. Reiss and L. Hellerman, J. Biol. Chem. 231, 557 (1958).
32. C. Lutwak-Mann, Biochem, J. 32, 1364 (1938).
33. E . S. G. Barron, Ζ. B. Miller, G. R. Bartlett, J. Meyer, and T. P. Singer, Biochem. J. 41, 69 (1947).
34. Α. Ο. M. Stoppani and A. S. Actis, Anales asoc. quim. arg. 40, 128 (1952).
35. Α. Ο. M. Stoppani, A. S. Actis, J. O. Deferrari, and E. L. Gonzalez, Bio
chem. J. 54, 378 (1953).
36. Α. Ο. M. Stoppani, A. S. Actis, J. O. Deferrari, and E . L. Gonzalez, Nature 170, 842 (1952).
36a. W. B. Jacoby, Arch. Biochem. Biophys. 70, 625 (1957).
36b. W. B. Jacoby, Proc. Natl. Acad. Sci. U. S. 46, 206 (1960).
36c. H. R. Mahler, B. Mackler, D. E. Green, and R. M. Bock, J. Biol. Chem.
210,465 (1954).
36d. R. W. Albers and G. J. Koval, Biochim. Biophys. Acta 52, 29 (1961).
36e. B. Mackler, H. R. Mahler, and D. E. Green, J. Biol. Chem. 210, 149 (1954).
37. E. C. Slater, Biochem. J. 45, 130 (1949).
38. T. P. Singer, Ε. B. Kearney, and H. Zastrow, Biochim. et Biophys. Acta 17, 154 (1955).
38a. V. R. Potter and C. A. Elvehjem, J. Biol. Chem. 117, 341 (1937).
38b. Ν. B. Das, Biochem. J. 31,1116 (1937).
39. R. Van Heyningen, Report to the Ministry of Supply, Great Britain, by M. Dixon, No. 10, 1942.
40. K. Bailey and E. C. Webb, Biochem. J. 42, 60 (1948).
41. P. B. Marshall, Brit. J. Pharmacol. 3, 8 (1948).
42. G. Chen, J. Infectious Diseases 82, 26 (1942).
43. G. E. Glock and P. McLean, Biochem. J. 55, 400 (1953).
43a. Β. Z. Dresel, Biochem. Z. 192, 35 (1927).
43b. Β. Z. Dresel, Biochem. Z. 178, 70 (1926).
44. T. P. Singer and E. S. G. Barron, J. Biol. Chem. 157, 241 (1945).
45. T. P. Singer, J. Biol. Chem. 174, 11 (1948).
46. L. H. Stickland, Biochem. J. 28, 1746 (1934).
47. R. Mamelak and J. H. Quastel, Biochim. et Biophys. Acta 12, 103 (1953).
48. J. S. Friedenwald and H. Herrmann, J. Biol. Chem. 146, 411 (1942).
48a. J. H. Quastel and Α. Η. M. Wheatley, Biochem. J. 29, 2773 (1935).
48b. M. Jowett and J. H. Quastel, Biochem. J. 29, 2189 (1935).
49. P. Rona and P. Gyorgy, Biochem. Z. I l l , 115 (1920).
50. P. Rona, Y. Airila, and A. Lasnitski, Biochem. Z. 130, 582 (1922).
51. L. A. Mounter and V. P. Whittaker, Biochem. J. 53, 167 (1953).
52. R. H. S. Thompson, Biochem. Soc. Symposia (Cambridge, Engl.) 2, 28 (1948).
20. SULFHYDRYL AGENTS'. ARSENICALS 117 53. C. Voegtlin, H. A. Dyer, and C. S. Leonard, J. Pharmacol. Exptl. Therap.
25,297 (1925).
54. L. A. Stocken, R. H. S. Thompson, and V. R. Whittaker, Biochem. J. 41, 47 (1947).
55. L. A. Stocken and R. H. S. Thompson, Biochem. J. 40, 535 (1946).
56. L. A. Stocken and R. H. S. Thompson, Biochem. J. 40, 529 (1946).
57. S. M. Rosenthal, U.S. Public Health Repts. 47, 251 (1932).
58. V. P. Whittaker, Biochem. J. 41, 56 (1947).
59. H. Eagle and G. 0 . Doak, Pharmacol. Revs. 3, 107 (1951).
60. H. Eagle, H. J. Magnuson, and R. Fleischman, J. Clin. Invest. 25, 451 ( 1 9 4 6 ) ; J. Wexler, H. Eagle, H. J. Tatum, H. J. Magnuson, and Ε. B.
Watson, ibid. 25, 467 ( 1 9 4 6 ) ; Μ. B. Sulzberger, R. L. Baer and A.
Kanof, ibid. 25, 474 ( 1 9 4 6 ) ; 25, 488 ( 1 9 4 6 ) ; W. Modell, H. Gold, and McK. Cattell, ibid. 25, 480 ( 1 9 4 6 ) ; A. B. Carleton, R. A. Peters, L. A.
Stocken, R. H. S. Thompson, and D. I. Williams, ibid. 25, 497 ( 1 9 4 6 ) ; W. T. Longcope, J. A. Luetscher, Jr., Μ. M. Wintrobe, and V. Jager, ibid. 25, 528 ( 1 9 4 6 ) ; J. A. Luetscher, Jr., H. Eagle, and W. T. Long- cope, ibid. 25, 534 ( 1 9 4 6 ) ; W. F. Hughes, Jr., ibid. 25, 541 ( 1 9 4 6 ) ; A.
Gilman, R. P. Allen, F. S. Philips, and E. St. John, ibid. 25, 549 (1946) ; W. T. Longcope and J. A. Luetscher, Jr., ibid. 25, 557 (1946).
61. L. A. Stocken and R. H. S. Thompson, Biochem. J. 40, 458 (1946).
62. L. J. Reed, in "The Enzymes" (P. D. Boyer, H. Lardy, and K. Myrback, eds.), Vol. 3, p. 222. Academic Press, N e w York, 1960.
63. Ο. K. Reiss, J. Biol. Chem. 233, 789 (1958).
64. D. R. Sanadi, M. Langley, and F. White, Biochim. et Biophys. Acta 29, 218 (1958).
65. D. R. Sanadi, M. Langley, and F . White, J. Biol. Chem. 234, 183 (1959).
66. R. A. Peters and L. A. Stocken, Biochem. J. 41, 53 (1947).
67. Ε . A. H. Friedheim and H. J. Vogel, Proc. Soc. Exptl. Biol. Med. 64, 418 (1947).
68. W. F. Riker, J. Pharmacol. Exptl. Therap. 87, Suppl. 66-67 (1946).
69. J. L. Sawyers, B. Burrows, and Τ. H. Maren, Proc. Soc. Exptl. Biol. Med.
70, 194 (1949).
70. F. Murgatroyd, Ann. Trop. Med. 31, 473 (1937).
71. G. F. Otto and Τ. H. Maren, Science 106,105 (1947).
72. G. F. Otto and Τ. H. Maren, Am. J. Hyg. 50, 92 (1949).
73. W. Hughes and G. A. Levy, Biochem. J. 41, 8 (1947).
74. R. H. S. Thompson, Biochem. J. 40, 525 (1946).
75. Η. H. Miiller, Naturwissenschaften 33, 253 (1946).
76. W. B. Savchuck, H. W. Loy, and S. S. Schiaffino, Proc. Soc. Exptl. Biol.
Med. 105, 543 (1961).
77. E. Shonbaum, Μ. K. Birmingham, M. Saffran, and E . Kurlents, Can. J.
Biochem. and Physiol. 34, 527 (1956).
78. G. S. Christiansen, L. J. Kunz, W. D. Bonner, Jr., and Ε . V. Thimann, Plant. Physiol. 24, 178 (1948).
79. D. E. Dolby, J. Gen. Microbiol. 12, 406 (1955).
80. A. Lacassagne, R. Royer, and G. Sudall, Comp. rend. soc. biol. 145, 1451 (1951).
118 R. Μ. JOHNSTONE 81. L. V. Beck, Cancer Research 9, 626 (1949).
82. L. V. Beck and R. L. Gillespie, J. Natl Cancer Inst. 12, 1223 ( 1 9 5 4 ) . 83. L. V. Beck, Proc. Soc. Exptl. Biol Med. 78, 392 (1951).
84. R. A. Roosa and E. D. DeLamater, Cancer Research 20, 1543 (1960).
85. J. D. Fulton and W. Yorke, Ann. Trop. Med. 37, 80 (1943).
86. Η. H. Anderson, V. P. Bond, and Β. E . Abreu, Federation Proc. 5, 162 (1946).
87. Η. H. Anderson and E . L. Hansen, Am. J. Trop. Med. 27,153 (1947).
88. Ε . A. H. Friedheim, Schweiz. med. Wochschr. 71, 116 ( 1 9 4 1 ) ; J. Am.
Chem. Soc. 66, 1775 ( 1 9 4 4 ) ; Ann. Trop. Med. Parasitol. 42, 357 (1948) ; Am. J. Trop. Med. 29, 173 (1949).
89. H. Eagle, G. O. Doak, R. B. Hogan, and H. G. Steinman, J. Pharmacol.
Exptl. Therap. 74, 210 (1942).
