Pharmaceutical Chemistry Journal

VoL 32. No. I. 1998

THIOUREA AND THIOSEMICARBAZIDE DERIVATIVES: STRUCTURE, TRANSFORMATIONS, AND PHARMACOLOGICAL ACTIVITY. PART IV. PROTECTIVE ACTION OF 1,2,4-TRIAZINOINDOLE DERIVATIVES WITH RESPECT TO PULMONARY EDEMA A. B. T o m c h i n I a n d A. V. K r o p o t o v t

Translated from Khimiko-Farmatsevticbeskii Zhurnal, Vol. 32, No. 1, pp. 2 2 - 26, January, 1998. Original article submitted December 10, 1996.

j : R l = R 2 -- H, R 3 = (CH2)3NMe2, n = 0; k: R I --- R 2 = H, R 3 = 3 - p i p e r i d i n o p r o p y l , n = I;

An important task of modem pharmacology is the search for new drugs providing prophylaxis and therapy for pulmonary edema, in particular, the toxic edema that is especially dangerous because of accumulation of a large amount of toxic substances (e.g., nitrogen oxides similar to those used for the military and commercial purposes). Irrespective of the reasons for the development of edemogenic processes in lungs, important factors of pathogenesis requiring immediate pharmacotherapy are hypoxia, increasing in the course of the edema is development, and activated lipid peroxidation. We may suppose that a protective effect with respect to pulmonary edema can be produced by antihypoxic agents optimizing respiration metabolism in the cells. However, these substances still have not been studied in this respect. The purpose of this work was to study the possibility of prophylaxis and therapy for pulmonary edema using 1,2,4-triazinoindole derivatives previously found to possess a high antihypoxic activity [1, 2]. For this purpose, we have synthesized a series of derivatives of 3-alkylthio-l,2,4-triazino[5,6-b]indole (Ia-Ir) and 3-alkylthio- 1,2,4-triazino[6,5-b]indole (IIa - IIc): R~r.-.-.-~

I: R t = H, R 2 = C H 2 C O N H 2, R 3 = (CH2)2NMe2, n = 0; m : R I = H, R 2 = C H 2 C O N H 2, R 3 = (CHz)2NEt2, n = I; n : R I = Me, R 2 = H, R 3 = (CH2)-zNMr z, n = 2; o : R I = Me, R 2 = 14, R 3 = (CH2)2NEt 2, n = 2;

p: R t = Me, R z = H, R 3 = 2 - m o r p h o l i n o e t h y l , n = 2; r: R I = O M e , R 2 = H, R 3 = 2 - m o r p h o l i n o e t h y l , n = 2; Iia: R = ( C H 2 ) 2 N M e 2, b: (CH2)2NEt2, e: 2 - m o r p h o l i n o e t h y l .

METHODS OF INVESTIGATION The synthesis of compounds I a - l r and l l a - I I c and the measurements of their physicoehemical characteristics log P~p, log P, and pKa were described previously [ 1, 2]. The biological tests were performed on white mongrel male mice weighing 1 8 - 24 g and on male rats weighing 180 - 200 g. The acute pulmonary edema model in the test animals was induced by two methods. The hydrodynamic edema was initiated by intraperitoneal injections of adrenaline hydrochloride [3]. The toxic pulmonary edema was modeled by placing animals in an inhalation chamber containing nitrogen oxides at a concentration of 4.3 mg/liter; the exposure duration was 10 and 6 min for mice and rats, respectively. The extent of the pulmonary edema development was characterized by gravimetric parameters. These included the lung coefficient (LC), representing the relative weight of lungs (expressed in % [4]) with respect to the total body weight, and the dry residue (DR) coefficient equal to the mass ratio of dry and raw lung tissues (also expressed as a percentage). These parameters provide an objective characterization of the liquid balance in lungs in the state of pulmonary edema [5]. Besides this, we have determined the percentage survival of test mice during one day and calculated the protection coefficient (PC) [6]. The adequacy of the pulmonary edema modeling and the resuits of therapy were checked by histological methods. The test compounds were introduced by intraperitoneal injections (insoluble compounds, in the form of fine suspen-

N ~N HCI "-N /

-. N / / " - . S R 3

N~ N

"

I

H la - l r

lla-

Ilc

la: R t = R 2 = H, R 3 = C H 2 C H = C H 2, n = 0; b: R t = R 2 = H, R 3 = C H P h C O P h , n = 0 ; c: P) = R 2 = H, R 3 = C H 2 C O O H , n = 0; d: R t = R 2 = H, R 3 = C H 2 C O O E t , n = 0; e: R ! = R 2 = H, R 3 = C H 2 C O N E t 2, n = 0; f: R I = R 2 = H, R 3 = p i p e r i d i n o c a r b o n y l m e t h y l , n = 0; g: R t = R 2 = H, R 3 = (CH2)2NH 2, n = 0; h: R t = R 2 = H, R B = ( C H 2 ) 2 N M e 2, n = 1; i: R I = R 2 = H, R 3 = (CH2)2NEt 2, n = I;

t State Medical U n i v e r s i t y , V l a d i v o s t o k , Russia.

20 0091-150X/98/3201-0020520.00 9 1998 Plenum Publishing Corporation

Thiourea and Thiosemicarbazide Derivatives: Structure, Transformations, and Pharmacological Activity

21

troi and test groups was based on the Student criterion anc sions containing Tween-80) 30 min before modeling the performed by the correlation analysis methods [13]. The acute pulmonary edema. The drugs were injected at optimum processing of correlated parameters was carried out on ar doses established preliminarily during the study of their antiHP- 1000 computer. hypoxic action [1, 2]. Animals in the control group were injected with the same volume of solvent. The reference preparations were represented by sodium hydroxybutyrate [7] and RESULTS AND DISCUSSION the antihypoxic drug amtizole [6]. The membranotropic effect of the antihypoxic agents As is seen from the experimental data presented in Tastudied was assessed by the methods of acid and osmotic ble 1, compounds Ia, Ie - Ih, lj, I!, In - Ip, and IIa offer prohemolysis of mice erythrocytes. The resistance of erythrotection against the adrenaline-induced pulmonary edema, cytes was determined by the acid erythrogram technique [8] with the protection coefficients PC = 1.30- 1.63, and prousing 0.024 M hydrochloric acid as a hemolytic agent. The duce a more than threefold increase in the survival of test osmotic strength of erythrocytes was studied in the buffered mice. Of maximum interest in this group are the compounds hypotonic solutions of sodium chloride [9]. The dynamics of lipid peroxidation was studied on rats as described in [10]. In this series of TABLE I, Protective Action of Triazinoindole Derivatives on the Acute Adrenaline-Induced Pulmonary Edema experiments, the reagent concen- Model in Mice trations were as follows: sodium Dose, Number Survival, % PC LC, % DR, % Compound m g / kg of animals hydroxybutyrate, 250 mg / kg; 20 100 0.92 _+0.06 22.3 _+0.41 amtizole, 40 m g / k g ; compounds Intact mice 15 72 15 2.13 _+0.25* 13.1 _+ 1.00" If and lj, 12.5 m g / k g . In 10 or Adrenaline (control I) 20 rain after the introduction of 15 60 I0 2.18_+0.32" 12.9_+0.77" adrenaline, the control and test Adrenaline + Tween-80 animals were slightly narcotized (control II) with ether and killed, after which Sodium 62.5 20 40 !.27 1.66 _+0.11 17.2 + 1.34"* hydroxybutyrate the lung lipids were extracted and 125 20 50 !.30 1.47 _+0.05** 17.4 -+ 1.08"* identified. Animals with the toxic 250 20 60 1.39 1.26 + 0.15"* 15.8 -+ 0.43** pulmonary edema model were 500 20 50 1.30 1.25 • 0.08** 17.3 • 0.97** killed 60 or 120 rain after withdrawal from the inhalation cham- Amtizole 40 40 40 1.27 1.45 -+ 0.10"* 16.9 + 0.82** ber. The extraction of lipids from la 100 20 50 1.30 1.39 -+ 0.33 16.1 + 1.00"** blood serum and lung tissues was lb 12.5 20 10 1.00 1.62 -+ 0.14 15.3 -+ 0.93 performed by the conventional Ic I00 20 30 1.18 1.83 -+ 0.43 15.1 --- 2.10 methods [11, 12]. The lipid ex- Id 50 24 25 1.14 1.72 + 0.06 16.8 + 0.74*** tracts were used for analysis of the le 12.5 20 60 1.51 1.91 + 0.76 15.4 + 1.10 total lipids, phospholipid and neu- If 6.25 20 30 1.18 1.62+0.20 14.1 _+0.82 tral fractions. The dynamics of 12.5 36 72.5 1.63 1.45 _+0.24 17.5 _+ 1.00"** variation of the lipid peroxidation lg 12.5 20 70 1.54 1.87 _+0.71 12.5 _+0.40 parameters during development of lh 25 20 50 1.30 1.72 _+0.17 17.2 _+0.95** the pulmonary edema models, in- li 12.5 20 0 0.87 1.73 + 0.13 15.7 _+ 1.12 duced by adrenaline chloride or nilj 12.5 20 60 1.51 1.64 _+0.15 17.9 _+0.49*** trogen oxides, and in the course of Ik 12.5 20 15 1.05 2.00 _+0.62 14.2 _+ 1.83 the hyperhydration process cor11 12.5 20 50 t.30 1.83 _+0.48 15.6 _+2.10 rected by antihypoxic agents were lm 12.5 20 30 1.18 2.22 _+0.36 16.7 +_ 1.83 studied in vivo by analyzing the In 12.5 20 60 1.51 2.13 _+0.37 14.4 _+ 1.82 blood and lung tissues. to 50 20 50 1.30 2.00 _+0.55 15.9 + 1.25 In order to estimate the pharIp 50 20 50 1.30 1.95 _+ 0.30 15.9 _+ 1.33 macological breadth of the synthelr 12.5 24 33 1.15 1.77 _+ 0.38 15.3 _+ 1.42 sized compounds, we have studied lla 12.5 20 50 1.30 1.66 _+ 0.41 20.0 _+0.73*** their acute toxicity (LDs0) for the lib 12.5 28 15 1.00 2.13 _+0.25 16.1 _+2.07 intraperitoneal introduction. 25 20 30 I. 18 2.15 _+0.34 14. I _+ 1.42 The results of experiments llc were mathematically processed us- Notes. Here and in Table 2, * denotes a difference statistically significant against the data for intact mice ( p < 0.05), ing the methods of variation statis- 9* denotes a difference statistically significant against the data for control 1 ( p < 0.05), and *** denotes a tics. The comparison between con- difference statistically significant against the data for control II ( p < 0.05).

22

A.B. Tomchin and A. V. Kropotov

In studying the relationship between activity and structure, it is convenient to divide the triazinoindole derivatives into water-soluble and insoluble. The former group includes the compounds containing side-chain amino fragments with a rather high basieity, facilitating the formation of water-soluble hydrochlorides. The latter group, representing compounds not possessing sufficient basicity (Table 3), includes compounds I a - If(all other compounds entering into the first group). We may conclude that the most interesting compound among the whole series of triazinoindole derivatives studied is (1,2,4-triazino[5,6-b]indolyl-3-thio)acetic acid piperidide (If). This compound exhibits a universal and high protective activity with respect to both adrenaline-induced and toxic pulmonary edema (PC = 1.63 and 1.48, respectively) and with respect to hypoxia in an altitude chamber (PC = 1.54) [ 1, 2], in combination with low toxicity and high therapeutic ratio. Unfortunately, compound If is insolTABLE 2. Protective Action of Triazinoindole Derivatives on the Acute Toxic Pulmonary Edema Model in Mice uble in water. Among the soluble Dose, Number Survival, % PC LC, % DR, % Compound m g / kg of animals compounds, the maximum activity with respect to all three models in0.92 • 0.06 22.3 • 0.41 Intact mice 20 100 dicated above was observed for Nitrogen oxides 3(3-dimethylaminopropylthio)1, 2.05 +_0.14" 12.3 • 0.24* (control I); 120 16.6 2,4-triazino[5,6-b]indole hydroNitrogen oxides + Tween-80 chloride (Ij) possessing sufficient (control II) 60 5 2.11 +_0.08* 13.2 +_0.54* pharmacological breadth (albeit Sodium lower compared to that of com1.03 2.04 • 0.06 12.2 + 0.49 hydroxybutyrate 125 20 20 pound If). It should be noted that 250 40 30 I.! 1 !.59 + 0.14"* 14.9 + 0.73 compound If and Ij (as well as 500 40 30 1.11 1.50 +- 0.12"* 14.8 + 0.59** some other 1,2,4-triazino[5,6-b]in1.29 1.72 + 0.11 14.9 +- 0.43** Amtizole 40 80 50 dole and 1,2,4-triazino[6,5-b]inla ! 00 20 30 1.24 1.82 • 0.28 13.9 + 0.94 dole derivatives) are superior to 1.14 2,20 • 0.10 12.1 +- 0.57 Ib 12.5 40 20 amtizole and sodium hydroxy1.24 !.82 • 0.23 14.0 + 0.72 Ic 100 20 30 butyrate with respect to the PC 1.65 1.47 • 0.13"** 13.9 • 0.83 Id 50 20 70 value. 1.48 1.90 +- 0.27 12.9 +- 0.86 le 12.5 20 50 Thus, our experimental data 1.43 2.09 +- 0.14 12.9 + 0.54 If 6.25 20 50 confirmed the assumption con12.5 60 55 1.48 1.56+_0.13"** 16.2+_ 1.12"** cerning protective properties of the 25 20 60 1.65 1.66+0.11"** 14.4+0.52 antihypoxic agents with respect to lg 12.5 20 70 1.65 1.87 + 0.71 12.5 + 0.50 the pulmonary edema. However, Ih 25 20 30 1.11 1.76 +_0.32 14.2 • 0.65 the correlation analysis revealed li 12.5 20 30 1.24 1.43 + 0.24** 15.7 -+ 1.12 no reliable relationship between lj 12.5 20 60 1.65 1.86 + 0.31 15.1 • 0.69*** the PC values of the triazinoindole Ik 12.5 20 20 1.14 1.84+_0.17 13.8+_0.43 derivatives with respect to the hyII 6.25 20 70 1.65 2.30 • 0.13 10.1 • 0.32 poxic hypoxia in the altitude 12.5 40 55 1.48 1.80 + 0.12 12.5 + 0.59 chamber [1, 2], on the one hand, 25 2O 70 1.65 2.30 +- 0.13 10.1 • 0.96 and the PC values for the adIm 12.5 20 20 1.14 1.91 + 0.16 13.6 +-0.30 renaline-induced and toxic pulmoIn 12.5 20 20 1.14 1.82 +-0.26 15.3 • nary edema models on the other Io 50 20 30 124 1.92 • 0.36 13.5 + 0.54 hand. Therefore, realization of the Ip 50 20 30 1.24 1.96 -+ 0.48 13. I +- 0.47 protective activity of the antihylq 12.5 20 30 1.11 1.28 +_0.12"* 16.5 + 0.72 poxic agents in the cases of hylla 12.5 20 70 1.65 1.66 + 0.16"** 14.3 _+0.51 poxia caused by different factors lib 12.5 40 30 1.14 1.66+_0.19"** 14.1 +_0.64 have certain distinctive features lie 25 20 20 1.14 1.90 • 0.43 13.5 +- 0.56 determined, probably, by the phar(If, lj, lla) capable of decreasing the parameters of lung hydration. Note that some other compounds (le, lg, In - lp) affect the LC and DR values rather insignificantly. With respect to the toxic pulmonary edema model, the triazinoindole derivatives studied can be divided into three groups offering weak (PC= 1.03-1.14), medium (PC= 1.24 - 1.33), and satisfactory (PC = 1.37- 1.65) protective action (Table 2). The second group includes compounds Ia, Ic, Ii, Io, and Ip; it is worth noting that compound Ii produced a 32% decrease in the LC value and a 27% increase in the DR coefficient (p < 0.05). The third group includes compounds l d - Ig, Ij, I1, and IIa. Note that compounds Ie, Ig, and I1, albeit providing the survival of 5 0 - 7 0 % animals, increase (rather than decrease) the tendency toward increased the lung hydration.

Thiourea and Thiosemicarbazide Derivatives: Structure, Transformations, and Pharmacological Activity

macodynamic spectrum of the particular compounds. At the same time, the correlation analysis of PC values for the adrenaline-induced and toxic pulmonary edema showed unidirectional protective action o f the compounds studied, with the correlation coefficient for the entire series of compounds being r = + 0.54 (p < 0.05). All triazinoindoles (except for compounds Id, I11, II, In, and IIa) obey a linear relationship between the PC values for the adrenaline-induced (PCa) and toxic (PCt) pulmonary edema: PCt = 1.021PCa (F = 4262, p < 0.0001, r = 0.996). All the coefficients are statistically significant forp < 0.0001. The presence of a correlation between PC values for the two pulmonary edema models is evidence for the existence of universal factors in pathogenesis of the edemogenic states o f lungs and similar mechanisms of its correction by the antihypoxic agents studied. By contrast, no reliable correlation was found between the parameters of survival (PC) and the LC and DR values, characterizing the extent of edematous changes in lungs. This fact is indicative o f a possibility that the protective action of the compounds studied is realized on the level of the whole organism, apparently at the expense of increasing the tissue resistance with respect to hypoxia. When the pairs of correlators corresponding to compounds Ib, Id, If, and IIa are excluded from the analysis of data for the toxic pulmonary edema model, the coefficient of correlation between PC and LC for the remaining pairs is r = + 0.5 (p < 0.05). As noted above, the survival of test animals upon prophylactic administration of the triazinoindole derivatives can be sufficiently high even if a considerable amount of transudate is present in the lungs. An analysis of the structure-activity relationship revealed the following correlations. The site of condensation of the triazine and indole nuclei in the triazinoindole derivatives studied has no significant effect on their protective action with respect to the pulmonary edema models. Exceptions are the dimethylaminoethylthio derivatives Ih and IIa, for which the 1,2,4-triazino[6,5-b]indole derivatives are superior to the [5,6-b]-counterparts with respect to the toxic pulmonary edema. In the series of soluble aminoalkylthio derivatives, the compound with an unsubstituted amino group (Ig) exhibited high activity with respect to both pulmonary edema models studied. Substituting alkyl radicals for hydrogen in the amino group led to a decrease in the activity. Of the alkyl-substituted compounds studied, the maximum activity was observed for the dimethyl derivatives, while the passage to ethyl and morpholine derivatives was accompanied by a decrease in the activity. Inclusion of the terminal nitrogen atom of a side aminoalkyl chain into the piperidine ring not only reduced the activity, but increased the toxicity as well. In compounds containing two methylene units between the amino group and the sulfur atom, introduction of the third methylene unit increased the protective action. The same effect was achieved upon substituting a carbonyl group for one of the methylene units adjacent to the amino group. This was

2~

also accompanied by a sharp drop in the toxicity and, accord ingly, by an increase in the therapeutic ratio. However, thL substitution decreased the basicity and, hence, the solubiliu in water. Substituting a carbamidomethylene group for hydroger at the indole nitrogen in the active compounds containing dimethylamino group in the side chain retained the level oi activity with respect to the adrenaline-induced pulmonary edema and increased the activity toward the toxic pulmonary edema model. Replacement of the hydrogen atom in positior 8 by the methyl radical also favored an increase in the protective action. In general, the above correlations agree with the conclusions made upon analysis of the relationship between structure and the antihypoxic activity [ 1]. Exclusions are only provided by data on the effect o f methyl radical in position 8 and the length of the side aminoalkyl chain. We have also studied the interrelation between the protective action of triazinoindole derivatives with respect to the acute pulmonary edema and the lipophilicity of these compounds. For the adrenaline-induced pulmonary edema model, the water-soluble compounds (except lib and the 8-methyl derivatives I n - I p ) were found to obey an approximation of the following type: 3

5

PC a = 1.16 + 0.973 log p2ap- 0.684 log Pap + 0.045 log Pap (F = 10.50, p < 0.0055, r = 0.82). All the coefficients are statistically significant for p < 0.008. The water-insoluble compounds (except for Ic) obey a different relation:

T A B L E 3. Physicochemical Characteristics and Acute Toxicity o f Triazinoindole Derivatives Compound

log Pap

log P

pK a

ia

3.48

3.48

...

Ib

4.79

4.79

...

Ic

-1.07

ld

2.29

2.29

...

> 25,600

le

2.54

2.54

...

3400

If

2.77

2.77

...

Ig

0.84

2.27

8.77

> 25,600

lh

1.53

2.35

8.08

384

li

1.81

3.05

8.55

250

Ij

1.15

2.60

8.74

560

Ik

2.01

3.77

9.11

< 50

I1

0.60

1.29

7.95

1720

Im

0.10

1.30

8.54

450

In

2.04

2.94

8.18

450

Io

2.51

3.77

8.60

280

lp

3.05

3108

6.17

550

Ir

2.62

2.64

6.04

1300

IIa

1.76

2.59

8.12

...

lib

1.94

3. I 1

8.50

246

Ilc

2.55

2.57

6.00

325

.

.

.

.

.

.

LDs0, mg / kg > 25,600 1450 .

.

.

4500

24

A.B. Tomchin and A. V. Kropotov

,'C a = - 21.44 + 20.68 log Pap - 5.99 log P a2p+ 0.554 log Pa3p ( F = 841.3, p -<0.025, r = 0.999). All the coefficients are statistically significant for p _<0.02). For the toxic pulmonary edema model, the water-soluble compounds (except Ih, Ip, and IIa) obey the the following correlation: PC, = 0.96 + 1.978 log Pap- 1.624 log P~p + 0.342 log p3p (F = 18.38, p < 0.0006, r = 0.87), the coefficients being statistically significant for p < 0.0013. A correlation for the insoluble compounds is as follows: PC, = 0.851 log P~ - 0.266 log Pa3p+ 0.0043 log P~ (F= 349.5, p < 0.0001, r = 0.997), where all the coefficients are statistically significant for p < 0.003). Note that eoeffi:ients at log P~ in the first and last relationships are very small and these terms may be neglected. Thus, the relationship between PC values and logarithmic apparent distribution :oeftieients in most of the compounds studied is satisfactorily described by an equation of the third-order parabola, the approximations for the water-soluble and insoluble compounds King different. Each of the two groups is characterized by its ,wn region of optimum lipophilieity, in which a high protective activity is most probable: the region for the insoluble :ompounds is shil~ed toward higher values of the distribution :oefficients. For the pulmonary edema models of both types, the most active water-soluble compounds are found in the region of log Pap = 0.6 - 1.8. In insoluble compounds, the optimum range of log Pap is 2.6 - 3.5 for the adrenaline-induced Lpulmonaryedema and 2.3 - 2.8 for the toxic edema. The above correlations include the PC values, that is, refer to the survival characteristics. The gravimetric parameters LC and DR are less correlated to the lipophilicity. Nevertheless, the soluble compounds (except Ip, Ir, and IIb) tested on the adrenaline-induced pulmonary edema model obey the following relation: PCa= 2.234- 1.556 log P~:p+ 1.256 log p3p-0.257 log Pa4 F = 8.70, p -< 0.009, r = 0.79). All the coefficients are statistically significant for p _<0.014. The water-insoluble compounds obey the relation PCap = 0.19 log Pap + 3.0(log Pap)- 1 (F= 177.2, p < 0.0008, r = 0.992). All the coefficients are statistically significant forp -<0.038). Thus the characteristics of lung hydration also have a region of optimum lipophilicity. It is interesting to note that, similarly to the case of antihypoxie activity [1], the protective action with respect to the pulmonary edema is best correlated to the apparent distribution coefficients. This gives evidence for a significant role of the ionization process. In the group of soluble compounds, no correlation is found between the protective activity and pK~

values. However, the most active compounds fall within a rather narrow range ofpKa = 7 . 9 - 8.8. This allows us to conelude that a region of optimum basicity still exists, being a necessary condition for the high activity. The lack of reliable correlation between PC of the triazinoindole derivatives with respect to the hypoxic hypoxia and the pulmonary edema models of different origin suggests that membranotropic effects probably are involved in the mechanism of protective action. Indeed, the "chaotropie" changes observed in the lipid spectrum [14] and a decrease in the level ofantioxidant protection factors in lungs, revealed in the state of developed acute edema, are indicative of increased membrane permeability. Ordering of the structural components of pulmonary membranes and of the lipid peroxidation process may contribute to the protective activity. In connection with this, we have studied the membrane-stabilizing action of compounds Ij, If and the reference drugs in the experiments in vivo and in vitro. It was established that all these compounds inhibit the process of osmotic and, even more, oxygen-induced hemolysis of erythrocytes, thus producing a membrane-stabilizing effect. This conclusion was confirmed on the mitochondrial model of nonenzymatic lipid peroxidation. Preliminary introduction of antihypoxic agents into a suspension of mitochondria prevented the accumulation of malonic dialdehyde. Similar results were observed for these compounds (except amtizole) on the model of enzymatic lipid peroxidation in mitochondria. In the experiments on rats, a property in common for compounds Ij, If, amtizole and sodium hydroxybutyrate with respect to the acute pulmonary edema, irrespective of the etiology of the edemogenic process development in lungs, is the ability to decrease the level of phospholipid lysoforms and increase the total phospholipid fraction in the total lung lipids in the state of developed pathology. Judging from the protective activity of antihypoxic agents in general, as manifested in the experiments on acute pulmonary edema models of different origin described above, we have concluded that, like sodium hydroxybutyrate and amtizole, antihypoxic agents of the triazinoindole series produce a membrane-stabilizing effect by normalizing the lipid spectrum in lungs and controlling the process of lipid peroxidation. These effects significantly contribute to the mechanism of protective action of the triazinoindole derivatives with respect to the acute pulmonary edema. ACKNOWLEDGMENTS The authors are grateful to T. I. Zhukova and T. A. Kuznetsova for their help in the experimental chemical part, and to V. N. Butakov and O. Yu. Uryupov, in the experimental pharmacological part of the work. REFERENCES 1. A. B. Tomchin, O. Yu. Uryupov, T. I. Zhukova, et al., Khim.Farm. Zh., 31(3), 19-27 (1997).

T h i o u r e a a n d T h i o s e m i c a r b a z i d e Derivatives: S t r u c t u r e , T r a n s f o r m a t i o n s , and Pharmacological A c t i v i t y

2. A. B. Tomchin, O. Yu. Uryupov, and A. V. Smirnov, Khim.Farm. Zh.,31(12),6- 11 (1997). 3. A. Kh. Kogan, L. O. Luk'yanova, and A. N. Kudrin, Patol. Fiziol., No. 1, 5 0 - 5 6 (1982). 4. N.C. Staub, Physiol. Rev., 54, 6 7 8 - 8 1 1 (1974). 5. V. B. Serikov, V. V. Rozental', and N. A. Belyakov, Patol. Fiziol., No. 1 , 4 0 - 4 5 (1986).

6. Methodological Recommendations on the Experimental Study of Preparations Proposed for Clinical Testing as Antihypoxic Agents [in Russian], L. D. Luk'yanova (ed.), Minzdrav SSSR, Moscow (1990), p. 8. 7. M. I. Zverev and M. Ya. Anestiadi, Toxic Pulmonary Edema [in Russian], Shtiintsa, Kishinev (1981).

25

8. I. I. Gitel'zon and I. A. Terskov, in: Problems in Biophysics Biochemistry, and Pathology of Erythrocytes [in Russian], Mos cow (1967), pp. 41 - 4 8 . 9. N. L. Vasilevskaya, Byul. Exp. Biol., 40(I 2), 68 - 72 ( i 955). 10. E. M. Kreps, in: Brain Lipid Evolution: Adaptation Function o Lipids [in Russian], Nauka, Leningrad (198 ! ). 11. J. Folch, M. Less, and G. H. S. Stanley, J. Biol. Chem., 226 497 - 509 (1957). 12. E. L. Bligh and W. L. Dyer, Can. J. Biochem. Physiol., 37, 911 917(1959). 13. V. I. Metelitsa, A Cardiologist "s Handbook of Clinical Pharma cology [in Russian], Meditsina, Moscow (1987). 14. Yu. A. Rakhmanin, A. V. Kropotov, N. F. Kushnerova, et al.. Gigiena Sanit., No. 6, 3 - 5 (1995).