NEW AMINOPHOSPHONATES WITH ANTIOXIDATIVE ACTIVITY

CELLULAR & MOLECULAR BIOLOGY LETTERS

Volume 6, (2001) pp 83 – 91 Received 18 January 2001 Accepted 23 March 2001

NEW AMINOPHOSPHONATES WITH ANTIOXIDATIVE ACTIVITY HALINA KLESZCZYNSKA and JANUSZ SARAPUK Department of Physics and Biophysics, Agricultural University, Wroclaw, Norwida 25, 50-375, Poland

Abstract: The antioxidative activity of some newly synthesized aminomethanephosphonic acid derivatives was studied. The compounds studied differed in their polarity and the hydrophobicity of the electronic substituents at their nitrogen and phosphorus atoms. It was found that all the aminophosphonates studied, both cyclic and acyclic, protected erythrocyte membranes against peroxidation to some extent. The effect was somewhat weaker in the case of cyclic compounds, and for erythrocytes irradiated with UV light. The cyclic compounds provided no protection of erythrocytes illuminated by natural light. The observed differences between the antioxidative activities of cyclic and acyclic compounds are probably related to differences in their ability to incorporate into the lipid phase of erythrocyte membranes. Once incorporated, they change the fluidity of the membranes. The extent of those changes was determined in fluorescence measurements. Generally, they were found to be more pronounced in the case of acyclic aminophosphonates, although as regards other structural differences between particular aminophosphonates, a clear picture of the relationship between structure and effect is more difficult to obtain. No correlation was found between the antioxidative efficiency of the compounds and the fluidity changes they induce. Key Words: Aminophosphonates, Antioxidative Activity, Erythrocytes INTRODUCTION Organophosphorous compounds have been known for over 50 years [1]. Many of them exhibit biological activity and are widely used as potent herbicides. Perhaps the best known are glyphosate [2], Trakephon [3] and amino-

84

CELL. BIOL. MOL. LETT.

Vol. 6. No. 1. 2001

phosphonic acid derivatives of fluorene [4, 5]. It was suggested that the biological activity of these latter is correlated to their lipophilicity [5] and hence, their ability to incorporate into the lipid phase of biological membranes. This leads to the destruction of the membrane when the concentration of the lipophilic compound is high enough. On the other hand, the presence of various compounds in membranes in low concentrations was shown to stabilize the membranes [6, 7]. It is also possible that some compounds applied at these low concentrations may exhibit other useful features. Indeed, it has been shown that some aminophosphonates can be used as antioxidants [8]. This work is a continuation of the studies on the potential use of aminophosphonates as antioxidants. Pig erythrocytes were used for this reason, and the compounds studied were used at sublytic concentrations. It was already shown that erythrocytes are a very convenient model for this type of experiment, as aminophosphonates readily interact with them due to their lipophilic character. MATERIALS AND METHODS The compounds studied were synthesized in the Department of Organic Chemistry, Biochemistry and Biotechnology of the Technical University of Wroc³aw. Heating the carbonyl compound with a corresponding amine yielded an imine, which was used without purification in the next step. After the addition of dialkyl phosphite to the imine, the reaction mixture was heated for Tab. 1. The structure and substituent groups of the acyclic aminophosphonates. R

1

C

R

2

OR HN 4

R

Compound no. 1 2 3 4 5 6 7 8 9

P

3

3

OR O

R1

R2

R3

R4

CH3 CH3 CH3 CH3 n-C5H11 CH3 CH3 CH3 CH3

CH3 CH3 n-C3H7 n-C4H9 CH3 CH3 n-C5H11 CH3 t-C4H9

CH3 i-C3H7 n-C4H9 CH3 C2H5 n-C4H9 n-C4H9 n-C4H9 n-C4H9

n-C4H9 n-C4H9 n-C4H9 n-C4H9 n-C4H9 n-C10H21 n-C8H17 n-C14H29 n-C4H9

CELLULAR & MOLECULAR BIOLOGY LETTERS

85

several hours. The final product was isolated and purified by column chromatography. The purity was checked by 1H-NMR and 31P-NMR spectra. The spectral data are given elsewhere [9, 10]. The general structure of the aminophosphonates studied and the particular substituents at the carbon, phosphorus and nitrogen atoms are shown in Tables 1 (acyclic compounds) and 2 (cyclic compounds). The oxidation studies were performed on erythrocyte ghosts prepared according to [11]. Lipid peroxidation in the RBC membrane was induced by natural illumination or UV irradiation (3.5 mW/cm2 lamp intensity) and measured after 1.5 h. The incubation time was 2 hours, and the temperature was 37°C. The degree of lipid peroxidation was determined by measuring the concentration of malonic dialdehyde released in the samples. Further experimental details were given earlier [12]. Tab. 2. The structure and substituent groups of the cyclic aminophosphonates.

(CH2 )n R1 NH

P(O)(OR2 )2

Compound no.

R1

R2

n

10 11 12 13 14 15 16

n-C4H9 n-C8H17 n-C8H17 i-C4H9 n-C8H17 (CH2)2OH C(CH2OH)2CH3

i-C3H7 i-C3H7 C2H5 n-C4H9 n-C4H9 n-C4H9 C2H5

2 1 2 2 1 2 2

Fluidity experiments were done on erythrocyte ghosts, which were subjected to the action of the compounds studied. The concentrations used were the same as those in the oxidation experiments. The fluorescent probe used was TMAPDPH at 1 µM concentration. The measurements were performed with an SFM 25 spectrofluorometer (KONTRON) at 25°C. The excitation and emission wavelengths were 354 nm and 429 nm, respectively. The anisotropy coefficient A was calculated according to the formula [13-15]: A = (III - GI⊥ )/(III + 2GI⊥)

(1)

86

CELL. BIOL. MOL. LETT.

Vol. 6. No. 1. 2001

where III is the intensity of fluorescence emitted in a direction parallel to the polarization plane of the exciting light, I⊥ is the intensity of fluorescence emitted in the perpendicular direction, and G is the diffraction constant. All reagents used were of analytical grade. The fluorescent probe N-((4-(6phenyl-1,3,5-hexatrienyl)phenyl)propyl)trimethylammonium p-toluenesulfonate) (TMAP-DPH) was purchased from Molecular Probes Inc. (Eugene, USA). The partition coefficients (P) were calculated as follows. Hemolytic curves were obtained for four hematocrits, and then used to plot a dependence between those hematocrits and the concentration of the compounds. The obtained linear relationships were used to calculate the concentrations of the compounds in the RBC membrane and the incubation solution. P was then calculated as a ratio of those concentrations. RESULTS AND DISCUSSION The results of antioxidation studies are shown as bar diagrams in Figs. 1-3. Cyclic compounds were not found to give any antioxidative protection of erythrocyte membrane lipids when natural light illumination was applied. It can be seen that the acyclic compound efficiency to protect erythrocytes against peroxidation was better when natural light was used. This is especially evident for compounds 3, 4 and 5 (Figs. 1 and 2). These compounds have the largest hydrophilic substituents at the carbon atom, and, in the case of compounds 4 and 5, hydrocarbon substituents at their phosphorus atom are shortest (Tab.1).

Inhibition [%]

60

0,01 mM 0,02 mM

50 40 30 20 10 0 1

2

3

4

5

6

7

8

9

Compound number

Fig. 1. The results of studies on the antioxidative efficiency of acyclic aminophosphonates with natural light illumination.

Higher energy light (UV) seems to remove any structural advantages. Some general conclusions about aminophosphonate activity can be drawn, which are independent of what kind of illumination was applied. Namely, the weakest

CELLULAR & MOLECULAR BIOLOGY LETTERS

87

antioxidative efficiency was exhibited by cyclic aminophosphonates with hydroxyl and branched or short hydrocarbon substituents at their nitrogen atom (compounds 13, 15 and 16; Tab. 2). An elongation of the hydrocarbon chain at the same position seems to enhance the protective properties of compound 6 (Tab.1, Figs. 1 and 2), however, when it becomes too long, this effect is negated (compound 8; Tab. 1, Figs. 1 and 2). The optimal chain length seems to oscillate between C10H21 and C4H9. This optimum effect is enhanced if a chain of such length is accompanied by an isopropyl substituent at the phosphorus atom (compounds 2, 10 and 11; Tabs. 1 and 2). It is possible that such a substituent makes a compound more bulky in the vinicity of its phosphorus

Inhibition [%]

60

0,01 mM 0,02 mM

50 40 30 20 10 0 1

2

3

4

5

6

7

8

9

Compound number

Inhibition [%]

Fig. 2. The results of studies on the antioxidative efficiency of acyclic aminophosphonates with UV irradiation.

40 35 30 25 20 15 10 5 0

0,01 mM 0,02 mM

10

11

12 13 14 Compound number

15

16

Fig. 3. The results of studies on the antioxidative efficiency of cyclic aminophosphonates. UV irradiation was applied.

88

CELL. BIOL. MOL. LETT.

Vol. 6. No. 1. 2001

atom, thus improving the protection of the lipid phase which the compound is incorporated into. The effect must be connected with conformational differences between compounds with isopropyl and transbutyl chains rather than with changes in the lipophilicity of these compounds (compare compounds 11 and 14; Tab. 2 and Fig. 3). The substitution of C5H11 or t-butyl group at the carbon atom diminishes the protective efficiency of aminophosphonates (5 and 7; Tab.1, Figs. 1 and 2). It must be emphasized that in all cases where the substituents at the phosphorus and nitrogen atoms were the same, acyclic compounds were clearly more effective than cyclic ones in the protection of erythrocytes against peroxidation (compare compounds 2 and 10 or 7 and 14; Tabs. 1 and 2, Figs. 2 and 3). It also seems that cyclohexane aminophosphonates were slightly more effective than cyclopentane aminophosphonates (compounds 10 and 11; Tab. 2, Fig. 3). It has previously been shown that the antioxidant activity of various bifunctional surfactants depended on the possibility of their incorporation into erythrocyte membranes to different depths [16, 17]. In order to find out if this is the reason for the differentiated antioxidative efficiency of the compounds studied, fluorescence studies were performed employing a TMAP-DPH probe. The calculated anisotropy of the fluorescence probe incorporated into erythrocyte ghosts should be the measure of fluidity changes in the ghost membranes induced by aminophosphonates inbuilt into these membranes. The percentage changes in fluorescence anisotropy relative to control are collected in Tables 3 and 4. Some hemolytic experiments were also performed enabling the calculation of the ratio of the number of incorporated molecules of a compound to the number of molecules remaining in the incubation solution P (see Materials and Methods). As can be seen (Tables 3 and 4), the values of P calculated for the majority of the compounds studied match the anisotropy changes ∆A quite closely. This is most evident in the case of the weakest (nos. 1, 7, 10 and 16) and strongest acting compounds (nos. 4, 6 and 13). It seems that changes in anisotropy can be used as a measure of the partition coefficient for those compounds. However, such an approach cannot be used for all the compounds studied. There are discrepancies between the ∆A and P values, for example for compounds 3 and 5, which are evidence that not only the partition coefficient determines the efficiency of a compound. As in the oxidation experiments, acyclic compounds influenced the fluidity of erythrocyte membranes to a greater extent than the cyclic compounds. The obtained anisotropy values in some cases match the antioxidative efficiency of the aminophosphonates. This is so in the case of compounds 2, 3, 4, 6, 9 (acyclic compounds) and almost all the cyclic compounds. The biggest discrepancies between oxidation and fluidization effects were found for compounds 8 and 9, i. e., for compounds that should incorporate deeply into erythrocyte membranes due to their long hydrocarbon chains. Compound 1,

CELLULAR & MOLECULAR BIOLOGY LETTERS

89

devoid of developed hydrophobic parts, and so probably incorporating very shallowly into the lipid bilayer, also showed a small influence on membrane fluidity. As mentioned, no big differences were observed for the anisotropy values calculated for cyclic compounds, except for compounds 10 and 11. Their antioxidative efficiency was relatively high, as follows from the fluorescence measurements. This is probably due to the isopropyl substituents at their phosphorus atom, promoting antioxidative protection. Tab.3. The percentage values of anisotropy changes ∆A found for acyclic aminophosphonates and the values of their partition coefficients P.

The ∆A and P values C MM 0.01 0.02 P

1 1.5 1.8 *

2 10.2 15.7 0.21

Compound no. 3 4 5 9.1 11.3 2.3 16.0 16.4 5.8 0.15 0.22 0.19

6 13.1 17.1 0.24

7 2.6 4.7 0.16

8 7.7 13.1 0.14

9 9.5 15.0 0.17

* - No hemolysis was observed up to 5 mM P - mean error did not exceed 20% Tab.4. The percentage values of anisotropy changes ∆A found for cyclic aminophosphonates and the values of their partition coefficients P.

The ∆A and P values C MM 0.01 0.02 P

10 2.6 8.7 0.14

11 8.3 10.2 0.20

Compounds no. 12 13 6.0 10.6 11.3 12.1 0.22 0.25

14 6.8 9.8 0.23

15 4.9 6.4 0.16

16 2.6 2.6 0.08

P - mean error did not exceed 20%

The general conclusion emerging from the results obtained is that at sublytic concentrations, the aminophosphonates studied may act as moderate antioxidative agents, their antioxidative efficiency depending on the localization of their active centres with respect to the membrane they are incorporated by. It seems that the optimal condition ensuring high effectiveness is a combination of an unbranched and not too long hydrocarbon substituent without any additional groups such as a hydroxyl at the nitrogen atom, small, up to isopropyl

90

CELL. BIOL. MOL. LETT.

Vol. 6. No. 1. 2001

substituents at phosphorus atom and developed hydrophobic substituents at the carbon atom. The presence of a ring at this atom is not favorable. Acknowledgements. This work was supported by the Polish Research Committee (KBN), grant no. 6 PO4 G 050 17. REFERENCES 1. Pikl J. US Patent 2.328.358 (1943). 2. Baird D. D., Upchurch R. P. and Selleck G. W. Phosphonomethyl glycine, a new broad-spectrum, postemergence herbicide. Calif. Weed Conf. Proc. 24 (1972) 94-102. 3. Perkow W. Wirsubstanzen der Pflantzenschutz und Schadlingsbekampfungsmittel Part II. (P.Parey, ed.), Verlag, Berlin und Hamburg (1983/1988) p. 24. 4. Lejczak B., Kafarski P. and Gancarz R. Plant growth regulating properties of 1-amino-1-methyl phosphonic acid and its derivatives. Pest. Sci. 22 (1988) 263-275. 5. Gancarz R. and Dudek M. Structure-activity studies of aminophosphonic derivatives of fluorene. Phosphorus, Sulfur and Silicon 114 (1996) 135142. 6. Hagerstrand H. and Isomaa B. Amphiphile-induced antihaemolysis is not causally related to shape changes and vesiculation. Chem. Biol. Ineractions 79 (1991) 335-347. 7. Isomaa B., Hagerstrand H. and Paatero G. and Engblom A.C. Permeability alterations and antihaemolysis induced by amphiphiles in human erythrocytes. Biophys. Biochim. Acta 860 (1986) 510-524. 8. Kleszczynska H., Sarapuk J. and Dziamska A. The physicochemical properties of some new aminophosphonates. Cell. Mol. Biol. Lett. 5 (2000) 415-422. 9. Wieczorek J. S., Gancarz R., Bielecki K., Grzys E. and Sarapuk J. Synthesis and physiological activities of new acyclic aminophosphonates. Phosphorus, Sulfur and Silicon, in press. 10. Wieczorek J. S., Gancarz R., Bielecki K., Grzys E. and Sarapuk J. Synthesis of new cyclic aminophosphonates and their physiological activities. Phosphorus, Sulfur and Silicon, in press. 11. Dodge J. T., Mitchell C. and Hanahan D. J. The preparation and chemical characteristics of hemoglobin-free ghosts of erythrocytes. Arch. Biochem. 100 (1963) 119-130. 12. Kleszczynska H., Oswiecimska M., Witek S. and Przestalski S. Inhibition of lipid peroxidation inthe erythrocyte membranes by quaternary morpholinium salts with antioxidant function. Z. Naturforsch. 53c (1998) 425-430.

CELLULAR & MOLECULAR BIOLOGY LETTERS

91

13. Lakowicz J. R. Fluorescence polarization. In: Principles of Fluorescence Spectroscopy. Plenum Press. New York and London, (1983) pp. 112-151. 14. Campbell L. D., Dwek R. A. Fluorescence. In: Fluorescence in Biological Spectroscopy. The Benjamin Cunnings Publishing Company Inc. Menlo Park and London (1984) pp. 91-120. 15. Lentz B. R. Membrane „Fluidity” from fluorescence anisotropy measurements. In: Spectroscopic Membrane Probes (L. M. Loew ed.), CRC Press Inc. Boca Raton, Florida (1988) vol. 1, pp. 13-41. 16. Kleszczynska H. and Sarapuk J. The role of counterions in the protective action of some antioxidants in the process of red cell oxidation. Biochem. Mol. Biol. Int. 2 (1998) 385-390. 17. Kleszczynska H., Oswiecimska M., Sarapuk J., Witek S. and Przestalski S. Protective effect of quaternary piperidinium salts on lipid oxidation in the erythrocyte membrane. Z. Naturforsch. 54c (1999) 424-428.