**5. Modified estrogens with antioxidant action as potential neuroprotective agents**

Estradiol and its analogues have been known to have pro- and antioxidant features. These properties of estrogens are still subject of debate and depend on many factors, including animals or tissues, administration routes, concentrations, peroxidative model and so on. Antioxidant action of estrogens has been widely studied in vivo and in vitro. Aside from its effects on LDL-oxidation (**Badeau et al., 2005**), it has reported that estrogens decreased lipid peroxidation in brain homogenates and neuronal cultures (**Thibodeau et al., 2002),** reduced the superoxide anion production in different cells (**Florian et al., 2004),**

In vivo studies allow usage of different experimental models of neurological disease (**Azcoitia et al., 2002**). Among the steroid family, only estrogens have the capability to prevent neuronal cell death caused by oxidative stress. Estradiol has been reported to reduce mortality and cerebral damage in the models of brain ischemia including middle cerebral artery occlusion and common artery occlusion (**Perez et al., 2005**). The neuroprotective effect of estrogens have also been shown in animal model of Parkinson's disease induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine and in experimental models with various toxicities including serum deprivation, amyloid peptide, glutamateinduced excitotoxicity, kainic acid and so on (see **reviews Green Simpkins, 2000; Amantea et al., 2005**). Estrogens may protect against injury via receptor-dependent and receptor-independent mechanisms and it has been suggested that antioxidant capacity is an important component of the complex neuroprotective effect (**Green Simpkins, 2000; Prokai Simpkins, 2007).**

Antioxidant activity of estrogens as well as other known antioxidants in vivo is determined by a lot of factors: concentrations, distribution, localization, fate of antioxidant-derived radical, interaction with other antioxidants, metabolism (**Niki Noguchi, 2000**). Natural fluctuation of ovarian hormones during estrous cycle may influence the effect of exogenous hormones; therefore ovariectomized animals are often used. Another problem with this approach is that estrogens can be transformed in tissues into metabolites, for example to

inhibitory action the presence of side chain (3-4 carbon atoms) at position 16 is optimal. Such substituents contribute to the appearance of antagonist activity to ER. It is desirable to have the tertiary amide group with substituents on the amide nitrogen. Authors assumed that primary bromide will inactivate 17-hydroxysteroid dehydrogenase. Thus, steroid was expected to be inhibitor of 17-hydroxysteroid dehydrogenase and agonist-antagonist to nuclear estrogen receptor. These suppositions have been confirmed: model steroid caused 25% stimulation of cellular growth at concentration 0.1 μM, at the same concentration steroid inhibited by 45% the 0.1 nM estradiol-stimulated growth of ZR-75-1 cells. It means, that compound **84** is partial agonist of ER and inhibitor of 17-hydroxysteroid

Sulphamates **85** and **86** are inhibitors of estrone sulfatase and 17-hydroxysteroid

The selection of ways for the treatment of oncological diseases mainly depends from individual peculiarities of patients, as result the methods of definition of aromatase, estrone sulphatase, 17-hydroxysteroid dehydrogenase or mRNA (which realizes their synthesis),

**5. Modified estrogens with antioxidant action as potential neuroprotective** 

Estradiol and its analogues have been known to have pro- and antioxidant features. These properties of estrogens are still subject of debate and depend on many factors, including animals or tissues, administration routes, concentrations, peroxidative model and so on. Antioxidant action of estrogens has been widely studied in vivo and in vitro. Aside from its effects on LDL-oxidation (**Badeau et al., 2005**), it has reported that estrogens decreased lipid peroxidation in brain homogenates and neuronal cultures (**Thibodeau et al., 2002),** reduced

In vivo studies allow usage of different experimental models of neurological disease (**Azcoitia et al., 2002**). Among the steroid family, only estrogens have the capability to prevent neuronal cell death caused by oxidative stress. Estradiol has been reported to reduce mortality and cerebral damage in the models of brain ischemia including middle cerebral artery occlusion and common artery occlusion (**Perez et al., 2005**). The neuroprotective effect of estrogens have also been shown in animal model of Parkinson's disease induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine and in experimental models with various toxicities including serum deprivation, amyloid peptide, glutamateinduced excitotoxicity, kainic acid and so on (see **reviews Green Simpkins, 2000; Amantea et al., 2005**). Estrogens may protect against injury via receptor-dependent and receptor-independent mechanisms and it has been suggested that antioxidant capacity is an important component of the complex neuroprotective effect (**Green Simpkins, 2000;** 

Antioxidant activity of estrogens as well as other known antioxidants in vivo is determined by a lot of factors: concentrations, distribution, localization, fate of antioxidant-derived radical, interaction with other antioxidants, metabolism (**Niki Noguchi, 2000**). Natural fluctuation of ovarian hormones during estrous cycle may influence the effect of exogenous hormones; therefore ovariectomized animals are often used. Another problem with this approach is that estrogens can be transformed in tissues into metabolites, for example to

dehydrogenase with moderate activity.

**agents** 

**Prokai Simpkins, 2007).**

dehydrogenase (**Potter & Reed, 2002; Messinger et al., 2006).**

the superoxide anion production in different cells (**Florian et al., 2004),**

content have a crucial importance (**Irahara et al., 2006**).

catecholestrogens, that have another antioxidant property and can produce prooxidants (**Picazo et al., 2003**). Most animal studied have utilized rodents (especially rats), that have a very high rate of estrogen degradation. Further, in case of brain the synthetic estrogens that are candidates for neuroprotective antioxidants in vivo should be able to cross the bloodbrain barrier after their systemic administration. In vivo estrogens probably do not exert their own antioxidant action but interact synergistically with other antioxidants or reductants (**Hwang et al., 2000**). They can also affect the redox state of the cell through alteration of glutathione concentration and enhance the production of high energy compounds, stimulate antioxidant enzymes, such as SOD, catalase or glutathione transferase (**Perez et al., 2005**; **Akcay et al., 2006; Siow et al., 2007; Kumtepe et al., 2009**).

Using in vivo methodology we can not understand molecular mechanisms of certain hormone antioxidant action, but it allows studying the manifestation of complex action of hormone treatment.

More than 12 different types of neuronal cells against the 14 different toxicities were used to investigate neuroprotective effect of estrogens, their analogues and derivatives (**Green Simpkins, 2000; Wise et al., 2001**). The concentrations of estrogens that have produced protective action in these models vary from physiological (0.1 nM) to pharmacological (50 M) and it is suggested that different neuronal types may have different sensitivities to estrogen-mediated protection. For example physiological concentrations of 17-estradiol were neuroprotective in cultures that contain multiple cell types and maintain intact cellular architecture (**Dhandapani Brann, 2007)**. Several studies provide a positive correlation between in vitro neuroprotective potency and antioxidant activity of estrogens: they inhibited lipid peroxidation induced by glutamate, iodoacetic acid, amyloid peptide (**Perez et al., 2005; Prokai-Tatrai et al., 2008**), reduced iron-induced lipid peroxidation (**Vedder, et al., 1999**), prevented intracellular peroxides accumulation induced by different toxicities (**Behl et al., 1997**). The neuroprotective antioxidant activity dependents on the presence of OH-group in the C-3 position on the A ring of the molecule. The formation of ether derivatives at C-2 position reduces effect because it abolishes the ability to donate a hydrogen atom (**Prokai et al., 2001**; **Perez et al., 2005**).

However, adjoin electron-donating methoxy groups to the phenolic ring may enhance antioxidant potency by weakening the phenolic O-H bond and provide stability of the formed phenoxyl radical (**Prokai Simpkins, 2007**). Nevertheless generally higher concentrations of the hormones were required for antioxidant action than were needed for neuroprotection. This observation indicates that antioxidant effect is not a significant mechanism involved in the neuroprotective activity of estrogens in vivo (**Wise et al., 2001; Manthey Behl, 2006**).

The simplest level of in vitro study of radical scavenging activity is investigation using cell-

free models. Oxidation is induced by different systems of free radical generation and different types of prooxidants in order to gain a more precise view of the mechanism of inhibition. This methodology allows to investigate basic chemical properties (antioxidant or prooxidant) of natural estrogen molecules or synthetic analogues and compare compounds in each other.

There are multiple reactive oxygen and nitrogen species (ROS and NOS) and free radicals: superoxide radical (O2•-), hydroxyl radical (HO•), hydrogen peroxide (H2O2), nitric oxide (NO•), peroxynitrite (ONOO-), hypochlorous acid (HOCl), peroxyl radical (ROO•), lipoperoxy radical (LOO•). The reactivity toward various ROS or NOS can be measured by

Approaches for Searching

problem.

examples.

**OH**

**Schwarz, 1997**).

of Modified Steroid Estrogen Analogues with Improved Biological Properties 193

As we have noted earlier, presence of hydroxyl groups at C-3 and at C17 is important for effective binding of estrogens with estrogen receptors. As the first group is necessary for antioxidant activity of estrogen, synthesis of analogue with substituents at C-17 and/or aromatic ring, which would prevent receptor binding, would be an easy solution of the

Thus, 17-methylensteroid **91** in the experiments on 23-day-old Sprague-Dowley rats exhibits only 1/70 of the uterotropic activity of estradiol. Moreover, this compound blocks the action of 7,12-dimethylbenzanthracene (**Jungblat et al., 1990**). Although in this work the investigation of antioxidant activity of compounds is not mentioned, it must be present in analogy with 17-difluoromethylene derivative **92** (**Bolhman & Rubanii, 1996**). In spite of that steroid has positive action on cardiovascular system; such compounds may have only restricted application because this steroid possesses contraceptive activity. We have to note that biological properties of 8-analogue **93** are close to properties of steroid **92** (**Bolhman &** 

The synthesis of steroid ethers in position 17 also could have lead to compounds having antioxidant activity coupled with lowered uterotropic action, as it was shown with compounds such as **96** (**Prokai et al., 2001; Prokai & Simpkins, 2007)**. Thus, the presence of large substituents in ring D does not result in lowered antioxidant properties, hence synthesis of new compounds which would possess double bond at position 8(9) and large substituents at D-ring can be considered as the next step in the creation of new drugs. It is desirable that these substituents had free phenol hydroxyl group, which would automatically increase its antioxidant properties. In fact, steroids **94** and **95** do have

The introduction of large substituent at position 2 must lead to decrease of estrogenic properties of modifies analogue at the presence of antioxidant activity, that was demonstrated with compounds **97** (**Miller et al., 1996**) and **98** (**Simpkins et al., 2004**) as

**OH**

**OH**

**OH**

**H**

**H**

**H H H**

The high antioxidant activity of steroid **99** has been shown on different models (**Klinger et al., 2003**). Unfortunately this compound strongly inhibits monooxidaze activity of

Hydroxyl group at С-17 in α-region of molecule does not prevent the appearance of antioxidant properties (compounds **100** and **101)**, which opens additional possibilities for the obtaining compounds with the selective action (**W. Römer, M. Oettel, P. Droescher, S.** 

Antioxidant properties of 6-oxa steroid estrogen analogues have undergone little investigation. Oxygen atom does obviously possess a big influence on electron density distribution in A ring, which can have an impact on antioxidant properties of such compounds. Difference in antioxidant action mechanisms between such compounds and

**Rubanii, 1996**). Analogues of type **93** decrease lipoprotein level.

**OH**

**OH**

 **99 100 101** 

those discussed earlier can be assumed (**Prokai-Tatrai et al., 2008**).

**H**

**H**

cytochrome Р450 that is undesirable.

desirable properties (**Römer et al, 1997**).

the inhibition methodology in which free radical species are generated in a special model system and the antioxidant effect is measured by inhibition of the reference reaction (**Sanchez-Moreno C., 2002**). For the superoxide anion radical generation a few model systems can be used: for example, hypoxanthine-xanthine oxidase system, autooxidation of riboflavine or non-enzymatic reaction of phenazine methosulphat in the presence of NADH and O2. Then O2•- reduces nitro-blu tetrazolium into formazan at 250C and pH=7.4.

**93 94 95** 

It is important to note that antioxidant action of estrogens is not necessarily depends on their receptors (**Xia et al., 2002**; **Perez et al., 2005**). This allows to search for modified analogues with lowered or depleted hormonal activity. It appeared that, for example, *ent*estradiol **87** has antioxidant properties **(Simpkins, 2004)**. The introduction of conjugated with aromatic ring double bond into steroid molecule leads to the increased antioxidant activity (steroids **88** and **89)** (**Römer, W.; Oettel, M.; Droescher, P.; Schwarz, S., 1997**) and compound **90** (**Römer et al., 1997**) in comparison with estrogen without double bonds. Steroids of type **88** have significant hypothalamic action (**Berliner, 1996**), that allowed them to be recommended for the treatment of autoimmune diseases.

the inhibition methodology in which free radical species are generated in a special model system and the antioxidant effect is measured by inhibition of the reference reaction (**Sanchez-Moreno C., 2002**). For the superoxide anion radical generation a few model systems can be used: for example, hypoxanthine-xanthine oxidase system, autooxidation of riboflavine or non-enzymatic reaction of phenazine methosulphat in the presence of NADH

**OH**

**OH**

**CF2**

**OH**

**OH**

**H H H**

**H**

**H H H**

**S**

**OH**

**H H H**

**OH**

**OH**

**OH**

**H**

**CH2**

**H H H**

**OH**

**<sup>H</sup> OH**

**H H H**

It is important to note that antioxidant action of estrogens is not necessarily depends on their receptors (**Xia et al., 2002**; **Perez et al., 2005**). This allows to search for modified analogues with lowered or depleted hormonal activity. It appeared that, for example, *ent*estradiol **87** has antioxidant properties **(Simpkins, 2004)**. The introduction of conjugated with aromatic ring double bond into steroid molecule leads to the increased antioxidant activity (steroids **88** and **89)** (**Römer, W.; Oettel, M.; Droescher, P.; Schwarz, S., 1997**) and compound **90** (**Römer et al., 1997**) in comparison with estrogen without double bonds. Steroids of type **88** have significant hypothalamic action (**Berliner, 1996**), that allowed them

**OH**

**OH**

**H**

and O2. Then O2•- reduces nitro-blu tetrazolium into formazan at 250C and pH=7.4.

**OH**

**OH**

**87 88 89** 

**OH**

**90 91 92** 

**93 94 95** 

**OH**

**96 97 98** 

**(CH3)3C**

to be recommended for the treatment of autoimmune diseases.

**H H H**

**OH**

**H**

**OH**

**O(CH2)3CH3**

**H**

**CF2**

**H H H**

> **H H H**

**OH**

**OH**

**OH**

**OH**

As we have noted earlier, presence of hydroxyl groups at C-3 and at C17 is important for effective binding of estrogens with estrogen receptors. As the first group is necessary for antioxidant activity of estrogen, synthesis of analogue with substituents at C-17 and/or aromatic ring, which would prevent receptor binding, would be an easy solution of the problem.

Thus, 17-methylensteroid **91** in the experiments on 23-day-old Sprague-Dowley rats exhibits only 1/70 of the uterotropic activity of estradiol. Moreover, this compound blocks the action of 7,12-dimethylbenzanthracene (**Jungblat et al., 1990**). Although in this work the investigation of antioxidant activity of compounds is not mentioned, it must be present in analogy with 17-difluoromethylene derivative **92** (**Bolhman & Rubanii, 1996**). In spite of that steroid has positive action on cardiovascular system; such compounds may have only restricted application because this steroid possesses contraceptive activity. We have to note that biological properties of 8-analogue **93** are close to properties of steroid **92** (**Bolhman & Rubanii, 1996**). Analogues of type **93** decrease lipoprotein level.

The synthesis of steroid ethers in position 17 also could have lead to compounds having antioxidant activity coupled with lowered uterotropic action, as it was shown with compounds such as **96** (**Prokai et al., 2001; Prokai & Simpkins, 2007)**. Thus, the presence of large substituents in ring D does not result in lowered antioxidant properties, hence synthesis of new compounds which would possess double bond at position 8(9) and large substituents at D-ring can be considered as the next step in the creation of new drugs. It is desirable that these substituents had free phenol hydroxyl group, which would automatically increase its antioxidant properties. In fact, steroids **94** and **95** do have desirable properties (**Römer et al, 1997**).

The introduction of large substituent at position 2 must lead to decrease of estrogenic properties of modifies analogue at the presence of antioxidant activity, that was demonstrated with compounds **97** (**Miller et al., 1996**) and **98** (**Simpkins et al., 2004**) as examples.

The high antioxidant activity of steroid **99** has been shown on different models (**Klinger et al., 2003**). Unfortunately this compound strongly inhibits monooxidaze activity of cytochrome Р450 that is undesirable.

Hydroxyl group at С-17 in α-region of molecule does not prevent the appearance of antioxidant properties (compounds **100** and **101)**, which opens additional possibilities for the obtaining compounds with the selective action (**W. Römer, M. Oettel, P. Droescher, S. Schwarz, 1997**).

Antioxidant properties of 6-oxa steroid estrogen analogues have undergone little investigation. Oxygen atom does obviously possess a big influence on electron density distribution in A ring, which can have an impact on antioxidant properties of such compounds. Difference in antioxidant action mechanisms between such compounds and those discussed earlier can be assumed (**Prokai-Tatrai et al., 2008**).

Approaches for Searching

call this process osteopeny.

of Modified Steroid Estrogen Analogues with Improved Biological Properties 195

changes in bone tissue, which mimic those occurring in aged women. At first phase of fast bone tissue loss triggered by accelerated bone remodeling occurs. Later on this process slows down. Trabecular tissue becomes depleted more so then cortical tissue. It should be noted that bone breaking is rarely observed in ovariectomized rats, thus is probably better to

There are known several models for testing steroids on animals. When investigating the action of compounds in mature rats at the age of 3 months intensive growth of bones in longitudinal direction should be taken into account. 75 Days old (**Turner et al., 1994**) rats are more suitable for the investigation; rats model (6-12 months), which has very few skeletal changes, are much harder to work with. To test bone state several parameters are used, such as «bone mineral density» (g/cm3) and «bone mineral content» (g/sm. length, g/sm2). Last parameter has linear correlation with bone mass (r=0.999) (**Sato et al., 1995**) and calcium contain in ash after bone burning (r=0.90) **(Gaumet M., 1996)**. By measuring dry weight of the bone (after drying at high temperature until the constant mass is obtained), ash weight after burning **(Bauss et al., 1996)**, and also by investigating the ratio of this two parameters (**Broulik & Schreiber, 1994**) changes in mineral bone content can be measured. Results of dual-energy X-rays absorptiometry BMD correlate with fracture incidence and useful in evaluation progression of osteoporosis (**Sharp et al., 2000**). Yet, the final conclusion about steroid activity can only be given after the investigation of their influence on mechanical durability of the bone (**Turner et al., 1994; Sato** 

**et al., 1999**). All the works in this part were investigating naturally occurring steroids.

**O**

**OAc**

**H**

**O**

**O**

**H**

**106** 

**H**

**H**

**H**

**H**

positions 2 and 4 cancels the osteoprotective effect.

**O**

**H**

**104 105** 

**H**

**O**

**H**

As we have already noted the significant structural similarity between natural steroids and their 8-analogues, the osteoprotective and uterotropic action of these compounds were investigated. In experiments with ovariectomized Wistar rats parallel change in those effects was observed. It was suggested that osteoprotective activity is dependent on nuclear receptor of estrogens (**Morozkina et al., 2007**). Hence analogues were chosen as model compounds, despite the fact that presence of methoxyl group at С-3 was expected to trigger the lowering of activity. However in this case carcinogenicity was also expected to be lowered. In the experiments with Sprague-Dowley rats compounds **104** and **105** were shown to possess the same osteoprotective activity as ethynylestradiol, although they were administered in higher doses **(Morozkina et al., 2008**). The presence of substituents in

We carried out the search for modified steroid estrogen analogues in the series of steroids with unnatural rings junction having antioxidant properties and lowered/depressed uterotropic activity. Such properties belong to compounds **102** (**Pison et al., 2009**) and **103**. Some results of investigations of antioxidant properties of steroid **102** are presented in the Table 2.


P – Student' coefficient.

Table 2. Results of investigation of action of steroid **102** on lipid peroxidation in brain of rats

Steroid was given *per os* in olive oil in dose 5 mg per 100 g of weight of rats, for the day before the slaughter. Solutions of the steroids contain 5 mg in 0.3 ml. Control group of animals were treated by olive oil in equal volume.

Solely further investigations of such compounds may show the perspectives of this class of steroids. Depressed hormonal activity of investigated steroids may be a negative factor, inasmuch as during their using the capacity for the induction of the formation of enzymes with antioxidant properties will be decreased.

The presence of antioxidant activity leads to expect neuroprotective action of such modified estrogens, in spite of the fact that exact mechanisms of these properties have not yet been established. A huge amount of patents in this area is a proof of it. We shall cite some of those works (**Simpkins et al, 1994; Covey, 2002; Wuelfert et al, 2002; Peri et al., 2005; Pei, 2005).** Unfortunately clinical trials did not confirm neuroprotective action. Most probably it is connected with multifunctional activity of estrogens, and as a result their positive properties are "compensated" by negative action, but how and which ones are unknown. Further progress in this area must be connected with success in detailed investigation of mechanism of neuroprotective properties and synthesis of analogues with restricted action.
