Synthetic and Plant Derived Thyroid Hormone Analogs

1. Introduction

Nuclear receptors (NRs) are transcription factors that regulate gene expression in response to small signaling molecules. The NR family includes receptors for thyroid hormone (TH) (Yen, 2001), retinoids, vitamin D, steroid hormones, fatty acids, bile acids and cholesterol derivatives, a variety of xenobiotics and other ligands. Additionally, other members are called orphans and could either bind to ligands that have not yet been identified or modulate gene expression in a ligand independent fashion (Webb et al., 2002).

Since NR play important roles in development and disease, they are important candidates for pharmaceuticals. This family of proteins can be modulated by natural and synthetic ligands and are therefore promising targets for drug discovery. The natural ligands may include different kinds of plant molecules or even a combination of compounds present in raw extracts of medicinal plants.

1.1. Compounds that bind NR

Ligands that target NRs include TH, glucocorticoids, estrogens for hormone replacement therapy (HRT), the diabetes drug thiazolidinedione, synthetic retinoids, and many others. Though the list of possible targets is extensive, it is restricted by the fact that NR ligands have both beneficial and deleterious effects. For instance, TH improves overall lipid balance and promotes weight loss by increasing metabolism, but causes tachycardia that can be severe enough to lead to heart failure, muscle wasting and osteoporosis (Felig & Baxter, 1995; Braverman et al. 2000). Likewise, estrogen use in HRT alleviates symptoms of hot flashes and reverses bone loss, but increases the risk of breast and uterine cancers, and stroke (Gustafsson, 1998; McKenna & O´Mally, 2000; McDonnel & Norris, 2002).

1.2. Thyroid hormone receptor (TR) isoforms

Thyroid hormone signals are transduced by two related thyroid receptor subtypes, TR α and TR β (Figure 1.) which are encoded by different genes (Gauthier et al., 1999).

Figure 1.

Chemical structures of thyroid hormones thyroxine (α) and triiodothyronine (β).

Studies of TR isoform-specific knockout mice and patients with resistance to thyroid hormone syndrome suggest that TR α mediates the effects of thyroid hormone on heart rate, whereas analogs that exclusively stimulate TR beta might have desirable effects without causing cardiac distress. Indeed, animal studies using thyroid receptor agonists with modest TR beta selectivity have validated this hypothesis (Taylor et al. 1997; Baxter et al., 2001; Grover et al., 2003). However, structure-based approaches to develop ligands with further improvements in isoform specificity are limited by the fact that the LBDs of TR alfa and TR beta are ~75% identical in amino acid sequence, and that the internal hydrophobic cavities that hold the hormone, called the pocket of the receptor, differ by just one amino acid (Ser-277 in TR alfa versus Asn-331 in TR beta). Therefore, it would be interesting to develop selective TR agonists that increase metabolism and improve lipid balance, but do not cause side effects on the heart. The first compound to show this property was GC-1 (Figure 2), an analog of T3 (Chiellini et al., 2002).

Figure 2.

Chemical structure of the β-specific thyroid hormone receptor agonist GC1.

2. TR antagonists

2.1. First generation of synthetic ligands

TR antagonists would be useful for short-term relief from the symptoms of hyperthyroidism, and might even be used on a long term basis. The first generation of T3 antagonists, which include DIBRT, HY-4, and GC-14, used the “extension hypothesis” as a general guideline in hormone antagonist design (Baxter et al., 2002; Yashihara et al., 2001; Chiellini et al., 2002). This extension in the ligand structure blocks normal receptor function by occupying the pocket region where the hormone normally binds.

Figure 3.

Chemical structure of thyroid hormone receptor antagonist DIBRT.

2.2. Novel series of antagonist compounds

Although the “extension hypothesis” is applicable to the design of nuclear receptor antagonists, the nature of chemical groups that convert agonist ligands to antagonists will likely depend on specific interactions between residues of the receptor and the ligand extension, to help stabilize the antagonist conformation. Following the first designed TR antagonists it was reported the design and synthesis of a novel series of compounds sharing the GC-1 halogen-free thyronine scaffold (second generation). One of them (NH-3) is a T3 antagonist with improved TR binding affinity and potency that allow for further characterization of its observed activity. One mechanism for antagonism appears to be the ability of NH-3 to block TR-coactivator interactions (Ngoc-Ha et al., 2002). NH-3 (Figure 4.) is the first T3 antagonist to exhibit potent antagonism in vivo and therefore may prove to be a generally useful tool for studying the effects of TR inactivation in a variety of animal models. Until now, such studies have been done primarily using TR-knockout mice because a pharmacological tool for inducing TR inactivation has not been available. TR inactivation was limited under previous ligands because they have only a modest affinity and potency for the thyroid hormone receptor (TR), which limits studies of their actions. T3 antagonists such as NH-3 may be useful therapeutic agents in the treatment of hyperthyroidism and other metabolic disorders.

3. Plant ligands of nuclear receptors

3.1. Estrogen analogs

Although modern research on drug discovery involves the design of hormone analogs based on the structure of the receptor, natural ligands can also be found in nature. Estrogen analogs are most common and have been discovered in a variety of plants. Ginsenosides (Figure 5.) for instance, found in Ginko biloba, have demonstrated pharmacological effects on the central nervous, cardiovascular, and endocrine systems. Although no direct interaction of the compound with estrogen receptor seems to be necessary for estrogenic action, the author classified this plant ligand as a novel class of potent phytoestrogen (Chan et al., 2002).

Figure 4.

Chemical structure of the thyroid hormone receptor antagonist NH-3 (Lim et al. 2002), designed by the extension hypothesis from GC-1 compound.

Figure 5.

Chemical structure of a ginsenoside.

Recently, Tephrosia candida (native to the tropical foothills of the Himalayas in India and introduced in South America) was reported to contain estrogenically active chemical constituents, which acted by binding to estrogen receptor ERα. Results were interpreted via virtual docking of isolated compounds to an ERα crystal structure (Hegazy et al., 2011). Also sesame ligands, from Sesamum indicum (flowering plant, native to Africa and widely naturalized in tropical regions around the world), and their metabolites have been evaluated for estrogenic activities (Pianjing et al., 2011). Two of them, enterodiol (Figure 6.) and enterolactone, have been indicated to have estrogenic/antiestrogenic properties on human breast cancer cells.

Figure 6.

Chemical structure of an enterodiol

3.2. Androgen analogs

As compared to estrogen analogs, androgen-like compounds in the flora are less frequently referred in scientific literature. But a few chemicals have shown androgen-like activity. Andrographolide (Figure 7.), an herbal medicine, inhibits interleukin-6 expression and suppresses prostate cancer cell growth (Chun et al., 2010). According to the author, this phytochemical could be developed as a therapeutic agent to treat both androgen-stimulated and castration-resistant prostate cancer. Another compound, Isoangustone A, present in hexane/ethanol extract of Glycyrrhiza uralensis, induces apoptosis in DU145 human prostate cancer (Seon et al., 2010). This species, also known as Chinese liquorice, is a flowering plant native to Asia, which is used in traditional Chinese medicine.

3.2.1. Androgen ligands in the diet

Some studies have specifically demonstrated that consuming one or more portions of broccoli per week can reduce the incidence of prostate cancer, and also induce the progression from localized to aggressive forms of prostate cancer (Trakka, et al. 2008). The reduction in risk may be modulated by glutathione S-transferase mu 1 (GSTM1) genotype, with individuals who possess at least one GSTM1 allele (i.e. approximately 50% of the population) gaining more benefit than those who have a homozygous deletion of GSTM1, according to the author.

Figure 7.

Chemical structure of an andrographolide.

From these studies, we can conclude that diet has a significant influence on the activity of the androgen receptors and possibly other types of nuclear receptor. If so, a wide range of diseases may be avoided by increasing intake of food that contains hormone analogs or other nuclear receptor modulators.

3.3. Plant thyroid hormone analogs

Concerning the thyroid hormone receptor, we find even fewer studies about thyroid hormone (T₃ and T₄) analogs in plants. In a work about patients where thyroid have been removed partially or totally due to thyroid cancer, the plant R. rosea was seen as a viable alternative treatment for the symptoms of short-term hypothyroidism in patients who require hormone withdrawal (Zubeldia et al., 2010). Some compounds of natural origin have also shown to affect the thyroid hormone feedback system by interfering with different components of this homeostatically regulated system: biosynthesis, secretion and metabolism, transport, distribution, and action of thyroid hormones, including the feedback mechanism.

Genistein (Figure 7.) and daidzein, the major components of soy, influence thyroid hormone synthesis by inhibition of the iodide oxidizing enzyme thyroperoxidase. This interferes with thyroid hormone transport proteins and 5′-deiodinase type I activities in peripheral tissues, leading to altered thyroid hormone action at the cellular level. Synthetic flavonoids, such as F21388, which is structurally similar to thyroxine, cross the placenta and also reach the fetal brain of animal models (Hamann et al. 2008).

The cruciferous family was also referred when we consider thyroid modulators in plant. In a study that examined the effects of both soy-foods and specific phytoestrogenic molecules on the development of thyroid cancer in humans it was demonstrated that intake of plants from that family decreases the risk of this kind of cancer (Horn-Ross et al., 2002). The compounds that may be associated with this effect are, according to the author, isoflavones, lignans and 2-hydroxyestrogens. Although anti-carcinogenic response was linked to those molecules, it was not explained how they may affect metabolism in humans or their physiological mechanism of action. Another compound, indole-3-carbinol, the most studied component of cruciferous vegetables, has been demonstrated to have chemopreventive activity in several different animal models of carcinogenesis, including mammary gland, but in another hand, the same compound has also been reported to exhibit adverse promoting effects, including liver and thyroid gland tumorigenesis (Murilo & Mehta, 2001).

Figure 8.

Chemical structure of a Genistein.

It seems to have a cross talk between the thyroid hormone receptor and the estrogen receptor. Our recent results (Cunha Lima et al, not published) have shown that ligands originally referred for thyroid diseases have activated estrogen receptor in transient transfection assays. This could explain why phytoestrogens have caused responses in thyroid cancer and thyroid hormone analogs have also effect in breast cancer, for example.

3.4. Toxicity of thyroid hormone analogs

Considering medications used for thyroid hormone replacement, as sodic levothyroxine, aT₄ analog, the side effects referred (Cunha Lima, 2008) may include headache, chest pain, rapid or irregular heartbeat, shortness of breath, trembling, sweating, diarrhea and weight loss. The most severe responses are those related to the heart, which can lead to serious cardiopathies and are due to the α isoform of the receptor, as cited previously. A single base difference in the pocket of the protein can lead to these harmful responses. This means that we have a two step work on the search for new agonists and antagonists: they should mimesis the response caused by the thyroid hormone (or antagonize it, depending on the disease) and second, they should have a β specificity to avoid tachycardia and more serious heart problems.

In the other hand, some compounds that modulate TR may not be specific for this receptor, as it is very common with estrogen ligands. Since women hormone analogs may interfere with the function of the thyroid, as referred before with some flavonoids, they may have beneficial effects in cases of thyroid cancer. Nevertheless those compounds may influence thyroid actions at the cellular level and could cause side effects harmful for healthy individuals.

3.5. Ethnobotanical search for TR plant ligands

Since thyroid hormone analogs have much fewer discovered natural ligands, and most of those nuclear receptor ligands are found from plant sources, ethnobotanical surveys can be a good strategy to discover hormone analogs in nature. This approach has an increased probability of success in locations with higher biodiversity; because they contain a privileged number of candidate species. Along with botanical diversity, ethnobotanical surveys are likely to succeed where the population has an in depth knowledge of medicinal plants and systematically uses those plants to treat a range of metabolic disorders.

In a recent work (Cunha Lima et al., 2008) we investigated the medicinal flora used for the treatment of metabolic disorders in Salvador, Bahia, in Northeastern Brazil. The city has hot, tropical weather, with average daily highs reaching 17⁰C in the winter and 38⁰C in the summer. Northeastern Brazil is the economically poorest region of the country, 60% of the active population has an income under $100 per month, (Brazilian Institute of Statistics and Geography, 2003) and many residents depend on medicinal plants to treat multiple ailments and diseases.

The referred study analyzed the knowledge of the urban population of Bahia city on the use of potentially therapeutic plants for the treatment of Diabetes mellitus type 2 (DM2), thyroid diseases, obesity and cardiopathies. Questionnaires were applied to traditional healers as well as to patients of the thyroid disease and diabetes ambulatory in the Hospital from Federal University of Bahia (UFBA). Thirty-one cited species were collected, taxonomically classified, and stored in Alexandre Leal Costa Herbarium (ALCB) from UFBA. Leaves were most commonly used in preparations (87%), followed by the whole plant (10%), and fruits and seeds (3%). The majority of the preparation (88%) required decoction (boiling the plant tea for at least 5 minutes); the rest includes infusion (liquid preparation without boiling) and ingestion of the fresh plant. Among the plant parts used the leaves were more frequent (87%), followed by the whole plant (10%), seeds and fruits (3%). The families Asteraceae (17%), Lamiaceae (15%) and Myrtaceae (12%) were the most cited among plants referred.

This survey identified botanical families frequently cited in other surveys of medicinal plant use in Brazil. In two studies conducted in the state of Rio de Janeiro, one in Rio city (Azevedo & Silva) and one in the reservation of Mangaratiba (Medeiros et al., 2005), the Asteraceae and Lamiaceae family of plants were the most frequently cited, the same happening in Conceição Açú-MT (Pasa et al., 2005). Species from the Asteraceae family were also the most frequently noted for medicinal use in a survey done in Ingaí-MG (Botrel et al., 2006) and by a “quilombola”(community of people descended from former Brazilian slaves), among the plants with possible action in the central nervous system (Rodrigues & Carlini, 2004). These data suggest that the Asteraceae and Lamiaceae family have excellent pharmacological potential on different kinds of diseases and they are currently being investigate in many clinical studies.

In the survey performed by our group in Salvador (Cunha Lima, 2008), the plant most used for the treatment of DM2 belongs to the genus Bauhinia (pata-de-vaca). The most commonly cited species in this work, B. forficate, has the flavonoid Kaempferitrina, Kaempferol-3-O-α-Diraminoside and the steroid Sitosterol as the hypoglycemic active principle (da Silva & Cechinel Filho, 2002). Terminalia catappa was the second most cited species for the treatment of DM2. Ahmed et al. (2005) demonstrated that the extract from this plant also has hypoglycemic activity and improves general clinical conditions.

Among the plants used for obesity control, with probable effect on the metabolism, Borreria verticillata (carqueija or vassorinha-de-botão) was the species with the highest number of references in our survey. This plant, also used for the treatment of diabetes type 2, is found across Brazil. Phytochemical studies have demonstrated the presence of alcaloids and iridoids (Vieira et al., 1999) associated with their antipyretic and analgesic properties, although no active principle linked to obesity was confirmed. The leaves of Bauhinia forficata, Costus spiralis and Theobroma cacao, were used as teas in combination with the commercial medical prescriptions sibutramine (an oral anorexiant used for weight control) and niphedipine (a dihydropyridine calcium channel blocker used for high blood pressure) indicated by the physicians from the Diabetes Ambulatory of the Federal University of Bahia Hospital (HUPES). The teas of Tragia volubilis leaves and the seeds of Ocimum gratissimum were also used in combination with niphedipine and aspirin.

The problems related to the thyroid attended at the Hospital of Federal University of Bahia include throat itch, tachycardia, arm pain, chokings, dizziness and fainting. The most extreme side effects symptoms are associated with the T4 hormone replacement for patients whose thyroid was partially or completely removed. The doses used vary from 50 to 200mcg/day of sodic levotiroxine. In addition to the plants cited for treatment of thyroid problems, watercress (Nasturtium officinale R.Br.) and spring-green (Brassica oleraceae L.) were eaten as iodine source. Although the majority of the patients did not tell their doctors they were using those teas, there are no reported adverse side effects due to the combination of the plant products and the medications indicated, nor any reference in the literature about harmful effect of such interaction.

Ethnobotanical surveys are good source of information for drug candidates and offer a less expensive way of finding hormone analogs than the design of synthetic compounds. The cited information represents an important source of regional knowledge on plants with pharmacological potential and presents 31 candidates (Table 1) that might contain triiodothyronine (T3) and thyroxin (T4) analogs, including agonists, antagonists and other compounds able to modulate thyroid receptor that may act against metabolic disorders.

Brazil has more than 55.000 species of cataloged plants (Simões & Schenkel, 2002), a significant portion of which has some phytotherapic activity known by the local population. However, the number of patents on plant-based pharmaceuticals is very small. In particular, the capital of Bahia has numerous plants used by inhabitants to treat diseases and this use is part of the local culture, based in the Candomblé (religion of African origin which uses many plants in rituals and treatments). Traditionally, information about medicinal plants is shared orally. Therefore, it is necessary to scientifically systematize and analyze this phytotherapic knowledge so that those species can be identified and their pharmacological properties tested.

Table 2 lists the species referred in this survey that had their active principles identified and/or properties confirmed, and the bibliographic references where the data was obtained. These works include results from clinical and experimental studies aiming the confirmation of therapeutic properties.

Vernacular Name	ALCB	Family	Species
aroeira	76103	Anacardiaceae	Schinus terebinthifolius Raddi
graviola	76101	Annonaceae	Annona muricata L.
jaca-de-pobre	76154	Annonaceae	Annona montana Macfad
carrapixo-de-agulha	76135	Asteraceae	Bidens bipinnata L.
carrapixo-preto	76111	Asteraceae	Bidens pilosa L.
chapéu-de-couro	76138	Asteraceae	Zinnia elegans Jacq.
urucum	76100	Bixaceae	Bixa orellana L.
cactus	78152	Cactaceae	Cereus sp. L.
pata-de-vaca	76159	Caesalpiniaceae	Bauhinia forficata Link
amendoeira	76096	Combretaceae	Terminalia catappa L.
cana-de-macaco	76122	Costaceae	Costus spiralis (Jacq,) Roscoe
mamona	76141	Euphorbiaceae	Ricinus communis (L.) Müll. Arg.
urtiga	76108	Euphorbiaceae	Tragia volubilis (L.) Müll. Arg.
alecrim ou alecrim-do-reino	76128	Lamiaceae	Rosmarinus officinalis L.
hortelã-grosso	76110	Lamiaceae	Plectranthus amboinicus (Lour.) Spreng
quiôiô	76112	Lamiaceae	Ocimum gratissimum L.
canela	76099	Lauraceae	Cinnamomum zeylanicum Breyn
erva-de-passarinho	76107	Loranthaceae	Struthanthus flexicaulis Mart.
Murici	78150	Malpighiaceae	Byrsonima sericea DC.
barbatimão	76158	Leguminosae	Abarema cochliocarpum (Gomez) Barnbey
jamelão	76156	Myrtaceae	Syzygium cumini (L.) Skeels
pitangueira	76163	Myrtaceae	Eugenia uniflora L.
capim- cidreira ou capim-santo	75150	Poaceae	Cymbopogon citratus Stapf.
roma	76162	Punicaceae	Punica granatum L.
carqueija ou vassourinha-de-botão	76132	Rubiaceae	Borreria verticillata (l.) G.Mey
laranjeira	76097	Rutaceae	Citrus aurantium L.
vassourinha	76114	Scrophulariacea	Scoparia dulcis L.
cacau	78148	Sterculiaceae	Theobroma cacao L.
erva cidreira	76105	Verbenaceae	Lippia alba N.E.Brown
melissa	76120	Verbenaceae	Lippia alba L..
levante	76123	Zingiberaceae	Alpinia nutans Roscoe

Table 1.

Medicinal plants candidates for thyroid hormone analogs according to ethnobotanical research in Salvador-Bahia, Brazil (Cunha Lima, 2008).

Species	Properties associated to the referred use	Reference
Bauhinia forficata	The flavonoids Kaempferitrin and Kaempferol-3-O-α-Diraminoside and the steroid Sitosterol found in the extract own hypoglycemic properties.	da Silva & Cechinel Filho (2002)
Terminalia catappa	Leaf extract prepared in different ways produced antidiabetic response with 1/5 of the lethal dose revealed by the lipid, creatine and urea profile as also serum alkaline phosphatase. The same dose caused anti-diabetic effects with fruit extracts.	Ahmed et al (2005); Nagappa et al (2003)
Rosmarinus officinalis	The anti-oxidants impair the mechanism of oxidation that occurs in cancer, heart disease , atherosclerosis and aging.	Ibanez et al (2000)
Cymbopogon citratus	Intense anti-oxidant activity due to the phenolic composition. The essential oil extracted from the leaf causes depression of the CNS in rats.	Prakash et al (2007); Negrelle (2007)
Bidens pilosa	Deposits of opaline silica in the leaves and extracts of the whole plant obtained with n- hexane demonstrated significant anti-cancer activity.	Parry (1986); Sundararajan et al (2006)
Lippia alba	Flavonoids found in this plant are active against different kinds of cancer including thyroid cancers.	Ren et al (2003)
Annona muricata	Graviola, a Brazilian fruit from the plant Annona muricata demonstrated anti-diabetic effect greater than the medication Clorpropamide, oral hypoglycemic from the sulphonilurea class.	Carvalho (2005)
Annona montana	The plant has kinase protein inhibitors that act creating obesity resistance and increasing insulin production.	Bialy et al (2005)
Syzygium cumini	The species presents anti-diabetic action in clinical and animal studies. Stem extracts stimulate the development of cells positive for insulin in the pancreatic epithelial duct.	Mentreddy (2007); Teixeira et al (2004); Schossler et al (2004)
Citrus aurantium	The combination of C. aurantium extract, caffeine and Saint John´s Herb (Hypericum perforatum) is safe and effective for weight lost and improvement of lipid levels in obese adults.	Colker et al (1999)
Alpinia nutans	The hidroalcoolic extract induces a dose-dependent decrease in artery pressure in rats and dogs.	Mendonça et al (1991)
Lippia alba	The aqueous extract of leaves from this plant, associated to the ones from Melissa officinalis and Cymbopogon citratus caused significant reduction in cardiac rhythm in rats, without changing the contractile strength.	Gazola et al (2004)
Bauhinia forficata	The rats treated with decoction of the plant leaves demonstrated significant reduction in serum and urine glucose. The results obtained with the purified extracts confirmed the therapeutic use for treatment of diabetes in clinical studies.	Pepato et al (2002); da Silva & Cechinel Filho (2002)
Eugenia uniflora	The empiric use of this plant is due to the hypotensive effect, mediated by vessel dilatation and weak diuretic effect that may be related to increased renal blood flow.	Consolini et al (1999)
Punica granatum	Flavonoid rich fractions obtained from fruit extracts demonstrated antiperoxidative effect. Malondialdehyde, hydroperoxide, and conjugated dienes were significantly decreased in the liver, while enzymatic activity of catalase, superoxide dismutase and glutathione reductase have shown significant increase.	Sudheesh (2005)
Scoparia dulcis	Plant extracts were effective on decreasing hyperglycemia and the susceptibility to free oxygen radicals in rats.	Latha & Pari (2004)

Table 2.

Plant species referred in the survey that have their therapeutic properties confirmed or active principles isolated according to scientific publications.

4. Conclusion

Studying medicinal plants can be a less expensive way of finding treatments for hundreds of diseases. This can be an important factor in areas where a great part of the population lacks financial conditions of buying allopathic medication and, in the other hand, have a big incidence of metabolic disorders.

The search for hormone analogs in medicinal plants is extremely promising. Over 100 existing nuclear receptors have been identified, not counting the orphan NRs that lack known ligands. Since those transcription factors modulate almost all genetic activity and human physiology, they are important targets for drug discovery. Besides the ligands, usually hormones, other molecules can also modulate nuclear receptors, including cofactors (co-activators and co-repressors), responsive elements, and other ligands (not exclusively the hormone that naturally binds this receptor). According to that, the molecules found in plants do not have to be only analogs of hormones, but also compounds similar to all other complementary modulators of NRs.

Countries with higher biodiversity are good targets for discovery of plant molecules that can control the activity of thyroid receptor. Unfortunately there is not enough scientific knowledge about their medicinal plants or about patent procedures that would guarantee intellectual property of discoveries made by local scientists. In addition to that, the forests are being devastated very fast before important plant compounds can be found. Therefore, additional research needs to be done to identify new ligands and other molecules in the flora that can modulate TR and may be used in the treatment of diseases related to thyroid malfunction.