Diabetes mellitus is a growing problem across the world. By the year 2010, it was estimated that over 221 million people would be afflicted with the disease. Type 1 diabetes is the result of absolute insulin deficiency and is treated through the addition of exogenous insulin. Type 2 diabetes, non-insulin dependent diabetes (NIDDM), is characterized by a relative insulin deficiency and increased insulin resistance and accounts for 90% of all cases of diabetes.
Insulin resistance is the inability of cells to use insulin effectively, which results in hyperglycemia even in the presence of adequate amounts of insulin. Insulin resistance contributes not only to diabetes but also to a plethora of other metabolic abnormalities, including dyslipidemia, hypertension, and vasculopathy, which are collectively termed the insulin resistance or cardiovascular dysmetabolic syndrome. Rosiglitazone, also known as Avandia, is effective only in the presence of insulin. Its antihyperglycemic effect is the result of lowered insulin resistance in cells. Its development as a drug is described in this paper.
Bioassay used to discover the lead compound
When GlaxoSmithKline started targeting insulin resistance in 1984, virtually nothing was known about the molecular mechanisms of insulin action, let alone what defects contribute to insulin resistance. Hence, in the absence of defined molecular targets, a mouse model of insulin-resistant type II diabetes was used as a “catch-all” screen for insulin-sensitizing molecules.
The antihyperglycemic activity was determined in genetically obese C57 B1/6 ob/ob mice, which are insulin-resistant, hyperinsulinemic, and glucose intolerant. The compound being screened was administered in the diet for eight days (an eight-day repeat dose screen at three dosage levels), and antihyperglycemic efficacy was assessed using an oral glucose tolerance test. Potency of ~1mg/kg and oral activity were criteria for potential lead compounds.
Lead compounds discovered
Clofibrate, a hypolipidaemic drug, was modified by Takeda and his colleagues to form ciglitazone. Ciglitazone was shown to be a very low potency (ED25 = 300 µmol/kg) insulin sensitizer and was chosen as a lead compound for further studies.
In dogs, rats, and men, several metabolic oxidation products of the cyclohexane ring of ciglitazone are formed. One of these metabolites, AD 47431, was found to have more potent antihyperglycemic activity than ciglitazone in genetically obese and diabetic kk (kkAy) mice and was thus adopted as the new lead compound. The enhanced activity of AD 4743 may be related to increased bioavailability due to greater hydrophilicity. Rosiglitazone emerged from an SAR program on AD 4743 in which the lipophilic cyclohexyl group was replaced by aromatic and polar groups. Replacement with a phenlyurea and conformationally restrained derivatives, such as benzoxazoles, gave substantially more potent analogues.
The minimal effective dose of BRL 48482 was shown to be 3 μmol/kg of diet, 300 times the activity of ciglitazone. Altering the lengths of the linking chains led to variations in the spatial separation between the thiazolidinedione and heterocycle. Since the potency was lower for all resulting compounds, it was assumed that BRL 48482 had the optimum spatial separation.
Replacing either oxygen link with sulfur led to decreased potency, as did branching with methyl or phenyl groups. Increasing the bulk of the group at the exocyclic nitrogen or acylating it also led to decreased potency. In the end, three compounds of identical efficacy (ED25 0.3 μmol/kg) were isolated: BRL 48482, BRL 48552, and BRL 49853.
To choose between these compounds, a selectivity screen was developed based on reductions in blood hemoglobin concentration. For BRL 49653 (now known as rosiglitazone), a dose level 100-fold greater than the minimally effective antihyperglycemic dose level had no significant effect on blood hemoglobin concentrations, whereas the others showed hemoglobin concentration reductions at dosages only marginally higher than those required to improve glycemic control. Thus, rosiglitazone was chosen as the preferred antihyperglycemic agent.
The Pharmacophore
Rosiglitazone belongs to a class of oral antidiabetic agents called thiazolidinediones, which seem to be ideally suited for the treatment of type 2 diabetes. All agents of this class have a thiazolidine-2,4-dione structure as shown. The various agents of this class differ in their side chains, which alter their pharmacologic and side-effect profiles.
The peroxisome proliferator-activated receptors (PPARs) form a subfamily of the nuclear receptor superfamily. The three isoforms of PPARs are ligand-dependent transcription factors that regulate target gene expression by binding to specific peroxisome proliferator response elements (PPREs) in enhancer sites of regulated genes. Each receptor binds to its PPRE as a heterodimer with a retinoid X receptor (RXR). Upon binding an agonist, the conformation of a PPAR is altered and stabilized such that a binding cleft is created, and recruitment of transcriptional coactivators occurs. The result is an increase in gene transcription. Through experiments involving dominant-negative mutations in human PPARγ, it was established that the receptor is indeed implicated in the cause of insulin resistance. Rosiglitazone and other thiazolidinediones are specific high-affinity ligands for PPAR.
Binding of Rosiglitazone and PPARγ
Rosiglitazone binds with the PPAR-gamma LBD and SRC-1 in the ternary complex. A ribbon drawing shows the ternary complex of PPAR-gamma LBD, BRL 49653, and the LXXLL helix domain of SRC-1. Residues around K301 and E471 that form the “charged clamp” are red, and the LXXLL SRC-1 helix is green. Rosiglitazone (stick diagram) binds in a deep cavity of the protein and provides a network of polar interactions that include the AF-2 domain. The secondary-structure elements are shown as a ribbon drawing, with amino acids involved in ligand binding labeled.
From a study conducted by Young et al. using radioiodinated ligand, it was determined that rosiglitazone bound to PPARγ effectively only in the S-conformation. The IC50 value of the S-enantiomer was 2.1 nm, compared to 2770 nm for the R-enantiomer. The acidic TZD heterocycle forms hydrogen bonds with His 323 on helix 5 and Tyr-473 on the AF2 helix.
Pharmacological Mechanism of Action:
Rosiglitazone reduces insulin resistance by increasing insulin-dependent glucose disposal in skeletal muscle cells and reducing hepatic glucose output by the liver. In subjects with dominant negative PPARγ mutations, adipocyte differentiation was inhibited, indicating that PPARγ is necessary in the process of adipocyte differentiation. Also, the receptor is present in much greater quantities in adipose tissue than in skeletal or liver tissue. The primary effect of PPARγ is on adipose tissue, with secondary insulin-sensitizing effects on skeletal muscle and liver cells. By stimulating glucose uptake into adipocytes through the glucose transporter GLUT-4, PPARγ causes liver and muscle cells to be more sensitive to existing levels of glucose, i.e., it decreases insulin resistance.
Rosiglitazone monotherapy is effective in patients with type 2 diabetes. In studies conducted, it reduced fasting plasma glucose levels by 3.22 mmol/L in 2 mg doses (bd) and by 4.22 mmol/L in 4 mg doses (bd). β-cell function was estimated to be improved over baseline by up to 60%.
Although effective in monotherapy, the insulin sensitizer is often used in conjunction with sulphonylureas or metformin. Sulphonylureas stimulate insulin secretion from β cells and thus treat the relative or absolute insulin deficiency of Type 2 diabetes rather than insulin resistance. Studies demonstrated that endogenous fasting insulin concentrations with rosiglitazone (2mg bd) + sulphonylurea were 6.4 pmol/L lower than those in patients undergoing treatment with just sulphonylurea. Metformin, a biguanide, was used in a 12-week trial of rosiglitazone combination therapy; fasting glucose levels decreased from. Because the mechanism of rosiglitazone differs from those of sulphonylurea and metformin, the effects of a combination of the two are additive, possibly synergistic. In addition to drug combination therapy, type 2 diabetes is treated through lifestyle modifications such as weight loss and increased pharmacologic agents that decrease the body’s requirement for insulin.
Synergism:
RXR ligands have been shown to be effective in activating PPARγ. This is because PPARγ forms a heterodimer with the retinoid X receptor (RXR) that can be activated by both PPARγ and RXR-specific ligands. Experiments were conducted to test whether LG100268, an RXR ligand, could enhance transcriptional activation by mutant PPARγ. At moderate concentrations, a combination of PPARγ and RXR ligands induced significantly greater transcriptional activation than either ligand alone, indicating the possibility of synergistic effects. Exercise stimulates glucose uptake by muscle cells with normal insulin sensitivity. Rosiglitazone therapy in conjunction with exercise improves this synergistic action.
Metabolism: The Fate of Rosiglitazone as it Journeys Through the Human Body
Rosiglitazone undergoes extensive metabolism, as no unchanged drug was detected in urine studies conducted using 14C-labeled rosiglitazone. The drug was rapidly cleared from plasma in all subjects and was quantifiable only up to 24 hours after dosing. The major routes of metabolism were N-demethylation and hydroxylation, followed by conjugation with sulfate and glucuronic acid. In vitro data show that cytochrome P450 (CYP) isoenzyme 2C8 is the predominant pathway for rosiglitazone metabolism, with CYP2C9 serving as a minor pathway. The metabolites formed are active but have significantly less activity than the parent compound. A proposed scheme for the metabolism of rosiglitazone in humans is shown below.
M10 and M4 together accounted for approximately 35% of the dose excreted over 8 days. The proposed scheme for metabolism in humans is closely similar to that proposed for metabolism in rats and dogs. Phase I metabolism in the rat and dog resulted in ring hydroxylation, N-demethylation, and oxidative removal of the pyridinylamino function, yielding a phenoxyacid derivative, just as in the proposed scheme for metabolism in humans. There were differences in species in the persistence of the circulating metabolites (measured as total radioactivity), but rosiglitazone’s principal metabolites were accurately predicted from preclinical studies. However, the phenoxyacetic acid metabolite M1 was a minor route of elimination in humans, accounting for less than 4% of the dose.
Prodrug
A prodrug of rosiglitazone was not found. This result seems reasonable, as the pharmacokinetics of rosiglitazone are within an optimum range without modification. It is already 99% bioavailable, and none of its metabolites are toxic. A search was conducted with keywords such as “Avandia prodrug,” “rosiglitazone prodrug,” “diabetes prodrug,” and “thiazolidinedione prodrug” using various search engines, including PubMed, All Ovid, Google, Lexis-Nexis, ISI Web of Science, Medline, EMBASE Drugs and Pharmacology, and Academic Search Elite.
Possible Prodrug
Cytochrome P-450 is an electron donor protein for several oxygenase enzymes found on the endoplasmic reticulum of most eukaryotic cells. It can oxidize tertiary amines, forming a carbinolamine that readily decomposes to the secondary amine with loss of formaldehyde. By this mechanism, a possible prodrug of rosiglitazone would involve methylating the nitrogen of the thiazole. The methylation prevents hydrogen bonding, making the molecule more lipophilic. However, since H-bonding is necessary for binding to PPARγ, P-450 must demethylate the prodrug before it can be effective. This might delay the onset of action of rosiglitazone, which would be useful in some circumstances. The mechanism by which cytochrome P-450 would demethylate the possible prodrug is outlined below.
Side Effects
In 26-week clinical trials, the mean weight gain in patients treated with rosiglitazone (8mg daily) monotherapy was 3.5kg. Edema was reported in 4.8% of patients receiving rosiglitazone, compared to 1.3% of patients on placebo. Decreases in hemoglobin and hematocrit of 1.0 g/dL and 3.3%, respectively, were observed in clinical trials of rosiglitazone monotherapy, as well as in combination with other hypoglycemic agents. There was also a slight decrease in white blood cell count, which is probably related to the increased plasma volume. In placebo-controlled trials, 0.2% of rosiglitazone-treated patients had reversible elevations in ALT (>3 times the upper limit of normal), compared with 0.5% of patients on active comparator agents. Headaches, back pain, and a slight cough are also minor side effects of rosiglitazone treatment.
Tolerance to Rosiglitazone
No indications of tolerance to rosiglitazone due to the effects of the drug itself were found. In studies conducted regarding the effects of thiazolidinediones, 75% of the patients exhibited glucose-lowering effects while 25% did not. Analysis of the individual data revealed that those who did not respond to the drug had the lowest levels of insulin secretion at the onset of the study. This indicates that rosiglitazone is not effective in the absence of adequate levels of insulin. Other factors that decrease glucose tolerance, i.e., increase insulin resistance, will cause rosiglitazone to be less effective.
In a study conducted by Gerben et al., the effects of caffeine on whole-body insulin sensitivity were observed. The calculated insulin sensitivity during caffeine administration was 0.39 ± 0.04, compared with 0.46 ± 0.04 μmol/kg in the placebo. This decrease in insulin sensitivity of ~15% is close in magnitude to the increase in insulin sensitivity obtained by rosiglitazone and may thus be involved in seeming tolerance.
Searches were conducted on PubMed, All Ovid, Google, Lexis-Nexis, ISI Web of Science, Medline, EMBASE Drugs and Pharmacology, and Academic Search Elite with keywords rosiglitazone tolerance, decreased effects of rosiglitazone (thiazolidinediones), increasing insulin resistance, and Avandia tolerance.
Reference:
- Smith, Steve. December 6th, 2001. SMR Drug Discovery Award Lecture. Avandia- targeting type 2 diabetes, the epidemic disease of the 21st century. http://www.prous.com/smr01/webcast.asp
- DeFronzo RA, Bonadonna RC, Ferrannini E. Pathogenesis of NIDDM: a balanced overview. Diabetes Care 1992;15:318-368.
- Cantello BCC, Cawthorne MA, Cottam GP, Duff PT, Haigh D, Hindley RM, Lister CA, Smith SA, Thurlby PL. [[.omega.-(Heterocyclylamino)alkoxy]benzyl]-2,4-thiazolidinediones as potent antihyperglycemic agents. J. Med. Chem. 1994; 37(23); 3977-3985.
- Thorp JM, Waring WS. 1962. Nature of hepatomegaly effect produced by ethyl-chlorophenoxy-isobutyrate in the rat. Nature 208:856-858.
- Chakrabarti R, Vikramandithyan RK, Prem Kumar M, Kumar SKB, Mamidi NVS, Misra P, Suresh J, Hiriyan J, Rao CS, Rajagopalan R. PMT13, a pyrimidone analog of thiazolidinedione improves insulin resistance-associated disorders in animal models of type 2 diabetes. Diabetes 2002: 4(5):319.
- Brooks DA, Etgen, GJ, Rito CJ, Shuker AJ, Dominianni SJ, Warshawsky AM, Ardeck R, Paterniti JR, Tyhonas J, Karanewsky DS, Kauffman RF, Broderick CL, Oldham BA, Rafizadeh CM, Winneroski LL, Faul MM, McCarthy JR. Design and Synthesis of 2-Methyl-2-{4-[2-(5-methyl-2-aryloxazol-4-yl)ethoxy]phenoxy}propionic Acids: A New Class of Dual PPAR/ Agonists. J. Med. Chem. 2001; 44 (13): 2061-2064.
- Barroso I, Gurnell M, Crowley VEF, Agostini M, Schwabe JW, Soos MA, Maslen GL, Williams TDM, Lewis H, Schafer AJ, Chatterjee VKK, O’rahilly S. Dominant negative mutations in human PPARγ associated with severe insulin resistance, diabetes mellitus, and hypertension. Nature 1999; 402:880-83.
- Xu EH, Lambert MH, Montana VG, Plunket KD, Moore LB, Collins JL, Oplinger JA, Kliewer SA, Gampe RT, McKee DD, Moore JT, Wilson TM. Structural determinants of ligand binding selectivity between the peroxisome proliferators activated receptors. Proceedings of the National Academy of Sciences 2001; 98(24):13919-13924.
