Antitumor and Quantitative Structure Activity Relationship Study for Dihydropyridones Derived from Curcumin

Problem statement: Pyridones are known to have variety of biological activities like antitumor, antibacterial, antiinflamatory and antim alarial activities. This study presented antitumor evaluation of dihydropyridones derived from curcumi n, as well as curcumin for comparison. Approach: The compounds evaluated for a preliminary estimati on of the in vitro tumor inhibiting activity against 11 of tumor cell lines by using Mi croculture Tetrazolium assay (MTT) method. The method is based on the metabolic reduction of 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide. The cell lines of tumor subpanels were inc ubated within five concentrations (0.01-100 μg mL ) of each tested compound for 48 h. Results: Antitumor biological activities represented as CC 50 were within the range >100-17±1 against leukaemia (MT4). The CC50 values were found to increase with increasing chain length of the substituent on the n itrogen atom. Conclusion: Antitumor activities of the tested dihydropyridones can be enhanced by incr easing chain length of the substituent on the nitrogen atom.


INTRODUCTION
Six-membered nitrogen heterocycles are key units in medicinal chemistry and versatile intermediates in organic synthesis (Dong et al., 2005;Comins and Ollinger, 2001). Dihydropyridones are important intermediates for the synthesis of natural products, particularly alkaloids (Elias et al., 2008) and they have been extensively investigated as valuable building block for the construction of piperidines, perhydroquinolens, indolizidines, quinolizidines and other alkaloid systems, with a wide range of a biological and pharmacological activities. These compounds known for their antiproliferative and antitubolin activities (Magedov et al., 2008) and as potential selective inhibitors of receptor tyrosyn kinase (Hu et al., 2008;Goodman et al., 2007). Their ability to induce leukaemic cell differentiation have been demonstarated (Pierce et al., 1981). In addition they have potent antimalarial activity (Yeats et al., 2008) and good anticonvulsant activity against acutely elicited Seizures (Revas et al., 2009). On the other hand curcumin is a principal curcuminoid of Indian curry and has known for its antitumor (Ran et al., 2009;Wohlmuth et al., 2010;Ljngman, 2009), antioxidant, antiinflamatory (Takahashi et al., 2009;Kuhad et al., 2007;Michaelidou and H-Litina, 2005) and antiarthritic properties (Patil et al., 2009).
Very little was published about the antitumor activities of dihydropyridones and the aim of this study is to investigate the relationship between structure and antitumor activity of a series of dihydropyridones derived from curcumin.

MATERIALS AND METHODS
The screened pyridones were synthesized by the reaction of curcumin and amines elsewhere (Elias et al., 2008). These compounds as well as curcumin were evaluated for preliminary estimation of the in vitro tumor inhibiting activity against a panel of tumor cell lines consisting of CD4 + human T-cells containing an integrated Human T-Leukaemia Virus type 1(HTLV-1), CD4 + human acute lumphoblastic leukaemia, human splenic B-lymphoblastoid cells, human acute Blymphoblastic leukaemia, human skin melanoma, human breast adenocarcinoma, human lung squamous carcinoma, human heptatocellular carcinoma, human prostate carcinoma, human foreskin fibroblasts and human lung fibroblasts, using microculture assay (MTT) method (Tang et al., 2010). This method is based on the metabolic reduction of 3-(4,5methylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). The cell lines of tumor subpanels were incubated within five concentrations (0.01-100 µg mL −1 ) of each tested compound for 48 h. Molecular descriptors for the studied compounds, logP, Hydration energy (∆H), Refractivity (Ref) and Polaraizability (POL) were calculated using HyperChem 8.5 program, after geometry optimization with the semi empirical RM1 Hamiltonian. The general molecular structure of the studied molecules is shown in Fig. 1.

RESULTS
The results of the antitumor activities, represented as CC 50 (µM) are summarized in Table 1.
The values of logP, Refractivity, Polarizibility increase with increasing molecular weight while hydration energy decreases with increasing molecular weight except for molecule 6.

DISCUSSION
All the tested compounds have antitumor activities less than those of curcumin against all tumor cell lines. This may be due to the lack to the β-diketone moiety in pyridones. It is obvious from Table 1 that the CC 50 value is increased with increasing chain length of the substituent on the nitrogen atom. Comparing the activity of compound 1 with other pyridones showed   Hydrophobicity constantof the substituent; Aobs: Observed biological activity expressed by Log (1/CC50); Apred: Predicted biological activity that the inclusion of a methylen or a phenyl group in the substituent moiety shifted the threshold of potency from inactive side towards activity in some of leukaemia lymphoma cell lines, particularly against the leukaemia cell lines MT4. For substituent longer than propyl group the compounds become active for most cell lines and in the case where R is hexyl group the antitumor activity becomes comparable to that of curcumin. Ignoring the data of compound 1 (CC 50 >100 for all cell lines) we tried to correlate the activity of the compounds 2-6 represented by Log(1/CC 50 ) against the leukaemia cell lines MT4 with the molecular descriptors, logP, refractivity, polarizability , hydration energy and carbon number of the substituent (C n ). Very good models with R 2 values 0.938, 0.957, 0.968, 0.957 and 0.955 respectively, were obtained when the data of compound 6 are not involved. The models are shown in Eq. 1-5: Equation 1-5 indicates a strong dependency of the activity on the alkyl chain length. However, when compound 6 involved in the regression equation poor models with low R 2 are predicted for all parameters except for ∆H. For example, in the case of the model including log P the correlation coefficient R 2 is 0.417, while for ∆H as a descriptor, a model with R 2 = 0.713 is obtained. This value became 0.957 when a double parameter regression equation including both ∆H and the hydrophobicity constant of the substituent (π) was used as shown in Eq. 6: The predicted biological activities for the dihydropyridones from Eq. 6 represented as Log (1/CC 50 ) are shown in Table 2.

CONCLUSION
This study has shown that the biological activity of the studied compounds increases with increasing chain length of the substituent on the nitrogen atom as well the activity could be predicted to good estimate on the basis of a model involving both hydration energy and the hydrophoibicity constant of the substituent.