Chemometric Modeling of 5-Phenylthiophenecarboxylic Acid Derivatives as Anti-Rheumatic Agents
Amit K. Halder,
Milan K. Maiti,
Arthritis involves joint inflammation, synovial proliferation and damage of cartilage. Interleukin-1 undergoes
acute and chronic inflammatory mechanisms of arthritis. Non-steroidal anti-inflammatory drugs can produce symptomatic
relief but cannot act through mechanisms of arthritis. Diseases modifying anti-rheumatoid drugs reduce the symptoms of
arthritis like decrease in pain and disability score, reduction of swollen joints, articular index and serum concentration of
acute phage proteins. Recently, some literature references are obtained on molecular modeling of antirheumatic agents.
We have tried chemometric modeling through 2D-QSAR studies on a dataset of fifty-one compounds out of which fortyfour
5-Phenylthiophenecarboxylic acid derivatives have IL-1 inhibitory activity and forty-six 5-Phenylthiophenecarboxylic
acid derivatives have %AIA suppressive activity. The work was done to find out the structural requirements of
these anti-rheumatic agents. 2D QSAR models were generated by 2D and 3D descriptors by using multiple linear
regression and partial least square method where IL-1 antagonism was considered as the biological activity parameter.
Statistically significant models were developed on the training set developed by k-means cluster analysis. Sterimol
parameters, electronic interaction at atom number 9, 2D autocorrelation descriptors, information content descriptor,
average connectivity index chi-3, radial distribution function, Balaban 3D index and 3D-MoRSE descriptors were found
to play crucial roles to modulate IL-1 inhibitory activity. 2D autocorrelation descriptors like Broto-Moreau autocorrelation
of topological structure-lag 3 weighted by atomic van der Waals volumes, Geary autocorrelation-lag 7 associated with
weighted atomic Sanderson electronegativities and 3D-MoRSE descriptors like 3D-MoRSE-signal 22 related to atomic
van der Waals volumes, 3D-MoRSE-signal 28 related to atomic van der Waals volumes and 3D-MoRSE-signal 9 which
was unweighted, were found to play important roles to model %AIA suppressive activity.
Keywords: Adjuvant induced arthritis, Interleukin-1, k-means cluster analysis, multiple linear regression, partial least square, validation, QSAR, tumor necrosis factor-α, topological descriptors, Balaban 3D index
Rights & PermissionsPrintExport