Zinc-dependent peptides are found to be encoded by 10 % of the genes in the human genome. Among all the metalloproteins, zinc-dependent proteins are the major part. The zinc ion of metalloenzymes plays a crucial role in the enzyme catalytic activity [1,2]. There are some significant zinc-dependent metalloenzymes like matrix metalloproteinases (MMPs), TNF-α converting enzyme (TACE), carbonic anhydrases (CAs), angiotensin-converting enzyme (ACE), carboxypeptidase A, thermolysin, etc. These metalloproteinases are responsible for executing various biological functions [[3], [4], [5], [6], [7]]. Therefore, these enzymes are an excellent target as they are engaged in a variety of human ailments namely cardiovascular and neurodegenerative disorders, rheumatoid arthritis, and cancer [2]. Regarding drug design and discovery, zinc-dependent metalloenzymes are the most desired and promising target [8]. Moreover, drug design may be beneficial for understanding the contribution of the zinc ion with the drug candidate physiochemically and quantitatively inside the zinc-containing proteins. Among these metalloenzymes, MMPs, which are under the zinc-dependent endopeptidase classes, are found to be linked with the remodeling of extracellular matrix (ECM) and these are essential for many physiological processes like tissue repairing, morphogenesis, development, tissue homeostasis, etc [8,9]. Moreover, MMPs are associated with the availability of growth factors and take part extensively in cellular signaling to regulate crucial processes such as proliferation, migration, and apoptosis [10]. Generally, all MMPs are released as inactive forms. The ‘cysteine switch’ of MMPs is one of the prime reasons for the activation of MMP where the thiol function is unpaired, and the amino acid residue cysteine is restored with a water molecule [11]. Moreover, in this activation mechanism, the cysteine switch empowers the hydroxylation of the pro-peptide of the relatively activated MMPs. All MMPs are triggered in the cell except MT-MMPs, MMP-11, -23, and −28 [10]. There are 26 different types of MMPs. Their sequential homology as well as the biological consequences have already been reported [12]. MMPs are categorized into 5 distinct classes depending upon their specificity towards the particular substrate. These are (i) collagenase, (ii) gelatinase, (iii) stromelysin, (iv) membrane type-MMPs, and (v) miscellaneous MMPs. MMPs involve a significant role in tumor development and the invasiveness of inflammatory cells [13,14].

Among these MMPs, MMP-2 and MMP-9, belonging to the group of gelatinase, have a substantial role in the digestion of denatured collagen (gelatins) having a relation with tumor progression. Again, gelatinases are found to be highly expressed in the brain, lung, breast, and colon cancer [15]. Apart from that, MMP-9 is responsible for tumor growth by degrading extracellular matrix through proliferation along with the migration of endothelial cells. MMP-9 is also associated with the physiological and pathological processes namely healing of wounds, remodeling of bone, rheumatoid arthritis, tumor invasion, metastasis, and several other inflammatory processes [16]. Therefore, MMP-9 is a promising biomolecular object for the design and discovery of specific inhibitors against these pathological conditions.

Regarding the structure of the MMPs, these consist of a signal peptide, a pro-peptide, a catalytic domain as well as a zinc-binding region, along with a hemopexin-like domain which is associated with the hinge region [12,16]. Structurally, MMP-9 comprises fibronectin repeats at the catalytic site with 5-stranded sheets and 3-helices. The catalytic center contains a Zn2+ ion which coordinates three histidine amino acid residues (namely His401, His405, and His411) along with a crucial water molecule [17]. MMP-9 is distinguished from the other MMPs by the three cysteine-rich domains which are similar to fibronectin type II repeats. It is essential as far as their interactions with various substrates (such as collagen, gelatin, and laminin) are concerned [18]. Moreover, the catalytic domain of MMP-9 consists of a tunnel-like S1՛ pocket that is responsible for selective inhibition. Despite direct inhibition with selective inhibitors, MMP-9 inhibition may also be taken into consideration by decreasing the PI3K/AKT signaling pathway responsible for cancer proliferation through the upregulation of MMP-9 [19]. AKT, a serine/threonine-specific protein kinase, has been taken into consideration as one of the key modulators and the principal biomolecular target for cancer [20]. AKT is also cleaved by caspase-3 which has prime roles in cell death [21]. This mechanism helps in the deactivation of cellular signaling and subsequently, irreversibly triggers cellular death [22]. Therefore, inhibition of MMP-9 may be an important approach in anticancer drug design and discovery [12]. Regarding several approaches related to anticancer drug design and discovery, quantitative structure-activity relationship (QSAR) studies have been carried out extensively. This approach may also enforce the designing of new drug candidates and determine possible hits. QSAR may be applied also in the design of effective MMP-9-selective inhibitors as well as to identify the lead MMP-9 inhibitors as a promising anticancer agent. Again, the classification-based QSAR methodologies have been verified to be more economical and productive for identifying promising drug-like molecules.

In this current exploration, a diverse set comprising potential MMP-9 inhibitors and less effective MMP-9 inhibitors are selected to figure out the structural and physiochemical requirements with the help of several classification-based methods such as linear discriminant analysis (LDA), structure-activity relationship in Python (SARpy), Bayesian classification and recursive partitioning (RP) analyses. This is supposed to be the first classification-based analysis taking into account all MMP-9 inhibitors with binding affinity data (Ki in nM) retrieved from the binding database (BDB) [23]. In this current work, the abovementioned classification-based analyses are carried out to extract the crucial structural features of these existing MMP-9 inhibitors. Therefore, this present work may be fruitful for the design and discovery of novel MMP-9 inhibitors in the future.