Herbal informatics: A unique model to identify the anti-cancerous agents for targeting lung cancer

The incidence of lung cancer has increased in recent years and causes major mortalities across the globe . Besides, the availability of the several chemotherapeutics modalities in the management, there is still a challenge to find out the efficient remedy with lesser or no toxic effects. Hence, there is a necessity to employ a complementary research to establish the effective management for the lung cancer. In this study, we have implemented a novel herbal informatics model to find out the alternative remedy in treatment of lung cancer. This model utilizes five major steps of bioprospection process based on the classical surge followed by the binary index and rationale-based selection of herbal products targeting the cancer-causing factors which is explained in detail in the methodology section of this model. This study revealed 07 herbals such as Withania somnifera (Ws), Berberis vulgaris(Bv), Glycyrrhiza glabra(Gg), Andrographis paniculate(Ap), Azadirachta indica(Ai), Cinnamomum verum(Cv), Piper longum(Pl) based on the fuzzy set optimization scoring(0.6-1) that could be further studied in vitro and in vivo level for utilization in the management of lung cancer.


1.Introduction
Lung cancer is a major health issue, that is caused due to a variety of genetic and environmental factors and is reported to cause the major mortalities all around the world [1,2]. Researchers, while predicting the mortalities related to lung cancer globally, estimated around 1.7 million deaths in 2018 [3]. Two major categories of lung cancer, namely, Small Cell Lung Cancer (SCLC) and Non Small Cell Lung Cancer (NSCLC) account for 15% and 85% of all Lung cancer cases respectively. Histologically and clinically, NSCLC is categorized into adenocarcinomas, squamous cell carcinomas and large cell carcinomas [4].
The various pathways like the tyrosine kinase family, epidermal growth factor receptors (EGFR) are over expressed in cancers. Various other anomalies have also been reported like, KRAS gene mutation, PI3K gene mutations, fusions in ALK genes etc. [5] Vascular endothelial growth factor (VEGF), that underlies angiogenesis is also reportedly increased in NSCLC and other cancers [6]. The gene encoding for Anaplastic Lymphoma Kinase(ALK) is reported to undergo rearrangement in some cases of non-small cell lung cancer (NSCLC). [7]. Anaplastic lymphoma kinase (ALK) and echinoderm microtubule-associated protein-like 4 (EML4) are over expressed in 3-13% of lung cancers. [8,9]. In hypoxic state Histone Deacetylases (HDAC) causes angiogenesis. [10]The phosphoinositide 3-kinase (PI3K) /downstream serine/threonine kinase (Akt )are found to be involved in anti-apoptotic signals in non-small cell lung cancers (NSCLC) [11,12]. Tubulins, the multifunctional proteins participate in chromosome separation during cell division, intracellular transport, cell shape modulation. The γ-tubulin in non-small cell lung cancer (NSCLC) is over expressed and also impacts patient survival at advanced stages of NSCLC [13] Fibroblast growth factor receptor (FGFR) is involved in regulation of cell proliferation, differentiation, shape and movement and causes angiogenesis [14,15]. FGF 9 factor, ligand of FGFR3, is over expressed in lung breast, prostate cancers . [16] Both autocrine and paracrine pathways simultaneously activate epidermal growth factor receptors (EGFR) to secrete epidermal growth factors molecules, which activate targeted signal transduction pathways and transfer cellular information from EGFR to the inner side of lung cells. [17] The epidermal growth factors promote lung metastasis through EGFR-receptor tyrosine kinases (RTK) by initiating and triggering signaling cascade in NSCLC and SCLC both. [18] Several factors/pathways can be targeted by active constituents present in herbal compounds to regulate the signaling pathways. As today apart from surgery, radiation, systematic therapy like chemotherapy, targeted therapy, antibody treatment and angiogenesis inhibitors medication are also given to lung cancer patients .
Various chemotherapeutics agents are also available for general and targeted therapy for lung cancer that is approved by FDA (The US Food and Drug Administration) and/or EMA (European Medicines Agency) such as carboplatin, cisplatin, paclitaxel, Gefitinib (anti-EGFR), Crizotinib (anti-MET, anti ALK and anti ROS1), Lorlatinib (anti-ALK) etc. [19]. The proliferating cancer cells are susceptible to chemotherapy but also has various side effects like nausea, diarrhea, weakness, lethargy, hair loss as routine. Some-times chemical drugs may have serious side effects like myelotoxicity [20], cardiotoxicity [21] , renal toxicity [22] Chemotherapy medicines may affect blood-forming cells in bone marrow. The long term usage of anti-cancer drugs is seen in increasing mutation in EGFR gene and amplifying mesenchymal-epithelial transition (c-MET) protooncogene and it also leads to the development of non-responsiveness to these therapies. [23] Herbal compounds are most popular among all medications known till date, and has several medicinal properties like healing, immune boosting, anti-bacterial, rejuvenating and purifier activity in them. They are being used as primary medicines to treat number of disorders with less or no side effects. [24] In Ayurveda , thousands of years ago ,the herbals with their therapeutic uses were described to treat various human disorders as various forms of concoction. The NIH National cancer institute (NCI) reported 3000 herbals with confirmed anticancer properties [25]. Herbal derived anticancer medicines have been categorized into four classes according to their mechanism of action such as Tubulin inhibitors (vinca alkaloids, taxanes), HDAC inhibitors (various isoflavonoids), Topoisomerase I and II inhibitors ( Camptothecins and podophyllotoxins) and DNA disrupters (epigallocatechin-3-gallate) [26]. Various herbals have been proven to exert anti-cancer mechanisms via regulating various pathways. Neem (Azadirachta indica), a well-known Ayurvedic herb, has been shown to exert various activities like antibacterial, antifungal, anti-inflammatory etc. It has also displayed anticarcinogenic activities via phytochemicals like quercetin, nimbolide etc. [27,28] In this research article, various other herbal compounds have been described in Table 2, with similar properties. These herbals have a number of active ingredients, but these molecules and their target hits have to be identified. The various side effects of chemotherapy in patients has persuaded scientist to study herbals derivatives as alternate medicines to alleviate disease like cancer. The complex nature of lung cancer poses challenges to present medicines available and persuades researchers to find new medicines. The present insufficient systematic research for herbals and their specific molecule is a limiting factor for lung cancer treatment. The traditional process is tedious and time consuming and thereby irrational. In this study in-silico herbal informatic model, we are bio-prospecting the several medicinal herbs to identify their potential for targeting specific factors/pathways that are responsible for lung cancer. Web based data search engine, matrix linked data mining and Fuzzy score  Table 1.
Similarly, database of herbals was prepared based on the selection criterions included Ethno-pharmacological properties, Traditional uses, Availability at ease, therapeutic potential associated with indirect indications and literature support as exemplified in i. Family of proteins that plays important role in apoptosis and it causes prevention of programmed death of the cells. ii. The expression of BCL-2 and thus the subsequent delay in apoptosis has been observed in both SCLC and NSCLC. iii. Research showed that phytochemical from Mentha arvensis, could cause down regulation of BCL-2 and this can also be used for its anticancer property.
The primary database set of Herbal Plants combined with the presence/absence of a given parameter was prepared using the web search engine, PubMed (N= first 20 hits) with a median value as cut off filter value. [32]

Relevance factor linked Binary weightage matrix-based analysis
The relevance of each cancer-causing factor was used for calculation of its respective Weightage score and subsequent analysis by consideration of every plant with respect to all the cancer causing factors taken together was performed. These parameters for weightage calculation increased the 'uncertainty factor' required for statistical analysis to be viable and bring about reduction in "Investigator's prejudice". Herbals with binary score ≥10, from previous step were taken into consideration for further study.

Fuzzy set membership analysis for decision matrix & optimization
The mathematical relationship as mentioned below (Equation. 2) was used to assign the relative relevance via Fuzzy score analysis within identified group of plants.

Classical literature surge:
The classical literature search brought forward 10 cancer causing factors as shown in   Figure 1).

Relevance factor linked binary-weightage matrix-based analysis
The binary matrix analysis of presence/absence of the considered factors in 50 herbals revealed that 20 herbals showed binary score from 10 to 14. The consequent weightage score matrix analysis of 20 herbals revealed that seven herbals are having more than 10 and these are selected for present study.

Decision matrix-based Optimization
Based on decision matrix analysis, 7 herbals showed highest fuzzy score range from 0.6-1. Amongst these, Withania somnifera (score=1) held the topmost position with μS score being 1, relative to the lowest μS score

Discussion
The chemotherapy kills the proliferating cancer cells but their long-term uses may cause other serious pathologies like myelotoxicity [20], cardiotoxicity [21], renal toxicity [22], mutation in EGFR gene to the patients [23]. As evident from large number of mortalities because of lung cancer every year or with poor recovery rates the chemotherapy alone is not able to control lung cancer completely . According to the latest lung cancer statistics, the five year survival rate is 19% which has suggested, there is still a need to find alternative medicines with the holistic approach in treatment. Herbal compounds have various medicinal properties like anti-oxidant , anti-inflammatory, immune booster, antiviral, antibacterial properties etc. as described in Table 2 and number of herbals have shown anti-tumor activities also like Withania somnifera,, Cinammomum verum, Nigella sativa, Vinca rosea.
In this present study 'in silico' bioprospection analysis is done, which includes using of PubMed as a random search engine for identification of the potential herbal agents based on rationale selection. Relevant factors (Bio-activity parameters) were selected with respect to their involvement (both direct and indirect) in the pathogenesis of lung cancer. The binary matrix approach is used to identify the herbal agents in this paper on the basis of fuzzy score analysis and these herbals have been shortlisted on the basis of all or none principle, with inclusion of herbal agents only with binary score ≥10 out of 14. After the first screening step, out of 50 herbals, 20 herbals were screened out for performing weightage matrix and the fuzzy set membership analysis, which finally provided a database of 07 herbals ( Figure 2) that showed an acceptable optimized score such as  and their anti-cancer potential can be further tapped in alleviating the disease for drug discovery process.

Conclusion
Herbal informatic study has provided 07 Herbal compounds i.e., Withania somnifera(Ws), Berberis vulgaris(Bv), Glycyrrhiza glabra(Gg), Andrographis paniculate(Ap), Azadirachta indica(Ai), Cinnamomum verum(Cv), Piper longum(Pl) with significant therapeutic potential targeting several disease factor of lung cancers. This study needs to be cross validated at docking level and at in vitro and in vivo level to further establish their role in management of cancer