In-vitro and in-silico evaluation of Brugmansia sauveolens' ability to treat asthma

1.        Introduction

Chronic respiratory disease is one of the main causes of mortality and morbidity worldwide. People of all ages are susceptible to asthma, a common chronic inflammatory disease of the airways. With more than 300 million people currently suffering from asthma worldwide, the burden of asthma is enormous (1). It has long been believed that asthma is the classic T helper type 2 (TH2) illness of the airways due to the high eosinophil count in the airways of persons with moderate asthma and results from mouse models. It is now known that some eosinophilic inflammation and some neutrophilic inflammation in asthma are controlled by type 2 innate lymphoid cells (ILC2 cells) acting in performance with basophils (2). The link between the main causes of asthma, airway inflammation, and oxidative stress in asthma is depicted in Figure 1. Exposure to external agents, such as allergens found indoors and outdoors, air pollutants like O3, NO2, PM (particulate matter), and respiratory tract infections, as seen in Figure 1(A), are the most frequent causes of these conditions. Oxidative stress is the phrase used to describe the imbalance between oxidants, such as reactive oxygen species (ROS) and reactive nitrogen species (RNS), and antioxidants in favour of oxidants after exposure. This illness could be categorized as a heterogeneous disorder. The primary environmental factors influencing the physiopathology of asthma are sensitization to dust, pollen, and diet. T lymphocytes and eosinophils release pro-inflammatory cytokines such as tumor necrosis factor- β (TNF-β), interleukins IL-2, IL-3, IL-4, IL-5, prostaglandins, histamine, and leukotrienes in response to antigens recognized by mast cells covered by IgE antibodies (Figure 1B).

Figure 1 Cellular mechanisms underlying the inflammation and oxidative stress in asthma. (A) The effect on airways after exposure to common external agents (air pollutants) such as O3, NO2, and PM; (B) Secondary mechanisms of pro-inflammation and aeroallergen deposition airway re-modeling that are involved in asthma.

In-silico studies have developed into a substantial and straight forward tool for analyzing diverse chemical compounds against numerous targets, with many advantages, including a significant decrease in the cost and time of drug discovery as well as the use of laboratory animals (3). There are a lot of pharmacologically active chemicals found in medicinal plants. Pharmaceutical components that can be employed in the creation of new drugs can be found in abundance in medicinal plants (4, 5). Beyond that, these plants have a big impact on how human societies have evolved all around the world. Anything about the biological function of possible drug candidates, including sequence-based properties and interactions with bodily structures, is considered a part of bioinformatics methods for drug development (6).

Bioinformatics methods for drug discovery include anything related to the biological function of potential drug candidates, including sequence-based characteristics, interactions with body structures (metabolites, proteins, cells, tissues, etc.), pathway perturbations, and toxicity, among others. Multi-omics and high-throughput sequencing are also major areas within bioinformatics (7). In the case of natural products; researchers make use of an entire range of techniques related to the organisms that produce the compounds. Drug discovery enabled by bioinformatics and computational methods is an increasing shift toward data-driven drug discovery. Traditionally, drug discovery has been performed as follows: scientists first find a target structure in the human body related to a disease or illness, followed by screening for “lead” compounds that show an affinity for the target. Although natural drug discovery has always been a hallmark activity in pharmacology, pharmaceutical researchers have only applied these chemo-informatics techniques to natural products recently (8).

This plant has been used in traditional medicine throughout cultures as a muscle relaxant, nematicide, aphrodisiac, hallucinogen, and treatment for inflammation, rheumatism, asthma, and allergies (10). Different chemical substances have been detected in the plant's flowers, fruits, stems, and roots, including alkaloids, volatile chemicals mostly terpenes, coumarins, flavonoids, steroids, and hydrocarbons. This medicinal species has been the subject of chemical investigations since 1996 (11). These investigations were the first qualitative research on particular classes of substances, which included amines, sugars, alkaloids, phenolic compounds, flavonoids, steroids, terpenoids, tannins, anthraquinone glycosides, saponins, and triterpenes (12). The ethanolic extract of the flowers was also used to quantify the alkaloids (5.903 0.01333 mg/g), phenolic compounds (3.435 0.0110 mg/g), and flavonoids (9.945 0.0256 mg/g) (13). The concentrations of these substances can alter, much like the volatile chemical profiles of living flowers, which are constantly changing due to both internal (genetic) and environmental (light, temperature, and water stress) variables. The loss of volatile chemicals and rapid deterioration are problems with cut flowers (14).

In the present study, our attention is on the anti-inflammatory process and the application of herbal plants that have anti-inflammatory properties. This study looks at the most recent information available in various search engines regarding the distribution, phytochemistry, pharmacology, and toxicology of B. suaveolens (Solanaceae) using chemical compounds or extracts at this time (9). To detect structural activity, we employed bioinformatics tools like PASS Online, Molinspiration, SEA, SWISS, and SLAP data. Data from these three phytoconstituents were compared to Phytoconstituents from B. sauveolens (15, 16). We evaluated B. suaveolens fractions and their in-vitro anti-inflammatory assay. We intended to conduct the anti-oxidative and the anti-inflammatory assays for all fractions, however, Covid-19 prevented us from doing so (17).

2.        Experimental Section

2.1.    Materials and Methods

2.1.1.   Materials for in-silico approach with the anti-inflammatory Phytoconstituents

Databases: We used biological databases such as PubChem, PDB (Protein Data Bank), and NCBI for a literature search for anti-inflammatory, Phytoconstituents in Indian medicinal plants used in asthma.

Software: Online tools were used for various in-silico predictions. PASS software was used to predict the biological activities of Phytoconstituents (9). Molinspiration software was used to predict the bioactivity score of the compounds against human molecular targets/ receptors such as GPCRs, ion channels, kinases, nuclear receptors, proteases, and enzymes (18-20). SEA is a tool for conducting structure-based fingerprint similarity searches and returns a ranked list of predictions, and comparative information on ligands known to share those targets (20). SWISS target returns a list of targets, with links to the relevant ChEMBL and Uniprot records, ranked by the pseudo-probability of ligand–targeted interaction (7). In SLAP, we sought to use such semantic methods to integrate and annotate the data about drug-target interaction (19).

Figure 2 Methodology for our bioinformatics studies: The different tools used to identify anti-inflammatory activity are shown in the figure. PASS software, Molinspiration, SEA, SWISS Target, and SLAP software are used to predict Phytoconstituents to identify anti-inflammatory activity.

2.2.    Materials for in-vitro approach (anti-inflammatory assay of Brugmansia extracts)

2.2.1.   Collection of plant material

Plants were collected from several areas in Kolkata during October and November.

2.2.2.   Chemicals and instruments

Methanol, petroleum ether, chloroform, ethyl acetate, n-butanol, sodium chloride, and aspirin were used. Centrifuge, electronic balance, and UV-VIS double beam spectrophotometer were used.

2.3.    Methods for in-silico approach

2.3.1.   Retrieval of phytochemicals and target protein

Firstly, we chose an anti-asthmatic formulation used in the market. Then we identified herbs in these formulations and categorized them based on their anti-inflammatory activity. The anti-inflammatory activities of herbs were collected in the literature analysis. Literature searches were based on experimentally determined properties. We chose Phytoconstituents such as cinnamaldehyde, piperine, and linolenic acid for further studies and standardization of the chosen parameters.

Protein Data Bank (PDB) archive is the single worldwide repository of information about the 3D structures of large biological molecules. 3D structures were downloaded in SDF format and canonical smiles were downloaded. 3D structures in SDF format were converted to PDB format.

2.3.2.   PASS (Prediction of Activity Spectra for Substances)

PASS software was utilized to predict the interactions of phytochemicals with selected targets (see Table 1). PASS can be accessed from the site www.way2drug.com/PASSOnline. PASS tool interpreted the biological active spectra using the 2D structure of molecules. PASS activity selected anti-inflammatory activity which has Pa (probability "to be active") greater than Pi (probability "to be inactive") and Pa greater than 0.5.

2.3.3.   Molinspiration, SEA, SWISS, and SLAP

The compounds were analyzed using the Molinspiration web tool based on Lipinski’s Rule of Five (www.molinspiration.com). Along with this, the bioactivity score was also predicted. A similarity ensemble approach (SEA) was used to study the potential targets of the test compounds (https://sea.bkslab.org/). Human targets related to inflammation were identified. Results showing Tanimoto coefficients (Tc) greater than 0.6 were selected. SWISS Target was used to study the potential human targets and a probability not to be less than 0.5 was used to collect significant data (http://www.swisstargetprediction.ch/). SLAP was also used to predict the targets of the test compounds and the benchmark p-value was set at 0.05 (http://cheminfov.informatics.indiana.edu:8080/slap/).

Figure 3 Structures of phytochemicals used for Molinspiration. The figure shows structures of Phytoconstituents used for this study. piperine, cinnamic aldehyde, and linoleic acid were used for forming a baseline. Phytoconstituents found in Brugmansia, atropine, and scopolamine (21).

2.4.    Methods for extraction, fractionation, and in-vitro approach for the anti-inflammatory assay

The leaves of B. sauveolens were subjected to shade drying. The dried plant material was further crushed to powder and stored in an airtight container for further analysis. The plant materials were first extracted with petroleum ether for defatting followed by methanol in the Soxhlet extractor. Further, the methanolic extract of the plant was fractionated with chloroform and ethyl acetate based on their polarity. The extracts and fractions were filtered and the solvent was evaporated to dryness over the hot water bath. All the dried extracts were kept in well-closed containers under refrigeration.

2.4.1.   In-vitro anti-inflammatory activity through membrane stabilization assay

Preparation of Red Blood cell (RBC) suspension: The Blood was collected from a healthy rat to the experiment and transferred to the centrifuge tubes. The tubes were centrifuged at 3000 rpm for 10 min and then washed three times with an equal volume of normal saline. The volume of blood was measured and reconstituted as a 10% v/v suspension with normal saline.

Figure 4 Extraction process in Soxhlet apparatus and fraction of plant material in separating funnel using a solvent

Heat-induced hemolysis: The reaction mixture (2 ml) consisted of a 1 ml test sample of different concentrations and 1 ml of 10% RBCs suspension and instead of a test sample only saline was added to the control test tube. Aspirin was used as a standard drug. All the centrifuged tubes containing the reaction mixture were incubated in a water bath at 56 ºC for 30min. At the end of the incubation, the tubes were cooled under a running tap water. The reaction mixture was centrifuged at 2500 rpm for 5 min and the absorbance of the supernatants was taken at 560 nm (22).

Percent Protection = 100 - [O.D of Sample]/ O.D of Control*100

3.        Results and Discussion

PASS Online: The pharmacological effects of a compound and its communication with biological entities were predicted by PASS online by uploading the SMILES string of natural compounds. PASS prediction results showed that the highest Pa value than the Pi value comes off for anti-inflammatory activity and hence selected compounds. However, all compounds have shown a significant Pa value as compared to the Pi value.

After that, we chose only those data which have Pa activity greater than 0.5. The Pa value is the active value of the compound so when Pa was greater than 0.5 it was good activity in comparison with Pi. We used the concept of ‘biological activity spectrum’ for a substance that shows a list of activity names, which reflects the result of the interaction of the chemical substance with different biological entities. If Pa < 0.5, the substance is unlikely to exhibit the activity in the experiment or the substance is likely to exhibit the activity in the experiment, but the probability is less. In Table 1, cinnamaldehyde and linolenic acid showed good anti-inflammatory activity. The highest Pa value of cinnamaldehyde is 0,596 and the highest Pi is 0,033; the highest Pa value of linolenic acid is 0,804 and the highest Pi is 0,006. This showed that the compounds exhibit potential anti-inflammatory activity.

Table 1 Biological activity spectrum of compounds- Pa (Active); Pi (Inactive)

Name of the compound	Pa	Pi	Activity
Cinnamaldehyde	0,596	0,033	Anti-inflammatory
	0,494	0,008	Anti-inflammatory, intestinal
	0,308	0,067	Anti-inflammatory, ophthalmic
	0,220	0,084	Non-steroidal anti-inflammatory agent
Piperine	0,314	0,060	Anti-inflammatory, intestinal
	0,344	0,127	Anti-inflammatory
	0,238	0,207	Anti-inflammatory, ophthalmic
Linolenic acid	0,804	0,006	Anti-inflammatory
	0,591	0,004	Anti-inflammatory, intestinal
	0,443	0,017	Non-steroidal anti-inflammatory agent
	0,373	0,017	Anti-inflammatory, ophthalmic

The molecular properties of the compounds were calculated using Molinspiration chemo-informatics software and are presented in Table 2 and Table 3. The Molinspiration tool retrieves results using the Canonical SMILIES in PubChem and the 3D structure of the compounds. A molecule having a bioactivity score of more than 0.00 is most likely to exhibit considerable biological activities, while values −5.0 to 0.00 are expected to be moderately active and if the score was less than −5.0 it was presumed to be inactive. The results reveal that the physiological actions of different compounds might involve multiple mechanisms and could be due to the interactions with GPCR ligands, nuclear receptor ligands, and inhibit protease and other enzymes. The identified compounds showed better bioactivity scores than Imiquimod and Acetylsalicylic acid for all drug targets.

Therefore, numbers of potential compounds and active human targets with their corresponding information of probability, p-VALUE, Max TC, p-value and score, and type information were retrieved, and are given in Table 3. Swiss targets of a molecule were predicted by identifying proteins with known ligands that are highly similar to the query molecule. In piperine target MAO-B has probability 1. As target prediction accuracy is affected by the molecule size probabilities were computed separately for each ligand size. In SEA data similarities were calculated for ligands and later successfully to drug target prediction. SEA calculates a raw score by summing up the Tanimoto similarity of the chemical structure. We found SEA and SLAP data on cinnamaldehyde and linolenic acid.

Table 2 Bioactivity scores of compounds tested. Imiquimod and acetyl-salicylic acid were used as standard anti-inflammatory chemicals

Drug/ compound	GPCR ligand	Ion channel modulator	Kinase inhibitor	Nuclear receptor ligand	Protease inhibitor	Enzyme inhibitor
Piperine	0.15	-0.18	-0.13	-0.13	-0.10	0.04
Linolenic acid	0.33	0.23	-0.19	0.35	0.13	0.42
Cinnamaldehyde	-1.09	-0.39	-1.24	-0.96	-0.76	-0.46
Imiquimod	0.27	0.21	0.60	-0.77	-0.29	0.48
Acetyl salicylic acid	-0.76	-0.32	-1.06	-0.44	-0.82	-0.28

 

Table 3 SEA, SWISS, and SLAP data of Piperine

 

Table 4 SEA, SWISS, data of cinnamaldehyde and linolenic acid

Name of compound	Target	Swiss target	SEA
		Probability	p-VALUE	Max TC
Cinnamaldehyde	TRPA1	1.0	2.819e-05	1.00
Linolenic acid	PPARG	0.739591175051	4.296e-09	1.00
	PPARA	0.739591175051	3.843e-08	1.00
	PPARD	0.739591175051	3.052e-06	0.85
	FABP4	0.714850037542	1.712e-07	0.79
	FFAR1	0.706805409756	0.000163	1.00
	PTGS1	0.665850737135	0.104	1.00
	FABP3	0.665850737135	6.282e-27	1.00
	OXER1	-	3.091e-62	1.00
	FFAR4	-	6.219e-11	1.00

3.1.    In-silico comparisons with Brugmansia sauveolens for anti-inflammatory properties

We compared the above data with B. sauveolens plant Phytoconstituents hyoscine, norhyoscine, atropine, apohyoscine. Pa value is the active value of the compound. So when the Pa value was greater than 0.5, it denotes a good activity. But when we compared Table 5, we observed that linolenic acid and cinnamaldehyde Pa>0.5 and hyoscine, norhyoscine, atropine, norhyoscine Pa<0.5. Therefore, the plant Phytoconstituents may show activity in an experiment, but the probability is less.

The molecular properties of the compound were calculated using Molinspiration Cheminformatics software and are presented in Table 6. The Molinspiration tool retrieves results using the Canonical SMILIES in PubChem and the 3D structure of the compounds.

A molecule having a bioactivity score of more than 0.00 is most expected to show significant biological activities, while values −5.0 to 0.00 are expected to be moderately active and if the score was less than −5.0 it was presumed to be inactive. In Table 6, B. sauveolens chemical constituents showed a good value of GPCR ligand, Ion channel modulator, Kinase inhibitor nuclear receptor ligand, Protease inhibitor, and Enzyme inhibitor. The identified compounds showed better bioactivity scores than Imiquimod and Acetylsalicylic acid for all drug targets.

Table 5 Biological activity spectrum of compounds – Pa (Active); Pi (Inactive) of B. Sauveolens chemical constituents

Name of compounds	Pa	Pi	Activity
Hyoscine	0,338	0,130	Anti-inflammatory
	0,292	0,092	Anti-inflammatory, ophthalmic
Norhyoscine	0,300	0,079	Anti-inflammatory, ophthalmic
	0,289	0,171	Anti-inflammatory
	0,200	0,189	Anti-inflammatory, intestinal
Atropine	0,338	0,037	Anti-inflammatory, ophthalmic
	0,379	0,107	Anti-inflammatory
	0,202	0,099	Non-steroidal, anti-inflammatory agent
	0,204	0,183	Anti-inflammatory, intestinal
Norhyoscine	0,300	0,079	Anti-inflammatory, ophthalmic
	0,289	0,171	Anti-inflammatory
	0,200	0,189	Anti-inflammatory, intestinal

 

Table 6 Bioactivity scores of B. sauveolens chemical constituents

Drug/compound	GPCR ligand	Ion channel modulator	Kinase inhibitor	Nuclear receptor ligand	Protease inhibitor	Enzyme inhibitor
Hyoscine	0.58	0.23	0.06	0.11	0.28	0.35
Norhyoscine	0.51	0.28	0.06	0.16	0.47	0.48
Atropine	0.44	0.26	-0.09	-0.06	0.13	0.21
Noratropine	0.43	0.35	-0.06	-0.10	0.35	0.35
Apohyoscine	0.48	0.50	-0.08	0.31	0.17	0.33

3.2.    In-vitro anti-inflammatory activities (Membrane Stabilization Assay)

The biological potential of plant extract fractions was studied for their ability to stabilize rat RBC membrane lyses in normal saline. The results obtained were also compared with standard anti-inflammatory agent aspirin based on the % protection. Initially, we decided aspirin was used as standard in the membrane stabilization assay and we performed a pilot experiment for aspirin to choose the correct dose of aspirin. We decide on the concentration for the following assay (Table 7, Figure 5)

 Table 7 Membrane Stabilization Assay of Aspirin for dose selection

 

Figure 5 Effect of aspirin on RBC membrane

The fractions were analysed at selected concentrations (500 µg/ml) for anti-inflammatory activity with membrane stabilization methods. Normal saline 500 µg/ml and aspirin100 µg/ml were used as a negative and positive control (Table 8).

Table 8 Membrane Stabilization Assay of Aspirin after dose selection

Concentrations	Mean	S. D
CCL4 ALK. (500µg/ml)	1.5	1.21
CCL4 NON.ALK (500µg/ml)	0.9	1.12
n-BUTANOL (500µg/ml)	1.5	1.23
ASPIRIN (100µg/ml)	1.6	0.74
Negative control	1.5	1.26

After that we decided to take only one fraction in different concentrations. So, we took CCl4 alkaloidal concentration in 100, 200, and 300 µg/ml, aspirin at 100 µg/ml and normal saline used as a negative control. But we performed only N=1 data in this (Figure 6 and Table 9).

Figure 6 Effect of different extracts on inhibition of Membrane Stabilization

Table 9 Effect of CCl4 alkaloidal fraction in red blood cells compared to the control group used as aspirin

Concentration	Sample 1	Sample 2	mean	control- sample	control – sample /control	*100
100 µg/ml	1.396	1.407	1.4	0.51	0.267	26.7
200 µg/ml	1.243	1.236	1.23	0.68	0.356	35.6
300 µg/ml	0.705	0.691	0.69	1.22	0.638	63.87
Aspirin 100 µg/ml	1.306	1.289	1.29	0.62	0.324	32.46
Negative control	1.901	1.925	1.91	-	-	-

4.        Conclusions

Firstly, we used in- silico methods for drug discovery in the natural plant during the previous decade. We choose three chemical constituents piperine, cinnamic aldehyde, and linoleic acid. After that, we identified structural activity by using bioinformatics tools such as PASS Online, Molinspiration, SEA, SWISS, and SLAP data. We compared the data of these three Phytoconstituents with B. sauveolens Phytoconstituents. On the other side, we have assessed the fractions of this plants and its in-vitro anti-inflammatory assay. Based on the in-silico and in-vitro study, we conclude that B. sauveolens has the potential for application in the treatment of asthma. Further studies such as in-vivo studies are required to validate the present findings.