Bangladesh, with its subtropical weather and fertile deltaic land, is rich in a variety of medicinal plants. The traditional medicine system of Bangladesh has developed over generations through trial and error. Bangladesh is the habitat of more than 500 medicinal plant species and the rural people have long depended on them for their primary healthcare needs (Ghani, 2003). Currently, about 250 medicinal plants are being used for the preparation of herbal formulations in Bangladesh and the estimated market value of medicinal plants is around 3.3 billion BDT (4.1 million USD) (Ghani, 2003; Bachar, 2013). The traditional healers of Bangladesh utilize these medicines of plant origin to treat various disease conditions including inflammation, hypertension, asthma, diabetes, gastro-intestinal disorders, cardiac problems, and skin diseases among different localities and communities throughout the country (Ghani, 2003; Rahmatullah et al., 2010a,b; Kadir et al., 2012; Islam et al., 2014).

Chrysopogon aciculatus (Retz.) Trin. of the Poaceae family is a vigorous creeping grass commonly growing in Bangladesh. Locally in Bangladesh, it is known as Premkata, Chorkata, Chui-kanta, and distributed all over the country. This tropical grassland species also grows in many other Asian countries including Malaysia, India, Nepal, Philippines, China and Indonesia, where it is commonly known as ‘Love Grass’(Ghani, 1998). A study conducted in Nepal revealed that this grass species is a popular cattle feed constituting 45% of total grazed pasture in lowland areas (Lehmkuhl, 1992). It is an exotic grass in Nigeria, where it is well established as a lawn grass (Stanfield, 1970). This medicinal herb plays an important ethnomedicinal role in different regions across the world and extensively used for the management of wide range of ailments. In lieu of diverse traditional uses, a few medicinal property of this weed has been evaluated and reported so far. In Bangladesh, the Garo ethnic community from Madhupur, Tangail uses C. aciculatus root juice to cure liver pain and the whole plant of C. aciculatus to treat cattle leg swelling, where it is known by the name Negraban and Negra bam (Anisuzzaman et al., 2007). Investigations revealed C. aciculatus to be rich in biologically important phytoconstituents belonging to several chemical classes including glycosylflavones, flavonols, sterols, flavones and germacranolides (Supplementary Table S1).

As a part of our continuing research on medicinal plants of Bangladesh, present study was undertaken to evaluate the local ethnomedicinal use, the antinociceptive activity of the ethanolic extract of C. aciculatus on animal model and to identify bioactive phytoconstituents by chemical profiling through LC-MS and GC-MS (Mondal et al., 2014; Sheikh et al., 2016; Zihad et al., 2018). Analgesic activity was observed in both peripheral and central model of analgesia. Furthermore, computational study was conducted with the identified compounds against human COX-2 enzyme. The aim of this study was to draw a scientific basis to the traditional use of this plant in mitigating pain among the traditional healthcare practitioners of Bangladesh.

The initial homology model of human COX-2 enzyme was selected based on the lowest dope profile and used for the docking (Figure 1). The sequence alignment is given in the Supplementary Materials (Supplementary Figure S3). The Ramachandran plot analysis showed that the number of residues in the favored, allowed and outlier regions are 97.8, 2.2, and 0.0%, respectively. In addition to the Z-score of -7.48 in knowledge base energy calculation, the low RMSD (root-mean-square deviation) of 0.8 to 0.9 Å between the overlaid model and template structure implied that the homology model was a theoretically reliable for molecular modeling study (Supplementary Figure S4). The homo-dimer model of COX-2 was generated from its monomeric units using GalaxyGemini following energy minimization step. The arachidonic acid binding pocket was identified with the following amino acid residues: Ala513, Val335, Gly512, Ser516, Tyr341, Phe343, Val102, Met99, Leu103, Val509, Leu338, Tyr371, and Ser339 (Supplementary Figure S5). Fourteen ligands obtained from the LC-MS and GC-MS analysis were docked into the active site of the human COX-2 homology model and nine different poses of the ligands in the binding site were analyzed (Supplementary Figures S6–S18) and the results are presented in Table 5. Six ligands, namely citronellyl isobutyrate, octadeca 9,12,15-trienoic acid, ([(4-oxo-3,5,6,7-tetrahydro-4Hcyclopenta[4,5] thieno [2,3-d]pyrimidin-2-yl)methyl] thio) acetic acid, campesterol, gamma oryzanol and tricin were found to interact with some amino acid residues at the catalytic site with hydrophobic interactions. Eleven out of thirteen compounds showed higher affinity for COX-2 than that of arachidonic acid (-6.4 kcal/mol) with the highest affinity observed for aciculatin and aurofusarin (-9.2 and -11.7 kcal/mol).

Acetic acid induced writhing reflex model in mice is a widely accepted and effective pain model for evaluating peripherally acting analgesics (Zulfiker et al., 2010). In this method, acetic acid induces pain sensation via localized inflammatory response resulting from local release of arachidonic acid and subsequent prostaglandins through cyclooxygenase pathway of the arachidonate cascade (Duarte et al., 1988). In other words, peritoneal administration of acetic acid raises the level of PGE₂ and PGF_2α in the peritoneal fluid and causes inflammatory pain (Zulfiker et al., 2010). In our study, ethanol extract of C. aciculatus significantly reduced the number of writhing in mice model and this decrease suggests that the test extract showed its analgesic activity through peripheral mechanism, which is the inhibition of prostaglandin biosynthesis by acting on visceral receptors sensitive to acetic acid (Sheikh et al., 2016). It is in agreement with the study reporting the presence of anti-inflammatory principle in this plant (Hsieh et al., 2011).

The hot plate method is a suitable assay for neurologic pain in evaluating centrally acting analgesics. This method measures a complex response to an acute and non-inflammatory nociception. Centrally acting analgesics like opioids usually act via supra spinal and spinal receptors; and therefore can increase the reaction time in hot plate test (Sabina et al., 2009; Mondal et al., 2014). Morphine, used as the standard drug in this test, is a renowned central analgesic that exerts its effect through binding with the opioid receptors (μ, κ, and δ) located in the post and pre synaptic membrane (Mondal et al., 2014). It was demonstrated that oral administration of C. aciculatus extract prolonged the response latency period to heat stimulus in test mice that increased at higher dose. The effect was observable from 30 min of treatment and declined at the end of the observation period (Table 2). As compared to the standard drug morphine, it can be suggested that the test extract has a mild central analgesic activity.

Results suggested that C. aciculatus has potential analgesic activity which is comparable to the standards, i.e., diclofenac sodium and morphine, and showed significant difference from the control. In both bioassays the extract showed positive dose dependent effect confirming that this plant has both peripheral and central analgesic activity. As per the results obtained in preliminary phytochemical screening, the ethanolic extract of C. aciculatus contains several important phytoconstituent classes including flavonoids, tannins, and alkaloids. There are several reports available that demonstrate analgesic and anti-inflammatory activity of different compounds from the stated classes (Pathak et al., 1991; Duke, 1992; Bittar et al., 2000; Bello-Ramírez et al., 2003; Wang et al., 2009; Chen et al., 2012). We also identified a number of compounds in the test extract through LC-MS and GC-MS that were previously reported from this plant and the Poaceae famly. Among these compounds, the anti-inflammatory activity of aciculatin is well established. Aciculatin inhibits the expression of cyclooxygenase-2 (COX-2) by suppressing NF-κB and JNK/p38 MAPK activation pathways (Hsieh et al., 2011). COX-2 is the inducible isoenzyme in the arachidonate cascade that is responsible for inflammatory pain (Sinatra, 2002). Thus inhibition of its expression will lead to pain inhibition. Moreover, this plant contains quercetin and daucosterol (Hu and Zheng, 2006). Quercetin acts as an analgesic both in neurogenic and inflammatory phases (Filho et al., 2008), whereas daucosterol reduces local inflammation (Mavar-Manga et al., 2008). So, these constituents along with aciculatin can be held responsible for the observed antinociceptive activity of C. aciculatus.

Molecular docking studies with the identified ligands against human COX-2 enzyme revealed that, among the thirteen phytoconstituents, six compounds interacted with a number of amino acid residues (Ala513, Val335, Gly512, Ser516, Tyr341, Phe343, Val102, Met99, Leu103, Val509, Leu338, Tyr371, Ser339) associated with the binding of arachidonic acid at the catalytic site, with binding affinity ranging between -6.2 and -8.4 kcal/mol. Among these six compounds, tricin, an O-methylated flavone, showed lower binding affinity (-8.2 kcal/mol) than that of aciculatin and aurofusarin, but with greater number (12) of interaction with the amino acid residues involved in the binding of arachidonic acid through hydrophobic interactions. Campesterol interacted with Leu338, Gly512, Met99, Val509, Leu103, Ser516, Tyr341, Phe343, Ala513 and Tyr371 with an affinity of -8.4 kcal/mol. Gamma oryzanol also interacted with 10 amino acid residues with an affinity of -7.3 kcal/mol. 9, 12, 15-octadecatrienoic acid (Binding affinity: -6.2 kcal/mol) and citronellyl isobutyrate (Binding affinity: -6.4 kcal/mol) showed interaction with nine and eight amino acid residues, respectively. The other seven compounds also showed some affinity for COX-2 enzyme but the amino acid residues were not involved in the binding of arachidonic acid. So the conclusion can be drawn from these results that, these identified secondary metabolites may have some part in the observed analgesic activity of this plant extract.

Results from the present study confirmed that C. aciculatus possesses significant and dose dependent analgesic activity, which supports the uses of the plant in folk medicine. This activity may be attributed to aciculatin, an anti-inflammatory glycosylflavone, identified in the plant and the identified phytoconstituents showing promising affinity toward human COX-2 enzyme in computational study.

This study was carried out in accordance with the internationally accepted principle for laboratory use and care (National Research Council of USA, 1996). The protocol was approved by a Animal Ethics Committee, Pharmacy Discipline, Life Science School, Khulna University, Khulna, Bangladesh (Protocol Number: KU/PHARM/AEC/15/006/025). Three members ethics committee consists of Dr. Ashish Kumar Das, Professor and chairman of the committee (dasasish03@yahoo.com), Dr. Jamil A Shilpi (jamilshilpi@yahoo.com), Professor and Dr. Shaikh Jamal Uddin, Associate Professor (uddinsj@yahoo.com); Pharmacy Discipline, Life Science School, Khulna University, Khulna, Bangladesh.

SZ, NB, and NS carried out the extraction and in-vivo studies under the supervision of SU, JS, AD, and SS. JS, HH, MI, and RR conducted the phytochemical profiling through LC-MS and GC-MS. SZ, SD, and MR conducted the computational study. SA carried out the statistical analysis. LN, SS, and SU designed the work. SZ and NS drafted the manuscript. EA, LN, and RR revised the manuscript. All authors have approved the manuscript for submission.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.