Ethyl p-methoxycinnamate Exhibits Superior Multi-Modal Anti-Inflammatory Activity Compared to Structurally Related Cinnamic Acid Derivatives

Abstract

Objectives

This study aimed to investigate the anti-inflammatory potential of cinnamic acid (CA) and its derivatives, ethyl p-methoxycinnamate (EPMC) and trans-4-methoxy cinnamic acid (APMC), using integrated in silico, in vitro, and in vivo approaches to identify safer alternatives to conventional nonsteroidal anti-inflammatory drugs (NSAIDs).

Methods

Molecular docking was performed to evaluate binding interactions with inflammation-related proteins, including heat shock protein 90 alpha family class A member 1 (HSP90AA1), Janus kinase 2 (JAK2), prostaglandin-endoperoxide synthase 2 (PTGS2), lipoxygenase, heat shock protein 90 beta family class B member 1 (HSP90AB1), and nitric oxide synthase 3 (NOS3). In vitro anti-inflammatory activity was assessed using bovine serum albumin (BSA) denaturation assays to determine the half-maximal inhibitory concentration (IC50). In vivo efficacy was evaluated using a carrageenan-induced paw edema model in mice (n = 3 per group), and hematological analysis was conducted 3 hours post-induction.

Results

Molecular docking revealed that EPMC exhibited superior multi-target binding affinities across five inflammatory proteins compared with allyl p-methoxycinnamate (APMC) and CA, with notable interaction with prostaglandin-endoperoxide synthase 2 (PTGS2/COX-2) interaction (–6.29 kcal/mol). CA uniquely bound NOS3 (–3.06 kcal/mol), suggesting distinct mechanistic pathways. BSA denaturation assays demonstrated comparable IC50 values for EPMC (170.02 μg/mL), CA (171.48 μg/mL), and diclofenac sodium (165.05 μg/mL), whereas APMC exhibited weaker activity (215.06 μg/mL). In vivo, EPMC produced the most rapid and complete resolution of inflammation, achieving significantly lower area-under-curve values than diclofenac sodium (p < 0.05). Hematological analysis revealed mechanistic divergence APMC (600 mg/kg) tended to normalize white blood cell (WBC) and lymphocyte counts, suggesting systemic immunomodulation, whereas EPMC and CA demonstrated localized anti-inflammatory action without hematological effects.

Conclusion

EPMC demonstrates superior multi-modal anti-inflammatory activity through multi-target engagement and localized tissue action, positioning it as a promising lead for next-generation anti-inflammatory therapeutics with potentially improved safety profiles compared with conventional NSAIDs.

INTRODUCTION

Inflammation is a biological response to harmful stimuli, including pathogens, damaged cells, or irritants [1], and plays a critical role in the body’s defense mechanisms. Inflammation has been identified as a contributing factor in the pathogenesis of numerous clinical diseases, including rheumatoid arthritis, neurodegenerative conditions, and certain types of cancer [2-4]. Anti-inflammatory agents that target key pro-inflammatory mediators, such as cyclooxygenase (COX) enzymes, tumor necrosis factor-alpha (TNF-α), and various interleukins, have been shown to manage inflammation effectively [5, 6].

However, current anti-inflammatory therapies, particularly nonsteroidal anti-inflammatory drugs (NSAIDs), have substantial limitations. NSAIDs use is associated with increased risks of gastrointestinal complications (2-4% annually) and cardiovascular events, with annualized excess risks of 7-9 non-fatal and 2 fatal cardiovascular events per 1,000 high-risk patients [7-9]. These adverse effects highlight the need for safer alternatives. Natural products with multi-target capabilities offer promising opportunities, as more than 50% of US Food and Drug Administration (FDA)-approved drugs from 1939 to 2016 originated from natural sources, demonstrating favorable safety profiles and multi-target modulation [10, 11]. Multi-target agents can enhance therapeutic efficacy while reducing resistance development and off-target effects, which is particularly valuable for complex diseases such as inflammation [12, 13]. Cinnamic acid (CA) and its derivatives have emerged as promising multi-target anti-inflammatory candidates. These compounds exhibit diverse pharmacological activities, including cyclooxygenase (COX) inhibition, nuclear factor kappa B (NF-κb) pathway modulation [14, 15]. One of derivative, ethyl p-methoxycinnamate (EPMC), derived from Kaempferia galanga, effectively blocks COX-1 and COX-2 pathways and inhibits protein denaturation [16, 17]. The unique anti-inflammatory properties of these derivatives arise from their chemical modifications: CA possesses a hydrophilic carboxyl group with limited membrane permeability, whereas EPMC’s ethyl ester group enhances lipophilicity and cellular absorption, resulting in superior anti-inflammatory activity [17, 18]. Another derivative, trans-4-methoxy cinnamic acid (APMC) retains the carboxylic acid group with a para-methoxy substitution that enhances reactive oxygen species (ROS) stabilization [18, 19]. Structure-activity relationship studies confirm that lipophilicity, as well as steric and electronic factors, substantially influence the biological activities of CA derivatives [20, 21].

Despite these advances, the mechanisms underlying the anti-inflammatory effects of CA and its derivatives remain poorly understood, particularly regarding their molecular interactions with multiple inflammatory protein targets and their differential effects on systemic versus local inflammatory responses. To address this knowledge gap, this study aimed to employ an integrated approach that combined in silico molecular docking to predict binding affinities for multiple inflammatory targets, an in vitro bovine serum albumin (BSA) denaturation assay to assess anti-inflammatory potency, and an in vivo carrageenan-induced inflammation model with comprehensive hematological analysis. This multi-modal investigation bridges computational predictions with experimental validation to elucidate the anti-inflammatory mechanisms and therapeutic potential of CA derivatives.

MATERIALS AND METHODS

1. Molecular docking studies

Molecular docking was performed to predict binding interactions between CA derivatives and inflammation-related protein targets. Protein structures were retrieved from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB, https://www.rcsb.org/), while compound structures were obtained from PubChem (https://pubchem.ncbi.nlm.nih.gov/) and subsequently optimized using Avogadro software. Docking simulations were performed using AutoDockTools, and visualization was conducted with BIOVIA Discovery Studio 2021.

Six inflammation-related protein targets were selected based on their established roles in inflammatory pathways: heat shock protein 90 alpha family class A member 1 (HSP90AA1, PDB ID: 2XJX) and heat shock protein 90 beta family class B member 1 (HSP90AB1, PDB ID: 3NMQ) for protein chaperoning in inflammatory responses; Janus kinase 2 (JAK2, (PDB ID: 6VGL) for JAK–signal transducer and activator of transcription (JAK–STAT) signaling; prostaglandin-endoperoxide synthase 2/cyclooxygenase-2 (PTGS2/COX-2, PDB ID: 5IKR) for prostaglandin synthesis; lipoxygenase (PDB ID: 6NCF) for leukotriene production; and nitric oxide synthase 3 (NOS3, PDB ID: 3EAH) for nitric oxide synthesis. These targets represent diverse mechanisms involved in the initiation, propagation, and resolution of inflammation.

Protein structures were prepared by removing water molecules, heteroatoms, and co-crystallized ligands that were irrelevant to the study. Missing hydrogen atoms were added [22], and Gasteiger charges were assigned. Non-polar hydrogen atoms were merged to ensure proper electrostatic interactions and compatibility with the docking software [23]. The candidate ligands employed were CA, EPMC, and APMC. Ligand structures were redrawn and energy-minimized using Avogadro software to obtain optimal three-dimensional conformations.

Docking protocol validation was performed by re-docking the native ligand into each protein structure. The root mean square deviation (RMSD) between the docked pose and the crystallographic pose of the native ligand was calculated. RMSD values < 2 Å indicate valid docking protocols [19]. All proteins exhibited RMSD values within this acceptable range (Table 1), confirming the reliability of the docking methodology.

Table 1.

Molecular docking parameters for CA and its derivatives with inflammation-related proteins

PDB ID	Native ligand	Grid box size	Coordinates	RMSD
2XJX	Onalespib	24; 26; 36	14.101 –2.869 5.072	1.41
6VGL	Ruxolitinib	28; 30; 30	7.779 –27.548 52.805	0.89
5IKR	Mefenamic acid	28; 22; 24	38.042 2.131 61.279	0.55
6NCF	AF7	44; 20; 22	11.277 –21.891 –18.408	1.03
3NMQ	7PP	28; 30; 22	2.323 10.362 27.13	1.41
3EAH	327	18; 12; 14	10.543 16.384 58.088	1.32

For molecular docking simulations, grid box dimensions and coordinates were adjusted to encompass the active site of each protein where the native ligand binds (Table 1). The Lamarckian genetic algorithm was employed with the following parameters: 100 genetic algorithm runs, population size of 150, a maximum of 2,500,000 energy evaluations, and a maximum of 27,000 generations, with other parameters set to default [24]. Docking results were analyzed to determine the binding energy values (kcal/mol) for each ligand-protein complex. Conformations with the lowest binding energies were selected for further analysis and compared with native ligand binding energies.

Protein-ligand complexes with the lowest binding energy values were subjected to detailed two-dimensional interaction analysis using BIOVIA Discovery Studio 2021 software. Visualization highlighted amino acid residues interacting with the ligand via hydrogen bonds, hydrophobic contacts, and other non-covalent interactions [22, 23]. Comparison with the corresponding native ligand interactions enabled the identification of shared binding residues and the assessment of binding-mode similarities, indicating comparable binding sites and potential competitive inhibition mechanisms.

2. BSA denaturation assay

Various concentrations of CA, EPMC, and APMC were prepared in ethanol. Diclofenac sodium was used as a reference drug. For each test, 500 µL of the test sample was added to 4.5 mL 0.2% BSA. A control was prepared using ethanol instead of the test sample. The mixture was incubated at 37℃ for 20 minutes, then heated to 70℃ for 1 minute to denature the protein. After heating, the mixture was shaken and allowed to cool to room temperature. Turbidimetric measurement was performed using ultraviolet–visible (UV-Vis) spectrophotometry at 660 nm. The percentage of inhibition was calculated using the following equation:

$$\%Inhibition=\frac{ControlAbsorbance-SampleAbsorbance}{ControlAbsorbance}\times 100$$

Concentration-response curves were generated by plotting percentage inhibition against compound concentration. Model selection for half-maximal inhibitory concentration (IC₅₀) calculation was based on visual inspection of the concentration-response patterns and goodness-of-fit assessment. CA displayed a sigmoidal pattern, suggesting receptor saturation kinetics, whereas APMC, EPMC, and diclofenac sodium exhibited linear responses within the tested concentration range. For CA, IC₅₀ values were calculated using a four-parameter logistic (4PL) model with least squares regression fitting and a 95% confidence interval in GraphPad Prism^®. For APMC, EPMC, and diclofenac sodium, IC₅₀ values were calculated using linear regression models in GraphPad Prism^®. All experiments were performed in triplicate.

3. In vivo carrageenan-induced anti-inflammatory assay

Male ddY mice (20-40 g, 8-12 weeks old) were obtained from Patriot Lampung and housed under standard laboratory conditions with a 12-hour light/dark cycle, temperature of 22 ± 2℃, and humidity of 55 ± 10%. Mice that appeared sick or died before or during the experiment were excluded from the study. The study protocol was approved by the Committee on the Ethics of Experiments of Yayasan Kartika Eka Paksi, Universitas Jenderal Achmad Yani, Yogyakarta, Indonesia (Skep/81/KEP/III/2025).

A total of 27 ddY mice were randomly assigned to nine groups (n = 3 mice per group). The sample size was determined using the resource equation, with the degree of freedom (DF) for analysis of variance (ANOVA) within the range of 10 to 20 [25]. Randomization was performed using a simple random allocation method during the acclimatization period. Blinding was not implemented in this experiment. To minimize potential confounders, all procedures were conducted on a single day by the same experimenter, with precise timing for each animal to ensure accurate response-time curves. Mice were acclimatized for 7 days with ad libitum access to food and water and were fasted for 2 hours before experimental procedures.

The nine experimental groups were: (1) normal control group – no carrageenan injection and no treatment; (2) negative control group – carrageenan-induced without treatment; (3) positive control group – carrageenan-induced and treated with diclofenac sodium (100 mg/kg body weight [BW]); (4) CA 150 mg/kg BW; (5) CA 600 mg/kg BW; (6) APMC 150 mg/kg BW; (7) APMC 600 mg/kg BW; (8) EPMC 150 mg/kg BW; and (9) EPMC 600 mg/kg BW.

The doses of 150 mg/kg BW and 600 mg/kg BW for APMC and EPMC were selected based on previous studies demonstrating that EPMC at 100-800 mg/kg BW significantly reduced carrageenan-induced paw edema in rats [24]. These doses correspond to 200-1,600 mg/kg BW in mice [26]. Additionally, a previous report indicated that 14.4 mg/kg BW of CA (equivalent to 28.8 mg/kg BW in mice [26]) exhibited limited anti-inflammatory activity [27]; therefore, higher doses were selected to ensure adequate anti-inflammatory responses.

Diclofenac sodium 100 mg/kg BW and the CA, EPMC, and APMC were prepared in 1% sodium carboxymethylcellulose (Na-CMC) solution and administered orally. The negative control group received only 1% Na-CMC solution. Treatments were administered 1 hour before carrageenan injection.

Baseline thickness of the right hind paw was measured using a micrometer. One hour after oral treatment, 0.05 mL of 1% kappa-carrageenan solution was injected intraplantarly into the right hind paw to induce localized inflammation. Paw thickness was measured at 1, 2, 3, and 4 hours post-injection to monitor the inflammatory response. At 3 hours post-induction, blood samples were collected from the lateral tail vein for hematological analysis.

Hematological parameters assessed included WBC count, lymphocyte count, lymphocyte percentage, platelet count, plateletcrit, platelet large cell coefficient, hemoglobin (Hb), and hematocrit (Hct). Analysis was performed using an ONETECH Hematology Analyzer OT300. These parameters were selected to evaluate systemic inflammatory responses and potential effects on immune cell populations and coagulation.

4. Statistical analysis

Statistical analyses were performed on paw thickness and hematological parameters. Before analysis, normality was assessed using the Shapiro-Wilk test, and homogeneity of variances was evaluated using Levene’s test. Data meeting the assumptions of normality and homogeneity (p > 0.05 for both tests) were analyzed using one-way ANOVA. For paw thickness, the area under the curve (AUC) over the 4-hour observation period was calculated using the trapezoidal rule to provide a cumulative measure of inflammation. AUC values were compared among groups using one-way ANOVA followed by Sidak’s multiple-comparison test in GraphPad Prism 9. Blood parameters were analyzed individually using one-way ANOVA with Dunnett’s multiple comparison test, comparing all groups to the negative control.

A compound was considered to have a statistically significant effect if the ANOVA yielded p < 0.05 and the multiple-comparison test showed p < 0.05 compared to the appropriate control group (negative control for inflammation parameters; normal control for hematological parameters). Data are presented as mean ± standard error of the mean (SEM). Statistical significance was set at p < 0.05 for all analyses.

RESULTS

1. Molecular docking

Validation of the docking protocol confirmed its reliability, with all protein targets demonstrating RMSD values < 2 Å (Table 1), ranging from 0.55 Å (PTGS2) to 1.41 Å (HSP90AA1, HSP90AB1). Molecular docking revealed distinct binding patterns among the tested compounds (Table 2). EPMC exhibited the most consistent binding energies across multiple targets: –5.27 kcal/mol (HSP90AA1), –5.55 kcal/mol (JAK2), –6.29 kcal/mol (PTGS2), –5.18 kcal/mol (lipoxygenase), and –5.22 kcal/mol (HSP90AB1), suggesting potential multi-target engagement. APMC showed intermediate affinities (–3.81 to –5.19 kcal/mol), whereas CA exhibited the weakest binding (–3.80 to –5.39 kcal/mol) for most proteins.

Table 2.

Binding energies (kcal/mol) of CA, EPMC, and APMC against inflammation-related proteins

Target protein (PDB ID)	Ligand name	Binding energy (kcal/mol)*
HSP90AA1 (2XJX)	Native ligand (Onalespib)	–9.14
	Cinnamic acid	–3.93
	Ethyl p-methoxycinnamate	–5.27
	Trans-4-methoxy cinnamic acid	–4.09
JAK2 (6VGL)	Native ligand (Ruxolitinib)	–8.68
	Cinnamic acid	–4.46
	Ethyl p-methoxycinnamate	–5.55
	Trans-4-methoxy cinnamic acid	–5.02
PTGS2 (5IKR)	Native ligand (Mefenamic acid)	–7.6
	Cinnamic acid	–5.39
	Ethyl p-methoxycinnamate	–5.56
	Trans-4-methoxy cinnamic acid	–5.19
Lipoxygenase (6NCF)	Native ligand (AF7)	–8.72
	Cinnamic acid	–4.67
	Ethyl p-methoxycinnamate	–5.18
	Trans-4-methoxy cinnamic acid	–4.96
HSP90AB1 (3NMQ)	Native ligand (7PP)	–9.42
	Cinnamic acid	–3.8
	Ethyl p-methoxycinnamate	–5.22
	Trans-4-methoxy cinnamic acid	–3.81
NOS3 (3EAH)	Native ligand (327)	–4.6
	Cinnamic acid	–3.06
	Ethyl p-methoxycinnamate	+21158.85
	Trans-4-methoxy cinnamic acid	+22350.66

*The highlighted value indicates the lowest binding energy among the candidate ligands.

Notably, for NOS3, CA was the only compound exhibiting negative binding energy (–3.06 kcal/mol), whereas EPMC and APMC showed highly unfavorable interactions (+21158.85 and +22350.66 kcal/mol), indicating CA’s unique structural compatibility with the NOS3 active site.

Two-dimensional visualization (Table 3) revealed that native ligands and candidate compounds shared similar amino acid residues, indicating comparable binding sites. For PTGS2, both mefenamic acid and EPMC formed hydrogen bonds with Tyr385, a key residue in the COX-2 active site. Similar shared residues were observed for HSP90AA1 (Leu48, Gly97), JAK2 (Lys1030, Pro1002), and lipoxygenase (Arg101, Val110), supporting EPMC’s potential for competitive binding across multiple inflammation-related targets.

Table 3.

Two-dimensional visualization of protein-ligand interaction

Target protein (PDB ID)	Ligand name	2D visualization	Hydrogen bond interaction*
HSP90AA1 (2XJX)	Native ligand (Onalespib)		Leu48, Asp93, Gly97, Asn51, Asp54
	Ethyl p-methoxycinnamate		Leu48, Gly97
JAK2 (6VGL)	Native ligand (Ruxolitinib)		Lys1030, Pro1002
	Ethyl p-methoxycinnamate		Lys1030, Pro1002
PTGS2 (5IKR)	Native ligand (Mefenamic acid)		Tyr385, Ser530
	Ethyl p-methoxycinnamate		Tyr385
Lipoxygenase (6NCF)	Native ligand (AF7)		Arg101, Val110, His130, Thr137, Arg138, Val109, Glu134
	Ethyl p-methoxycinnamate		Arg101, Val110, Ile126
HSP90AB1 (3NMQ)	Native ligand (7PP)		Lys58, Asp93
	Ethyl p-methoxycinnamate		Asp39
NOS3 (3EAH)	Native ligand (327)		Glu327, Trp322
	Cinnamic acid		Gln213, Tyr323
Scale	-		-

*The highlighted value indicates similar amino acids between the native and candidate ligands.

2. BSA denaturation assay

All tested compounds demonstrated dose-dependent inhibition of BSA denaturation (Fig. 1). CA exhibited a sigmoidal concentration-response curve consistent with saturation kinetics, whereas APMC, EPMC, and diclofenac sodium displayed linear relationships. Maximum inhibitory activities were as follows: diclofenac sodium, 77.74%; CA, 71.16%; EPMC, 69.06%; and APMC, 54.49%.

Figure 1.

Concentration response curve of CA (a), APMC (b), EPMC, and diclofenac sodium in the BSA denaturation assay. CA exhibits a sigmoidal inhibitory action, whereas AMPC, EPMC, and diclofenac sodium follow linear responses.

IC₅₀ values (Fig. 2) were: CA, 171.48 μg/mL; EPMC, 170.02 μg/mL; diclofenac sodium, 165.05 μg/mL; and APMC, 215.06 μg/mL. Statistical analysis indicated no significant differences between diclofenac sodium and either CA (p > 0.05) or EPMC (p > 0.05), whereas APMC’s IC₅₀ was significantly higher than both CA and EPMC (p < 0.05), demonstrating that CA and EPMC possess anti-inflammatory potency comparable to diclofenac sodium, while APMC exhibits weaker activity.

Figure 2.

IC50 values of diclofenac sodium, CA, EPMC, and APMC in the BSA denaturation assay. One-way ANOVA followed by Sidak’s post hoc shows that APMC IC50 is significantly higher than CA with EPMC, whereas no significant differences are observed between CA, EPMC, and diclofenac sodium.

3. In vivo carrageenan-induced inflammation assay

No mice were excluded during the experiment. Following carrageenan injection, the negative control group exhibited a rapid increase in paw thickness, peaking at approximately 3.0 mm at 1 hour (p < 0.05 vs. normal control) and remaining elevated throughout the 4-hour observation period (Fig. 3).

Figure 3.

Time-course of right hind paw thickness following carrageenan injection in mice treated with different treatment interventions. The negative control exhibits the highest paw thickness at 1, 2, and 3 hours post-injection. treatment groups (CA 150 and 600 mg/kg BW, APMC 150 and 600 mg/kg BW, EPMC 150 and 600 mg/kg BW), and diclofenac sodium 100 mg/kg BW attenuate edema, with EPMC showing the lowest paw thickness at all time points.

Diclofenac sodium (100 mg/kg BW) attenuated the early peak (approximately 2.5 mm at 1 hour) with sustained but incomplete resolution. CA at both doses produced moderate suppression of paw edema. APMC at 150 and 600 mg/kg BW produced more pronounced effects, with paw thickness approaching baseline by 4 hours (approximately 1.9-2.0 mm).

EPMC at both doses demonstrated the most rapid and pronounced anti-inflammatory effects, with early peaks virtually absent (approximately 2.0-2.1 mm at 1 hour) and paw thickness below baseline by 4 hours (approximately 1.8 mm). No substantial differences were observed between the 150 and 600 mg/kg BW doses for any compound, suggesting a plateau in the dose-response relationship.

AUC analysis (Fig. 4) confirmed sustained inflammation in the negative control group (p < 0.05 vs. normal control). All treatment groups except CA 600 mg/kg BW, exhibited significantly lower AUC than the negative control (p < 0.05). Notably, EPMC at both doses produced significantly lower AUC than diclofenac sodium (p < 0.05), indicating superior anti-inflammatory efficacy. EPMC 150 mg/kg BW was significantly lower than both APMC 150 mg/kg BW and CA 150 mg/kg BW (p < 0.05), and EPMC 600 mg/kg BW was significantly lower than CA 600 mg/kg BW (p < 0.05).

Figure 4.

AUC of paw thickness over 4 hours after carrageenan induction. One-way ANOVA followed by Sidak’s post hoc test shows a significant increase in AUC in the negative control compared with the normal control (p < 0.05). Treatment with CA (150 mg/kg BW), AMPC (150 mg/kg BW and APMC 600 mg/kg BW), EPMC (150 mg/kg BW and EPMC 600 mg/kg BW), and diclofenac sodium (100 mg/kg BW) significantly reduces paw thickness AUC (p < 0.05). EPMC at 150 mg/kg BW and 600 mg/kg BW significantly reduced paw thickness AUC compared with diclofenac sodium 100 mg/kg BW (p < 0.05). EPMC at 150 mg/kg BW also significantly reduced paw thickness AUC compared with CA or APMC at the same dose (p < 0.05), and EPMC at 600 mg/kg BW significantly reduced paw thickness AUC compared with CA at the same dose (p < 0.05). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

4. Hematological analysis

Blood samples collected at 3 hours post-induction were analyzed for hematological parameters (Fig. 5). Hb levels in the negative control were significantly higher than in the normal control (p < 0.05), potentially indicating hemoconcentration. No treatment groups significantly altered Hb levels (p > 0.05).

Figure 5.

Hematological parameters are measured from the lateral vein of mice 3 hours after carrageenan induction into the right hind paw of mice under different treatment conditions. Hb levels are significantly elevated in the negative control group compared with the normal group. Non-significant trends of increased WBC, lymphocyte, platelet, and hematocrit values are observed in the negative control. Treatment with APMC 600 mg/kg BW tends to normalize WBC and lymphocyte counts toward normal levels, although differences are not statistically significant. *p < 0.05.

WBC counts in the negative control showed a non-significant upward trend compared with the normal control. APMC 600 mg/kg BW tended to reduce WBC counts toward normal levels, although this did not reach statistical significance (p > 0.05). Diclofenac sodium, CA, and EPMC did not substantially alter WBC counts. Lymphocyte counts and percentages showed similar trends, with APMC at 600 mg/kg BW tending to normalize values, although this did not reach statistical significance (p > 0.05).

Platelet-related parameters showed non-significant elevation trends in the negative control, with no significant treatment effects. Overall, although the negative control exhibited trends consistent with systemic inflammation (elevated Hb, WBC, lymphocytes, platelets), only Hb reached significance. APMC 600 mg/kg BW showed the most consistent trends toward normalization of WBC and lymphocyte parameters, suggesting potential systemic immunomodulation. In contrast, CA and EPMC, despite superior anti-inflammatory activity as measured in paw edema, showed minimal hematological effects, indicating predominantly localized anti-inflammatory action.

DISCUSSION

This integrated study demonstrates that EPMC exhibits superior multi-target anti-inflammatory activity through mechanisms distinct from structurally related CA derivatives. Molecular docking revealed EPMC’s consistent binding across multiple inflammatory targets, including HSP90AA1, JAK2, PTGS2/COX-2, lipoxygenase, and HSP90AB1, with particularly notable interaction with PTGS2/COX-2 (–6.29 kcal/mol) involving the critical tyrosine 385 (Tyr385) residue shared with mefenamic acid. This multi-target profile aligns with contemporary polypharmacological strategies, in which simultaneous modulation of multiple pathways can enhance therapeutic efficacy and reduce resistance compared with single-target approaches [12, 13].

Natural products often possess privileged scaffolds evolved through biosynthetic pathways, which confer inherent multi-targeting capabilities valuable for complex diseases such as inflammation [28, 29]. Notably, CA uniquely bound NOS3 (–3.06 kcal/mol), whereas EPMC and APMC displayed unfavorable interactions, suggesting that CA modulates inflammation via nitric oxide pathways. This observation is consistent with report demonstrating that CA reduces serum nitric oxide (NO) levels in rats with gentamicin-induced liver and kidney dysfunction [15, 30]. The structure-dependent target selectivity, in which CA contains a free carboxylic acid while EPMC and APMC are esterified, highlights mechanistic pathways among these derivatives.

The BSA denaturation assay confirmed that CA and EPMC exhibit anti-inflammatory potency comparable to diclofenac sodium (IC₅₀: 171.48, 170.02, and 165.05 μg/mL, respectively), whereas APMC displayed significantly weaker activity (215.06 μg/mL). The concentration-response patterns provide mechanistic insights: CA’s sigmoidal curve suggests receptor saturation kinetics, whereas EPMC, APMC, and diclofenac sodium exhibited linear responses, indicating concentrations below saturation thresholds. EPMC’s superior activity compared with CA is attributable to its enhanced lipophilicity, achieved through ethyl esterification, which increases membrane permeability and cellular uptake [19, 20]. In contrast, CA’s hydrophilic carboxyl group limits membrane penetration and bioavailability [18], whereas EPMC’s ester modification facilitates target engagement [31]. This structure-activity principle—where lipophilicity correlates positively with anti-inflammatory efficacy—is well documented for CA derivatives [21, 32]. APMC’s weaker activity suggests that retention of the carboxylic acid restricts efficacy, even though it contains a para-methoxy group that enhances ROS stabilization [33], underscoring that both methoxy substitution and esterification are required for maximal activity.

In vivo validation using the carrageenan-induced inflammation model confirmed EPMC’s therapeutic superiority. EPMC at both doses (150 and 600 mg/kg BW) produced rapid-onset, complete suppression of inflammation, significantly outperforming diclofenac sodium (p < 0.05 for AUC comparison). The absence of an early inflammatory peak and below-baseline paw thickness at 4 hours indicates immediate and potent anti-inflammatory action with full edema resolution. This below-baseline phenomenon may reflect enhanced resolution mechanisms, including vasoconstriction reducing tissue fluid content or enhanced lymphatic drainage, consistent with reports that NSAIDs can modify vascular reactivity [34]. The dose-response plateau, where 150 and 600 mg/kg BW produced comparable effects, suggests that the lower dose approaches maximum efficacy. For EPMC and APMC, this likely reflects near-complete suppression of inflammation at 150 mg/kg BW, whereas for CA, potential explanations include receptor saturation (consistent with sigmoidal BSA patterns), limited bioavailability at higher doses, or compensatory metabolic activation. Nonlinear dose-response relationships have been reported for CA derivatives [33], supporting the concept of an optimal therapeutic window and highlighting the need for pharmacokinetic studies evaluating lower dose ranges (25-100 mg/kg BW).

Hematological analyses revealed mechanistic divergence among the compounds. APMC at 600 mg/kg BW consistently normalized WBC and lymphocyte counts toward values observed in normal controls, suggesting systemic immunomodulatory effects, potentially through inhibition of chemokine-driven leukocyte mobilization. Carrageenan induces chemotactic mediators, including interleukin-8 (IL-8), activated monocyte chemoattractant factor II (AMCF-II), and monocyte chemoattractant protein-1 (MCP-1), promoting leukocyte mobilization [35], and APMC’s attenuation of leukocytosis may involve inhibition of toll-like receptor 4 (TLR4)/myeloid differentiation primary response 88 (MyD88) pathways, mechanisms reported for structurally related cinnamic aldehyde [36]. This systemic immunomodulatory profile resembles the action of diclofenac sodium. In contrast, CA and EPMC exhibited minimal hematological effects despite superior anti-inflammatory activity in paw measurements, indicating predominantly localized tissue-level action via direct COX inhibition (EPMC), NOS3 modulation (CA), and membrane stabilization, without substantial systemic leukocyte modulation [6, 37]. This functional divergence has clinically relevant therapeutic implications. Conventional NSAIDs are associated with substantial risks, including 2-4% annualized gastrointestinal complication events, 20-30% prevalence of ulcers in chronic users, and cardiovascular risks of 7-9 non-fatal and 2 fatal events per 1000 high-risk patients annually [8-10, 38, 39]. EPMC’s localized action without systemic hematological modulation may reduce the risk of infection and bone marrow suppression compared with systemically acting agents [40], potentially benefiting patients requiring long-term therapy. Conversely, APMC’s systemic immunomodulatory profile suggests its suitability for conditions that require immune regulation, although careful monitoring for potential immunosuppressive effects may be necessary.

Limitations of this study include a small sample size (n = 3 per group), the absence of blinding, a single blood collection time point, and a short observation period. Future research should prioritize: (i) comprehensive pharmacokinetic studies to optimize dosing regimens; (ii) mechanistic studies, including COX isoform selectivity and cytokine modulation; (iii) chronic inflammation models to assess long-term safety; (iv) gastrointestinal and cardiovascular safety profiling versus NSAIDs; (v) dose-response studies at lower doses; and (vi) larger, blinded trials to confirm hematological trends. Collectively, these findings position EPMC as a promising lead for next-generation anti-inflammatory agents with improved safety profiles, warranting progression toward clinical development.

CONCLUSION

EPMC demonstrates superior multi-modal anti-inflammatory activity through engagement of multiple inflammatory targets, potent in vitro efficacy comparable to diclofenac sodium, and rapid, complete in vivo suppression of inflammation, surpassing conventional NSAIDs. The divergent profiles of the derivatives—APMC showing systemic immunomodulation and EPMC acting locally—underscore the potential for tailored therapeutics based on inflammatory phenotypes. EPMC’s multi-target nature aligns with polypharmacological strategies and may offer advantages over single-target NSAIDs, while its localized action may reduce systemic toxicities.