Anti-inflammatory potential of liraglutide, a glucagon-like peptide-1 receptor agonist, in rats with peripheral acute inflammation

The present study aimed to explore the possible anti-inflammatory actions of liraglutide (LRG), a glucagon-like peptide-1 (GLP-1) receptor agonist, and to compare with tramadol (TR) or LRG, and TR combination treatment by investigating the inflammatory signs such as pain hypersensitivity, edema, and fever in carrageenan (CG)-induced acute peripheral inflammation model in rats. The levels of several biomarkers for inflammatory status, angiogenesis, and oxidative stress were also measured in inflamed tissues. CG induced inflammation in the paws of rats identified by hypersensitivities, redness, edema and fever. LRG, significantly improved the hypersensitivity to mechanical (from 4 to 35.5 g) or cold (from 5 to 44.2 s) stimuli, reduced the edema (paw mass, from 2.54 to 1.85 g), and fever (paw temperature, from 33.6 to 27.3 °C). LRG dramatically suppressed the inflammatory signs when compared to those of TR. In addition, co-administration of TR and LRG resulted in further reduction of sensitivity to mechanical and cold stimuli. Anti-inflammatory potential of LRG altered depending on their inhibitory effects in the biomarkers of inflamed paws. Consequently, the suppressive actions of LRG in the inflammation induced hypersensitivities, edema, and fever, indicating that these drugs have significant anti-inflammatory potential with anti-hypersensitivities, anti-edema, and anti-pyretic effects. LRG with anti-inflammatory actions may be a highly promising therapeutic option for the management of inflammatory conditions or inflammatory-related various diseases.


Introduction
While inflammation is defined as a defensive mechanism of the body against any harmful factor, acute inflammation is the first response involving a series of vascular and cellular events (Libby 2007;Zhang and An 2007). The degree and characteristics of the inflammatory responses depend on the nature and severity of the damaging effect. Carrageenan (CG)-induced local acute inflammation in the paws of rats is a well-known experimental model to investigate local inflammatory processes and the anti-inflammatory effects of drug treatments (Annamalai and Thangam 2017). This model is characterized by inflammatory signs, such as swelling, heat, redness, and pain behavior (Barrot 2012;Annamalai and Thangam 2017).
Inflammatory responses involve complex processes mediated by inflammatory mediators produced by various immune cells, such as cytokines, chemokines, and growth factors (Dawes and McMahon 2013;Annamalai and Thangam 2017). The inflammatory mediators including proinflammatory cytokines and reactive oxygen species released by various inflammatory cells, such as macrophages and neutrophils, are responsible for the inflammatory process (Conner and Grisham 1996;Halici et al. 2007) Moreover, the balance between pro-inflammatory cytokines, TNF-α, and anti-inflammatory cytokines, IL-10, as well as the balance between pro-oxidant and the anti-oxidant defense system, plays a decisive role in the development of the acute inflammatory process (Conner and Grisham 1996;Kidd 1 3 and Urban 2001;Ferrara 2002;Halici et al. 2007). Also, angiogenesis and inflammation are processes that are closely related to each other. Several potent angiogenic factors, such as VEGF and TGF-β, produced by inflammatory cells play key roles in the development of inflammatory responses (Sporn and Roberts 1992).
Steroid and non-steroidal anti-inflammatory drugs are frequently preferred in the management of inflammatory diseases (Rao and Knaus 2008). However, in recent years, a number of experimental and clinical studies have reported the beneficial effects of many different pharmacological treatments in the management of inflammatory conditions. Previous papers conducted with the inflammation model have suggested that inflammatory conditions can cause an increase in opioid receptor formation and their transport to peripheral endings and μ opioid receptor agonists can cause more effective peripheral analgesia than agonists of the other opioid receptors (Raffa et al. 1992;Mert et al. 2007;Stein and Zöllner 2009;Reeves and Burke 2008). Tramadol (TR) hydrochloride (1RS, 2RS)-2-((dimethylamino)methyl)-1-(3methoxyphenyl)-cyclohexonal HCl) is a synthetic μ opioid receptor agonist (Raffa et al. 1992). Experimental and clinical studies to investigate the effectiveness of TR show that TR has anti-nociceptive and analgesic effects with fewer side effects than the other opioids. It can exert its effects through both μ-receptor-mediated and non-receptor-mediated mechanisms (Raffa et al. 1992;Mert et al. 2007; Reeves and Burke 2008).
Several papers on the clinical treatments of glucagonlike peptide 1 (GLP-1) receptor agonists for diabetes mellitus have indicated that GLP-1 receptors have a widespread distribution and that the action mechanism of GLP-1 analogs is also closely related to various pathophysiological processes, such as inflammation, in addition to regulating blood glucose levels (Cho et al. 2013). GLP-1 analogs, such as liraglutide (LRG), are clinically used in the management of diabetic patients due to their insulinotropic properties (Cho et al. 2013;Calsolaro and Edison 2015). LRG is a polypeptide GLP-1 analog drug produced by recombinant DNA technology. While the half-life of natural GLP-1 is approximately 1-2 min, LRG, which binds more tightly to albumin due to its molecular feature and thus is more resistant to dipeptidyl peptidase-4 activity, has a longer half-life and reaches approximately 12-13 h (Hölscher 2012;Meier 2012;Tang et al. 2016).
A growing body of preclinical evidence of anti-hypersensitive and anti-nociceptive activities of glucagon-like peptides indicates that, GLP-1 agonists, such as exenatide, can produce anti-nociception in various animal models of pain hypersensitivity states, including peripheral and diabetic neuropathic pain. It has also been reported that the anti-nociceptive effect of GLP-1 agonism can be inhibited by the opioid receptor antagonist naloxone, and suggested that GLP-1 receptor agonists can reduce drug supplementation (Gong et al. 2014a, b;Aykan et al. 2019;Zhang et al. 2020).
Although several studies have displayed that LRG may have systemic anti-inflammatory potential in addition to glucose-lowering effects, to our knowledge there is no study showing the possible actions of LRG on local inflammatory processes and inflammatory responses. Furthermore, the possible efficacy of LRG in reducing opioidmediated behaviors has not been fully explored. Therefore, in the presented study, it was hypothesized that LRG treatment may produce remarkable anti-inflammatory actions due to the widespread distribution of GLP-1 receptors and LRG may potentiate the efficacy of TR, a synthetic opioid, in acute inflammatory process. To test this hypothesis, the possible effects of alone and combined applications of LRG and TR (LRG + TR) on inflammatory signs in a local acute inflammation model in rats were explored by examining the inflammatory induced responses (pain, edema, and heat) and analyzing various biomarkers of paw tissue for inflammatory status, oxidative stress, and angiogenesis.

Animals
Adult Wistar albino rats (240-250 g, provided by Bolu Abant İzzet Baysal University experimental animals application and research center) were used in the experiments. Animals were housed under standard conditions with ad libitum access to food and water under a standard 12 h light/dark regime at room temperature (23 ± 2 °C). Bolu Abant Izzet Baysal University animal experiments local ethical committee, Turkey, approved the animal procedures (number: 2020/30). All experiments were made according to the guidelines of International Association for the Study of Pain.

Carrageenan-induced acute peripheral inflammation
Acute peripheral inflammation by λ-carrageenan (lambda CG, Sigma, Germany; 100 μL of 2% (wt/vol) is highly sensitive to investigate the inflammatory signs. Following the baseline behavioral measurements, CG was intraplantarly (I.pl.) injected (1-mL syringe and a 27-gauge needle) into the right hindpaw at the midline of the heel and anteriorly at the base of the second or third toe. An equal volume of saline (vehicle) was injected (I.pl) into the other hindpaw of rats for the control studies.

Drugs and experimental groups
In this present study, considering the previous studies using similar experimental models, and after power analysis (α = 0.05, power ≥ 0.80), 6 (n) animal studies were performed for each experimental group.
TR (Sigma-Aldrich Chemie GmbH, Munich, Germany) or LRG (Novo Nordisk A/S, Bagsvaerd, Denmark) was administered to rats in 200 µl saline solution at 1 mg/kg intraperitoneally (IP) 1 h after CG injection. The dose of 1 mg/kg for both drugs was chosen based on our pilot and previous studies, together with published literature demonstrating the acute effects of these drugs. In TR + LRG combination treatments, 1 mg/kg (in 200 µl saline) dose of both drugs was applied together.
Possible anti-inflammatory effects of systemically (IP) treated TR, LRG, or their combination were investigated comparatively in localized inflamed paw (I.pl. CG-injected right paw) and non-inflamed paw (I.pl. saline-injected left paw) of rats in this present study. In the experiments, recordings from the untreated paw (non-inflamed left paw) in rats were used to compare the effect of drugs. The data to be obtained before drug administration were evaluated as control data to determine drug efficacy. Each experimental group (n = 6)contained two rat cages.

Sensory testing procedure
Experiments were carried out at 23-24 °C room temperature, in a quiet environment, between 9 and 16 h. Before the study, all rats were habituated to the experimental environment and setup by placing the rats on the test systems described below for at least 30 min on the day of the experiment and 3 days before the experiments. Also, because the experimental procedures involved animal handling, the rats were handled by the experimenter for 3 consecutive days before testing before the experiments.
The experimenter was blinded to the treatments of the rats until the experiments were completed. During the experiments, every effort was made to minimize the number of rats used and to alleviate their suffering, which were carried out with due regard for ethical considerations. In this study, the effects of drug administration were examined 1 h before, 1 h after CG injection, and 5 h after CG injection to limit the exposure time of rats and observe optimum inflammationinduced responses.

Von Frey test
The development of mechanical or tactile hypersensitivity was evaluated with testing of mechanical withdrawal thresholds by applying calibrated Von Frey filaments (4-60 g) (Stoelting, Wood Dale, IL, USA) (Chaplan et al. 1994).
Rats were placed on a metal mesh floor in transparent boxes separated by individual opaque separators. Filaments were applied vertically for 2 s until they were bent semicircularly to the plantar surface of both hind paws. If no response was received, a larger diameter filament was likewise applied. The filaments were applied to the mid plantar area five times in ascending order until paw withdrawal, which was considered a positive response. The filament with which the paw withdrew was recorded as the threshold value. This withdrawal threshold was determined twice at 2-min intervals and was used as the mean withdrawal threshold for data analysis. During the testing, the rat was considered to have an intense mechanical nociceptive sensory blockade when no response was observed using the filament with the largest diameter. Even with the lowest stimulus, it was considered the development of a sign of tactile allodynia if the rats withdrew their paws.

Cold plantar test
Cold plantar test was performed to evaluate the development of cold hypersensitivity in rats (Brenner et al. 2012). Rats were placed in clear boxes separated by opaque separators on a 3 mm-thick glass plate. The cold application probe was made from a modified 5 mL syringe filled with freshly powdered dry ice. After the dry ice was pelleted and the surface was flattened, the cold probe was applied to the hindpaw of the rat from under the glass. The cold probe induced a cooling ramp in the rat's paw in the temperature range of 5-10 °C and elicited paw withdrawal responses. The withdrawal latency was measured with a chronometer.
Withdrawal latency for each rat was taken as the average of three trials at intervals of at least 2 min. Paw withdrawal latency is defined as the time it takes for rats to shake or withdraw their paw from the cold probe.

Paw temperature measurement
Thermal changes in the paws are used to determine the antiinflammatory activities of the inflammatory process and applications, especially in the inflammation model. Temperature changes were evaluated by measuring the temperature of the inflamed and non-inflamed paws with an infrared thermometer (2 cm from the middle surface of the paw) at the end of the experiments. Data were obtained on the local anti-pyretic effects of the treatments.

Evaluation of oedema activity
Intense oedema occurs after CG injection to the paws of rats. Changes in paw masses were used to determine the possible anti-edema activities of the treatments. Paw masses of rats were measured at the end of the experiments. After killing, paw edema was determined by measuring the mass (g) of the paws cut from the ankle joint.

Determination of the biomarkers in paw tissue
At the end of experiments, animals were killed under anesthesia and paw tissue samples were quickly removed with hygienic vehicles on ice and kept at − 80 °C until analysis. On the test day, the tissues were homogenized with phosphate buffer at a ratio of 1/9 (0.1 g tissue: 0.9 ml 50 mmol phosphate buffer pH: 7.40) and then centrifuged at 7000 rpm for 5 min. Supernatants formed after centrifugation weres studied using commercial kits.

Measurements of total anti-oxidant status and total oxidant status
The measurements of total anti-oxidant status (TAS) and total oxidant status (TOS) levels provide the accumulative actions of all antioxidants and oxidants existing in the inflamed tissues. TAS and TOS levels were measured using commercially available kits (Relassay, Turkey). The automated method for TAS level is based on the bleaching of characteristic color of a more stable 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonic acid (ABTS) radical cation by antioxidants. The reaction rate is calibrated with Trolox, which is widely used as a traditional standard for TAC measurement assays. The results were expressed as mmol Trolox equivalent/L (Erel 2004).
For measurement of TOS levels, oxidants present in the sample oxidized the ferrous ion-o-dianisidine complex to ferric ion. The oxidation reaction was enhanced by glycerol molecules abundantly present in the reaction medium. The ferric ion produced a colored complex with xylenol orange in an acidic medium. The color intensity, which could be measured spectrophotometrically, was related to the total amount of oxidant molecules present in the sample. The assay was calibrated with hydrogen peroxide and the results were expressed in terms of micromolar hydrogen peroxide equivalent per liter (μmol H 2 O 2 equivalent/L) (Erel 2005).

Measurements of VEGF-A and TGF-β1
Sandwich enzyme-linked immunosorbent assay (ELISA)based kits were used for the measurements of VEGF-A or TGF-β1 (Elabscience Biotechnology Co. Wuhan, China). Standards or samples are added to the micro ELISA plate wells and combined with the specific antibody. Then, a biotinylated detection antibody specific for universal VEGF-A or TGF-β1 and avidin-horseradish peroxidase (HRP) conjugate are added successively to each micro plate well and incubated. Free components are washed away. The substrate solution is added to each well. Only those wells that contain universal VEGF-A or TGF-β1, biotinylated detection antibody and avidin-HRP conjugate will appear blue in color. The enzyme-substrate reaction is terminated by the addition of the stop solution and the color turns yellow. The optical density (OD) is measured spectrophotometrically at a wavelength of 450 nm ± 2 nm. The OD value is proportional to the concentration of universal VEGF-A or TGF-β1.

Measurements of TNF-α and IL-10
Sandwich enzyme-linked immunosorbent assay (ELISA)based kits were used for the measurements of TNF-α and IL-10 (Bioassay Technology Laboratory (BT Lab), Inc., Shanghai, China). TNF-α or IL-10 present in the sample is added and binds to antibodies coated on the wells. Then biotinylated rat anti-TNF-α or anti-IL-10 antibody is added and binds to TNF-α or IL-10 in the sample. Then streptavidin-HRP is added and binds to the biotinylated TNF-α or IL-10 antibody. After incubation, unbound streptavidin-HRP is washed away during a washing step. The substrate solution is then added and color develops in proportion to the amount of rat TNF-α or IL-10. The reaction is terminated by the addition of acidic stop solution and absorbance is measured at 450 nm.

Statistical analyses
Findings were presented as mean ± standard deviation (SD). All statistical analyses were performed using analysis of variance (ANOVA, Statistical Package for Social Sciences 15.0, SPSS, Chicago, IL, USA). Data were tested for normal distribution using the Kolmogorov-Smirnov test. Thresholds or latencies (dependent variables) measured during the pretreatment and all of the posttreatments were analyzed by using two-way, repeated-measures ANOVA (time × treatment) followed by Bonferroni's post hoc test. To evaluate the effectiveness of treatments, difference between groups was tested using a one-way ANOVA (treatments as independent variables), followed by post hoc Bonferroni test. A value of P < 0.05 was considered as significant.

Results
While experiments in the inflamed right paw were conducted to identify possible anti-inflammatory actions of drugs, findings from non-inflamed left paw were used to identify possible changes in the sensor sensitivity.
It was determined that the basal cold latency was 21.0 ± 3.5 s, and the basal mechanical threshold value was 26.0 ± 0.5 g before the CG injection ( Figs. 1 and 2). I.pl. injection of saline for CG control did not change the basal threshold or latency values.

Changes in mechanical thresholds
Responses to mechanical stimuli from the non-inflamed (left) paws were used to evaluate the effects of the treatments on the tactile sensation of rats (Fig. 1A). In noninflamed rats, saline administration (control for drug) did not cause a significant change in threshold, while treatments of drugs significantly decreased the tactile sensitivity 4 h after administration (P < 0.05). LRG increased the threshold more than that of TR (P < 0.05). The TR + LRG combination increased the threshold even more in noninflamed rats (P < 0.05) (Fig. 1A).
CG significantly decreased the threshold value (Fig. 1B). CG-induced reduction in the threshold indicates the development of mechanical allodynia in rats. The mechanical allodynic response, which appeared 1 h after CG injection, reached its highest level after 5 h (P < 0.05). TR, LRG, or TR + LRG treatments caused significant enhancements in the decreased threshold by CG (P < 0.05). While LRG increased the threshold more than TR, the Fig. 1 Changes in the mechanical thresholds of the noninflamed (A) and inflamed (B) paws of rats. Tramadol (TR), liraglutide (LRG), or their combination (TR + LRG) was used to treat rats 1 h after carrageenan (CG) or saline (Sal) injections during the experiments. Each point represents the mean value of six rats, and the vertical bars indicate ± SD. Statistical evaluation was performed by repeated-measures ANOVA with Bonferroni's post hoc test. + P < 0.05 indicates significant differences as compared to baseline value. *P < 0.05 indicates significant differences as compared to Sal-treated group. **P < 0.05 indicates significant differences as compared to the TR-treated group. ***P < 0.05 indicates significant differences as compared to LRG-treated group increases in the threshold values reached the highest level in the TR + LRG combined treatment (P < 0.05) (Fig. 1B).

Changes in cold latencies
Responses to cold stimuli from the non-inflamed (left) paw were used to evaluate the effects of the treatments on cold sensation of rats ( Fig. 2A). In non-inflamed rats, saline administration (IP) did not change the latency, while all treatments significantly decreased the cold sensitivity 4 h after administration. LRG caused a greater increase in latency compared to TR, while the combination of TR + LRG increased latency even more in non-inflamed rats (P < 0.05) ( Fig. 2A).
Paw withdrawal latency to cold stimulus significantly reduced after CG injection. This decrease indicates the development of cold allodynia in inflamed rats. CG-induced cold allodynia peaked after 5 h (Fig. 2B).
TR, LRG, or TR + LRG treatments caused significant increases in the decreased latencies by CG (P < 0.05). LRG increased the withdrawal latency more than TR (P < 0.05). However, when LRG and TR + LRG administrations were Fig. 2 Changes in cold latencies of the non-inflamed (A) and inflamed (B) paws of rats. Tramadol (TR), liraglutide (LRG), or their combination (TR + LRG) was used to treat rats 1 h after carrageenan (CG) or saline (Sal) injections during the experiments. Each point represents the mean value of six rats, and the vertical bars indicate ± SD. Statistical evaluation was performed by repeated-measures ANOVA with Bonferroni's post hoc test. + P < 0.05 indicates significant differences as compared to the baseline value. *P < 0.05 indicates significant differences as compared to the Sal-treated group. **P < 0.05 indicates significant differences as compared to the TR-treated group. ***P < 0.05 indicates significant differences as compared to the LRG-treated group compared, it was found that the latency values did not statistically differ in inflamed rats (P > 0.05) (Fig. 2B).

Alterations in paw temperatures of inflamed rats
In the study, paw temperatures were measured with an infrared thermometer 5 h after the CG injection (4 h after the drug administration), to determine the possible anti-pyretic effects of the drugs.
Although TR decreased the increase in paw temperature due to the inflammation, the change was not statistically significant (P > 0.05). While the LRG and TR + LRG combination significantly suppressed the increase in paw temperature, there was no significant difference between the results of the two treatments (Fig. 3A). These results are important in terms of showing the anti-pyretic activities of LRG and TR + LRG combination, despite the partial anti-pyretic activity of TR. However, TR, LRG and TR + LRG treatments did not show any significant changes in the temperature of non-inflamed paws.

Alterations in the paw mass of inflamed rats
To evaluate the anti-edema effects of the drugs, the paw masses were measured 5 h after the CG injection (4 h after the drug administration).
Saline injection did not significantly change the paw mass compared to the paw mass of intact rat (1.50 ± 0.07 g) (Fig. 3B). When compared to the non-inflamed contralateral paw mass of saline-injected rats (1.47 ± 0.13 g), CGtreated paw mass significantly increased (2.54 ± 0.31 g) (P < 0.05) (Fig. 3B). TR, LRG, or TR + LRG combination treatments significantly suppressed the increase in paw mass (P < 0.05). These results are important in terms of demonstrating the anti-edema efficacy of TR, LRG, and TR + LRG combination. When compared, in the antiedema actions of all drug treatments, there were no statistically significant differences. However, TR, LRG and TR + LRG treatments did not show any significant changes in temperature of non-inflamed contralateral paws. Each bar represents the mean value of six rats, and the vertical bars indicate ± SD. Statistical evaluation was performed by one-way ANOVA followed by Bonferroni test. + P < 0.05 indicates significant differences as compared to the Sal-treated group. *P < 0.05 indicates significant differences as compared to the CG + Sal-treated group. **P < 0.05 indicates significant differences as compared to the CG + TR group

Changes in the paw tissue levels of biomarkers in inflamed rats
The TAS level did not significantly change in the inflamed paw tissue. While the TR application did also not cause any significant change in the TAS level, LRG application caused a significant increase in TAS level of inflamed paw (P < 0.05) (Fig. 4A). TR + LRG combination application caused a statistically insignificant increase in the TAS level (P < 0.05). The level of TOS in the paw tissue significantly increased after the CG-induced inflammation (P < 0.05) (Fig. 4B). However, TR, LRG, or TR + LRG combination treatments did not show statistically significant changes in the increased TOS level of inflamed paw tissue.
TNF-α levels significantly enhanced in the inflamed paw tissue (P < 0.05) (Fig. 5A). While TR significantly suppressed this increase in the TNF-α level (P < 0.05), LRG or TR + LRG combination did not cause any significant change. The level of IL-10 significantly increased in inflamed paw tissue when compared to the non-inflamed paw (P < 0.05) (Fig. 5B). The treatment of TR did not change the increased level of IL-10 in inflamed paw tissue, while LRG or TR + LRG combination significantly increased the IL-10 level (P < 0.05).
The level of VEGF significantly increased in the inflamed paw tissue of rats compared to the non-inflamed paw (P < 0.05). Although not statistically significant, TR showed a tendency to suppress the VEGF level in inflamed tissue. The treatment of LRG or the combination of TR and LRG resulted in statistically significant reductions in tissue VEGF level (P < 0.05) (Fig. 6A). TGF-β1 levels also significantly increased in the inflamed paw tissue of rats (P < 0.05) (Fig. 6B). While TR did not cause a significant change in TGF-β1 level, LRG or TR + LRG combination significantly suppressed TGF-β1 level in inflamed paw tissue (P < 0.05).
All treatments did not cause any significant changes in the levels of biomarkers mentioned above in non-inflamed contralateral paws.

Discussion
The findings showed the anti-hypersensitive efficacies of systemically administered TR, LRG, or TR + LRG combination in the increased mechanical or cold sensitivity caused by CG injection in the rat paw. The findings also suggested that notable reductions in tactile and cold sensitivities by LRG compared to TR in the non-inflamed paw may indicate their analgesic and/or anaesthetic potential. In addition, LRG treatment suppressed the CG-induced rat paw edema and paw temperature increase, indicating that it has significant anti-inflammatory potential, together with their anti-edema and anti-pyretic effects. To the best of our knowledge, this is the first report showing the anti-hypersensitive, anti-edema, Each bar represents the mean value of six rats, and the vertical bars indicate ± SD. Statistical evaluation was performed by one-way ANOVA followed by Bonferroni test. + P < 0.05 indicates significant differences as compared to the Sal-treated group. *P < 0.05 indicates significant differences as compared to the CG + Sal-treated group. **P < 0.05 indicates significant differences as compared to the CG + TR group and anti-pyretic actions of LRG in rats with acute peripheral inflammation.
In this present study, development of edema, redness, and inflammatory pain behaviors, which are indicators of inflammatory reaction, were observed within the first hour following CG injection. Hypersensitive responses to mechanical or cold stimuli peaked 5 h after CG injection.
In addition, remarkable increases in edema and temperature were detected in the inflamed paws. Moreover, responses to mechanical and cold stimuli from the noninflamed paw were used to evaluate the effects of the treatments on the tactile and cold sensitivities of rats, and to determine possible analgesic/anesthetic efficacies. Effects of tramadol (TR), liraglutide (LRG), or their combination (TR + LRG) on tumor necrosis factor alpha (TNF-α) (A) and interleukin (IL)-10 (B) in carrageenan (CG)-induced inflamed paws of rats. Each bar represents the mean value of six rats, and the vertical bars indicate ± SD. Statistical evaluation was performed by one-way ANOVA followed by Bonferroni test. + P < 0.05 indicates significant differences as compared to the Sal-treated group. *P < 0.05 indicates significant differences as compared to the CG + Sal-treated group. **P < 0.05 indicates significant differences as compared to the CG + TR group Fig. 6 Effects of tramadol (TR), liraglutide (LRG), or their combination (TR + LRG) on vascular endothelial growth factor A (VEGF-A) (A) and transforming growth factor β1 (B) in carrageenan (CG)induced inflamed paws of rats. Each bar represents the mean value of six rats, and the vertical bars indicate ± SD. Statistical evaluation was performed by one-way ANOVA followed by Bonferroni test. + P < 0.05 indicates significant differences as compared to the Sal-treated group. *P < 0.05 indicates significant differences as compared to the CG + Sal-treated group. **P < 0.05 indicates significant differences as compared to the CG + TR group Inflammatory pain hypersensitivity is the cardinal sign of acute peripheral inflammation that occurs depending on the process (Kidd and Urban 2001;Coutaux et al. 2005). Consistent with previous works, decreases in mechanical threshold and cold latency values are indications of the development of mechanical or cold hypersensitivities (Coutaux et al. 2005;Lippoldt et al. 2016). Considering the contribution of sensory afferents to hypersensitivity, peripheral inflammation may cause an increase in the sensitivity of peripheral nerve terminals of A-delta and C nerve fibers in the inflammation site (Libby 2007;Basbaum et al. 2009;Barrot 2012).
The effects of TR and LRG on latency and threshold parameters were different from each other. TR enhanced the threshold and latency in inflamed rats, while mechanical and cold sensitivity in the non-inflamed paw was reduced by TR treatment. Findings, as expected, showed that TR may have anti-hypersensitive effects. These results are consistent with the data obtained from previous studies (Raffa et al. 1992;Mert et al. 2007). It has been previously reported that TR may produce analgesic/anaesthetic activities due to its high affinity for μ-opioid receptors as well as inhibition of both serotonin and norepinephrine reuptake (Raffa et al. 1992;Reeves and Burke 2008;Stein and Zöllner 2009). There is also lots of evidence that TR inhibits the activity of voltage-gated Na channels, delayed rectifier K channels, N-methyl-d-aspartate receptors, and substance P receptors in vitro (Hara et al. 2005;Mert et al. 2006Mert et al. , 2007Reeves and Burke 2008).
Compared to TR, LRG treatment prolonged cold latency and raised mechanical threshold values of rats with paw inflammation, suggesting potent anti-hypersensitive effect. The present study reports for the first time the anti-hypersensitive potential of LRG in rats with peripheral acute inflammation. Furthermore, a growing body of experimental evidence has reported that GLP-1 agonist exenatide can also produce anti-nociceptive actions in various pain hypersensitivity conditions, such as peripheral and diabetic neuropathy (Gong et al. 2014a, b;Aykan et al. 2019;Zhang et al. 2020).
Moreover, LRG treatment also resulted in greater increases in both threshold and latency compared to TR in non-inflamed rats. In addition to the anti-hypersensitive effects of LRG in inflamed rats, the analgesic/anesthetic effects of LRG in non-inflamed rats showed that LRG may also modulate the conduction and function of sensory nerves. Previous studies have also reported that the activation of GLP-1 receptors by the agonists can produce antinociception in a variety of animal pain models in addition to neuroprotective effect in diabetic neuropathy LRG (Palleria et al. 2017;Moustafa et al. 2018).
Previous preclinical studies have demonstrated that antinociceptive actions of exenatide can be prevented by opioid receptor antagonist naloxone (Gong et al. 2014a, b). It has also been suggested that activation of opioid receptors may be a mechanism for anti-nociception by GLP-1 agonist in pain hypersensitivity states and GLP-1 agonists may reduce drug reinforcement (Gong et al. 2014b;Aykan et al. 2019;Zhang et al. 2020). Therefore, in the present study, LRG and TR applications were combined and its anti-inflammatory effects were investigated to demonstrate the efficacy of LRG in alleviating pain behaviors mediated by tramadol, an opioid. When two drugs co-administered, the treatment can produce independent effects or additive effects (equal to the sum of the effects of each). Combined treatment can also inhibit or decrease each other's effects (antagonism), or, the effect can be greater than the expected effect (synergy). In this present study, combined administration of TR and LRG resulted in further reductions in sensitivities of both the inflamed and non-inflamed paw. The remarkable suppression of cold and mechanical allodynia with anti-hypersensitive activity may suggest that this combination treatment may be very effective in inflammatory pain behaviors.
Paw masses were measured at the end of the experiments (5 h after CG application) to determine the anti-edema effects of the treatments. Increased edema activity following CG administration resulted in an increase in paw mass up to approximately two times after 5 h. LRG, TR, or TR + LRG combination treatments showed significant decreases on the paw masses at approximately the same levels. These results may indicate the anti-edema effects of LRG, TR, and TR + LRG combination.
One of the most important signs of inflammation after CG was a significant increase in the temperature of the inflammation area (Libby 2007;Zhang and An 2007;Dawes and McMahon 2013). The increase in temperature in inflammation is an indication that the cellular immune defense mechanisms are activated against the harmful factor, which is caused by different mediators released in the inflammation area. This increase in temperature of the inflammation area was significantly reduced by all drug treatments. However, compared to TR, LRG caused a greater reduction in the temperature of the inflammation site. These results may clearly demonstrate the anti-pyretic effects of TR, LRG, or TR + LRG combination treatments.
The presented findings may imply that the anti-hypersensitive, anti-pyretic and anti-edema effects of treatments may be closely related with the inflammatory environments including various mediators such as cytokines, chemokines, and growth factor proteins, alone or all together. To obtain more detailed information about the molecular action mechanisms of treatments, the changes of biomarkers in the inflammation site were measured.
Findings showed that the levels of both pro-inflammatory and anti-inflammatory cytokines increased in the inflammation area. It has been known that inflammatory cells in the inflammation area cause an increase in the levels of proinflammatory cytokines, and anti-inflammatory cytokines suppress the effect produced by pro-inflammatory cytokines and thus maintain the balance between pro-and anti-inflammatory cytokines (Dawes and McMahon 2013;Annamalai and Thangam 2017). LRG or LRG + TR treatment significantly increased the levels of IL-10 in inflamed paw tissue. This increased level of IL-10 may contribute to the antiinflammatory activities of LRG by suppressing the release of pro-inflammatory cytokines from inflammatory cells. The finding also suggested that the anti-hypersensitive effects of TR may be related to suppressing the increase of the proinflammatory cytokine TNF-α in the inflamed tissue.
Angiogenesis plays a role in many physiological and pathological processes (Ferrara 2002). The presented findings may suggest that increased secretion of pro-inflammatory cytokines results in angiogenesis in inflamed tissue. VEGF is one of the most important factors stimulating angiogenesis, and its inhibition can also be preferred as a therapeutic target to reduce angiogenesis (Sporn and Roberts 1992;Ferrara 2002). It is known that TGF-β1 increases the release of VEGF angiogenic factors and is important in the occurrence of complications. In this present study, the suppression of VEGF-A and TGF-β1 levels by LRG suggests that the anti-angiogenic property of LRG may be an important mechanism contributing to its anti-inflammatory activity.
Many previous studies have shown that oxidative stress can also play important roles in the inflammatory process (Conner and Grisham 1996;Halici et al. 2007). In this study, it was found that TAS did not change in inflamed paw tissues, while TOS increased significantly. While TR did not cause any change, LRG and LRG + TR increased the TAS level in inflamed tissue. The findings may suggest that LRG treatment may also be important for the prevention of inflammation-induced oxidative stress due to its anti-oxidant properties.
Consequently, compared with TR, with known effects, reduction of sensitivities to the sensory stimuli of the noninflamed paw by LRG treatment may indicate that it has analgesic/anesthetic effects, which is described by reducing or slowing of the sensory nerve conduction and functions. LRG may also ameliorate the pain hypersensitivity, edema and fewer. Anti-hypersensitive, anti-edema and anti-pyretic effects of LRG in rats with peripheral acute inflammation may be due to its anti-inflammatory, anti-angiogenic and anti-oxidative actions in tissues (Fig. 7). Furthermore, LRG, when combined with TR, may be an important treatment option for controlling inflammation and inflammatory signs. However, more experimental and molecular studies are needed, especially to examine the anti-inflammatory potential of LRG and its mechanisms of action in the underlying inflammatory processes. In addition, studies on the pharmacological mechanisms of the additive/synergistic effect occurring in the combined application of LRG and TR are required.

Conflict of interest
The authors declare no competing interests.
Ethical approval All rat studies were conducted according to protocols approved by the Bolu Abant Izzet Baysal University animal experiments local ethical committee, Turkey, with the approval number 2020/30.

Consent for publication
All authors have reviewed the manuscript and given consent for publication.