Natural populations of Galphimia spp. attenuates peripheral and central inflammation


 The genus Galphimia is widely distributed in Mexico, and is represented by 22 species, including medicinal species. The sedative and anti-inflammatory effects of galphimines produced by the species Galphimia glauca have been documented. Formerly, molecular studies using DNA barcodes demonstrated that nine populations botanically classified as Galphimia glauca belong to four different species of the genus Galphimia, and that only one exhibited the sedative properties; however, all the collected species showed anti-inflammatory activity. Other bioactive compounds like quercetin, galphins, galphimidins and glaucacetalins have been identified from methanolic extracts of plants botanically classified as Galphimia glauca. The aim of this work was to determine the anti-inflammatory activity of methanolic extracts of nine collected Galphimia spp. populations grown in Mexico. The possible modes of action were analyzed by evaluating the inhibition of LPS-induced inflammation processes both in vitro and in vivo. The nine populations were evaluated by an in vitro model using RAW 264.7 murine macrophage cells, and two populations (a galphimine-producing and a non-galphimine-producing population) were selected for the in vivo experiments of systemic inflammation and neuroinflammation in mice. Results suggest that an anti-inflammatory in vitro effect was present in all the studied populations, evidenced by the inhibition of nitrite production. An inhibitory systemic inflammation in mice was exerted by the two analyzed populations. In the neuroinflammation model, the anti-inflammatory effect was demonstrated in methanolic extract of the non-galphimine-producing population. For the populations of Galphimia spp. studied herein, the anti-inflammatory effect could not be correlated to the presence of galphimines.


Introduction
The genus Galphimia (Malpighiaceae) is represented in Mexico by 22 of the 26 existing species worldwide (Anderson 2007). At present, the taxonomic classi cation of species belonging to the genus Galphimia has been challenging and confused. The misunderstanding has been caused by the similarity in the morphology of various of the species, or by mistakes in labelling all specimens from Mexico as G. gracilis and G. glauca (Anderson 2007). It is reported that plants botanically classi ed as G. glauca have been used since pre-Hispanic times to treat different illnesses, including in ammation and central nervous disorders (Estrada 1985). Many investigations have been conducted to understand the phytochemical and pharmacological properties of G. glauca (Dorsch et al. 1992;Tortoriello and Ortega 1993;Müller et al. 1998;del Rayo et al. 2002;Cardoso-Taketa et al. 2004, 2008Herrera-Ruiz et al. 2006;Náder et al. 2006;Ortíz et al., 2010;Sharma et al., 2012a;Abarca et al., 2014). In Mexico, Doctor Mora, Guanajuato is the locality where the rst studies in a natural population botanically classi ed as G. glauca were carried out (Tortoriello and Lozoya 1992;Tortoriello and Ortega 1993;Toscano et al. 1993;Osuna et al. 1999). Subsequent studies showed that plants from this locality have anxiolytic and sedative activities in both, mice (Cardoso-Taketa et al. 2008;Sharma et al. 2012a) and humans (Herrera-Arellano et al. 2007Romero-Cerecero et al. 2018). It has been demonstrated that galphimines are the bioactive compounds with the anxiolytic and sedative effects (Tortoriello and Lozoya, 1992;Cardoso-Taketa et al., 2008;Sharma et al., 2012a). Furthermore, anti-in ammatory effects of these metabolites have been proposed using an in vivo model in mice (González-Cortazar et al. 2014). Galphimines constitute a family of 15 (named from A to O) nor-secofriedelane-type triterpenes (Toscano et al. 1993;Cardoso-Taketa et al. 2004;Ortega et al. 2020). Additionally, two investigations developed by our group in seven natural populations collected in the states of Chiapas, Guanajuato, Jalisco, Morelos and Querétaro, and botanically classi ed as G. glauca, showed that only two populations produce galphimines (Cardoso-Taketa et al. 2008;Sharma et al. 2012a), exhibiting anxiolytic and sedatives activities in mice; however, all of them had anti-in ammatory activity, using the tetradecanoylphorbol acetate-induced mouse ear in ammation model (TPA) (Sharma et al. 2012a). Other studies have documented the anti-in ammatory activity of G. glauca (Müller et al. 1998;González-Cortazar et al. 2014), in extracts and with pure galphimines (González-Cortazar et al. 2014). Other bioactive compounds as methyl gallate, gallic acid, quercetin, tetragalloylquinic acid, ellagic acid, galphins A-C, galphimidin, galphimidin B and glaucacetalins A, D and E, have been isolated from plants botanically classi ed as G. glauca (Dorsch et al. 1992;Neszmélyi et al. 1993;Müller et al. 1998;del Rayo-Camacho et al. 2002;Ortíz et al. 2010;Rios et al. 2020).
In order to clarify the identity of Galphimia species, we performed two molecular studies, using DNA barcoding analysis of natural populations botanically classi ed as G. glauca, including the populations studied in the present investigation. These studies suggest the presence of four species of the genus Galphimia among the collected populations (Sharma et al. 2012b;Gesto-Borroto et al. 2019). In consequence, it was considered to use the term Galphimia spp. to refer to the populations here studied.
In the present investigation, individuals (six per population) from nine natural populations of Galphimia spp. were collected in different geographical locations in Mexico; ve of these populations were studied for the rst time. To determine the anti-in ammatory activity of methanolic extracts of all of these populations, one in vitro and two in vivo bioassays were used. These bioassays were employed for the rst time to evaluate the anti-in ammatory activity from plants of Galphimia spp. in a cool and dry place without direct sunlight, and then powdered by mortar and pestle. Samples were kept under −70°C before to extract preparation. The methanolic extracts were prepared from all nine populations to evaluate the in vitro anti-in ammatory effect of Galphimia spp. Pulverized dried material (100 mg) for each sample was mixed with 1 mL MeOH. Samples were vortexed for 2 min, sonicated for 15 min and then centrifuged at 10,000 rpm for 15 min. The material residue containing the pellet was reprocessed four times to achieve exhaustive extraction. The four supernatants were collected, mixed and dried at room temperature. For both in vivo anti-in ammatory assays, four samples (5g each) of the pulverized dried material of QC and MS populations were mixed with 50 mL MeOH. Samples were vortexed for 2 min, sonicated for 30 min and then centrifuged at 4,000 rpm for 30 min. Further steps of the procedure were followed as describe above. antibiotics, in a 25 cm 2 ask, at 37 º C, and 5% CO 2 atmosphere in a humidi ed incubator.

Determination of nitrite concentration
Nitrite is a stable product of the NO oxidation and its concentration was determined in the cultured medium via the Griess reaction as an indicator of NO production. Speci cally, 100 µL of supernatant from each well was mixed with 100 µL of Griess reagent in separates 96-well plate. After an incubation of 10 min at room temperature, the optical density was determined at 540 nm with a microplate reader SpectraMax® iD3.

Cell viability
Cell viability was assessed using the rezasurin assay (Ahmed et al. 1994). The cell viability assay was performed after the NO determination. Brie y, 180 µL of DMEM/F12 medium supplemented with 10% heat-inactivated FBS and 20 µL of rezasurin 2 mM (Sigma Aldrich) were added to each well, and further incubated for 12 h at 37ºC in a 5% CO 2 humidi ed incubator. The uorescence was recorded using an excitation and emission wavelength of 544 and 590 nm, respectively, with a microplate reader SpectraMax® iD3.
In vivo anti-in ammatory effects of Galphimia spp. in LPS-treated mice Mice Male C57BL/6 mice, 7 to 8 weeks-old (25 g approx.), from a breeding colony of the Instituto de Investigaciones Biomédicas (IIB) at Universidad Nacional Autónoma de México (UNAM) were employed.
Mice were divided into groups of ve to six animals, and kept in polysulfon boxes with food and water ad libitum before and during the experiments. The holding room was maintained at 22 ± 3°C with a 12:12 h light-dark cycle.
All housing and experimental procedures were approved and conducted under the guidelines established by the Institutional Committee on the Care and Use of Experimental Animals of the IIB at UNAM (approval number ID 232).
In vivo anti-in ammatory effect of Galphimia spp. extracts in the periphery A model for systemic LPS-induced in ammation was employed (Qin et al. 2007;Meneses et al. 2016Meneses et al. , 2017. Mice received 1 mg/kg of LPS or and equivalent volume of the vehicle (0.9% NaCl; endotoxin-free isotonic saline solution (ISS); PiSA, Guadalajara, Mexico) administered intraperitoneally (i.p.). Treatments The peripheral anti-in ammatory activity of methanolic extracts of Galphimia spp. was evaluated in LPS treated mice (Fig. 1A). The anti-in ammatory effect of methanolic extracts of Galphimia spp. was evaluated employing a galphimine-producer population (QJ) and a non-galphimine-producer population (MS). Two doses of extracts from each population (200 and 600 mg/kg) were injected i.p. after the administration of LPS dose. Two hours later mice were sacri ced by cervical dislocation, and peritoneal uid was collected to obtain macrophages and dendritic cells.

Flow cytometry
Isolated peritoneal cells were treated following the methodology reported by Meneses et al. (2016Meneses et al. ( , 2017. For analysis, cells were distinguished using antibodies against CD86 and MHC-II, for macrophages, as well as CD86 and CD11b for dendritic cells. Macrophage and dendritic cells activation status was assessed by examining the medium uorescence intensity of the membrane receptors mentioned above.
Anti-in ammatory effect of Galphimia spp. extracts in the LPS-induced neuroin ammation Mice received 5 mg/kg of LPS or and equivalent volume of the vehicle (0.9% NaCl; endotoxin-free isotonic saline solution (ISS); PiSA, Guadalajara, Mexico) administered i.p.

Treatments
The anti-in ammatory effect of methanolic extracts of Galphimia spp. was evaluated employing a galphimine-producer population (QJ) and a non-galphimine-producer population (MS). One dose of each extract (600 mg/kg) was injected i.p. 48 h after the administration of LPS dose. Mice were sacri ced 24 h later by cervical dislocation, and brains were extracted. The expression of ionized calcium binding adaptor molecule (Iba1) and glial brillary acidic protein (GFAP), which are expressed speci cally in microglia and astrocytes, respectively, was performed by immuno uorescence analysis (Fig. 1B).

Immuno uorescence analysis
For the immuno uorescence analysis, each brain was processed according to the methodology reported by Meneses et al. (2016Meneses et al. ( , 2017. Brain sections were labeled with rabbit anti-GFAP (Invitrogen, Carlsbad, CA, USA), anti-Iba1 (Wako Chemicals, Inc., Richmond, VA, USA) and 4',6-diamidino-2-phenylindole (DAPI), to detect astrocytes, microglia and nuclei imaging, respectively. Photographs were obtained using a digital camera attached to a light microscope (Nikon Digital Sight DS-Ri1). Regions of the hippocampus (cornu ammonis (CA1 and CA2)) and the cortex (CR1 and CR2) were selected and processed using ImageJ software (National Institute of Health, Bethesda, MD, USA).

Statistical analysis
For the analysis of nitrite concentration, and for the anti-in ammatory effect in the systemic in ammation and neuroin ammation models, data were reported as mean ± standard deviation. For nitrite concentration determinations, and for the anti-in ammatory effect in the systemic in ammation model, statistical analysis was done by one-way ANOVA, followed by Dunnett's t-test, p value 0.05 was considered to show signi cant difference among groups. In the neuroin ammatory experiment, statistical analysis was done by a non-parametric test (Kruskal-Wallis followed by Mann-Whitney U-test), p value 0.05 was considered to show signi cant differences among groups. All statistical analyses were carried out using the program GraphPad Prism 6.01 (GraphPad Software Inc.).

Results
In vitro inhibition of macrophage activation Determination of nitrite concentration and cell viability A signi cant decrease using one way ANOVA in the production of nitrite (stable product of the NO oxidation) was observed in the cells treated with the four concentrations (25, 50, 75 and 100 µg/mL) of the methanolic extracts of all studied populations of G. glauca, in comparison with the cells that were only stimulated with LPS (Fig. 2). A signi cant difference among the galphimines-producer and nongalphimines producer populations was not demonstrated. The value of nitrite production of macrophages that only were LPS-treated was 9.74 ± 2.21 µM. The highest effect in the inhibition of nitrite production (1.71 ± 0.83 µM) was observed with the HZ population extract at 75 µg/mL. Five populations showed the better results inhibiting the nitrite production in the LPS-stimulated macrophages at 100 µg/mL and 75 µg/mL, from which two (HZ and QJ) are galphimines-producer populations, and three (MM, MS and ZV) are non-galphimines producer populations. At the concentration mentioned above, the methanolic extracts of these ve populations, did not showed signi cant differences in the production of nitrite, in comparison with the cells treated with aminoguanidine (100 µg/mL), or with the control macrophages without LPS (Fig. 2).
The four concentrations of the methanolic extracts were assessed to determinate its effect on the viability of the cell line RAW 264.7. None of the methanolic extracts, at any evaluated concentration showed a signi cant reduction on viability of the macrophages, in comparison with the cells that did not received any treatment (Fig. 3).
In vivo anti-in ammatory activity Activation of peritoneal macrophages and dendritic cells The activation of macrophages was evaluated identifying the CD86 and MHC-II molecules. The statistical analysis by one-way ANOVA allowed to identify a signi cant reduction in the percentage of macrophages CD86+, CD86+/MHC-II + and the medium uorescence intensity (MFI), in mice treated with methanolic extracts of both populations; the one that produces galphimines (QJ) and the non-producer (MS), at both evaluated doses (200 and 600 mg/kg), in comparison with mice that were only LPS-stimulated (Fig. 4C, D and E). The decrement in CD86 + macrophage activation was higher at 600 mg/kg with values of 44.00 ± 13.45% and 47.86 ± 10.78% for MS and QJ, respectively, in comparison with the group that only received LPS (74.16 ± 4.83 %). The same results were obtained in the analysis of macrophages CD86+/MHC-II+, the dose of 600 mg/kg was more effective in the inhibition of the macrophages activation for both populations (MS, 50.04 ± 13.61% and QJ, 53.90 ± 10.51%), in comparison with LPS-stimulated (76.80 ± 4.79%) (Fig. 4D). Besides, the MFI signi cantly decreased for CD86 + macrophages for the methanolic extracts of both populations at both analyzed doses (Fig. 4E). The reduction of the MFI was higher at 600 mg/kg with values of 986.80 ± 358.70 and 1031 ± 314.40, for MS and QJ populations, respectively.
Dendritic cells activation was analyzed by the detection of the constitutive membrane receptor CD11b and the costimulatory receptor CD86. In the analysis of CD11b+/CD86 + dendritic cells it is not possible to refer to an anti-in ammatory effect, since signi cant differences between the mice LPS-stimulated, the mice without any treatment, or the mice that only received ISS, were not present. However, the MFI of CD36 was signi cantly reduced in mice that received the methanolic extracts of both populations evaluated at both doses (Fig. 5C). The decrease of the MFI was higher at the dose of 600 mg/kg for MS (790.85 ± 462.3) and QJ (572.00 ± 252.40) populations, in comparison with mice LPS-treated only (1931.00 ± 440.10) (Fig. 5C). No signi cant differences was observed for macrophages MHC-II + or dendritic cells CD86 + or CD11b+ (data not shown).

Immuno uorescence analysis of the neuroin ammation model
The anti-in ammatory effect of two methanolic extracts of Galphimia spp. was evaluated in an LPSinduced neuroin ammation model. The samples corresponded to a galphimine-producer population (QJ) and a non-galphimine-producer population (MS). Considering the results arose in the experiment of systemic in ammation, only the dose of 600 mg/kg was injected i.p. 48 h after the administration of LPS (5 mg/kg).
The visualization of Iba1 and GFAP proteins, which are expressed speci cally in microglia and astrocytes, respectively, was performed by immuno uorescence analysis (Fig. 6). The quanti cation of Iba1 (Fig. 7) and GFAP (Fig. 8) expression was developed in the hippocampus (CA1 and CA2) and cortex (CR1 and CR2) determining the MFI of two brain sections for each mouse, in order to calculate the area on every photomicrograph and to analyze the resulting data.
The immuno uorescence analysis demonstrated that LPS increased the expression of Iba1 in microglia in the hippocampus (Fig. 7). A statistically signi cant reduction of the expression of Iba1 in CA1 and CA2 was observed in mice that received the methanolic extract of MS population in comparison with the group in which LPS was only administered (Fig. 7). On the other hand, the ones which received the methanolic extract of QJ population, exhibited a reduction in the expression of Iba1, in CA1 and CA2 regions; however this decrease was not signi cant different with the MFI of sections from mice treated only with LPS (Fig. 7). In the cortex region, no increased of Iba1 expression was observed (Fig. 7).
In the analysis of the expression of GFAP; LPS was able to increase its presence in the CA1 region. The administration of the methanolic extracts of MS showed a reduction in the expression of GFAP, but was not signi cant in comparison with the group that received only LPS (Fig. 8). A reduction in the expression of GFAP in mice treated with the methanolic extract of QJ was not observed. For GFAP, neither in CR1 or CR2, an induction of in ammation was observed (Fig. 8).

Discussion
In a previous work carried out by our group, the tetradecanoylphorbol acetate (TPA)-induced mouse ear in ammation test, was performed to determine the anti-in ammatory effect of methanolic extracts of seven populations of the genus Galphimia (Sharma et al. 2012a). Otherwise, the inhibition of the nitrite production at the highest concentrations (100 and 75 µg/mL) of ve of the methanolic extracts (HZ, QC, MM, MS and ZV) is similar to the level of nitrite production of amininoguanidine-treated cells or macrophages without any treatment, indicating a reduction of the nitrite concentration to basal levels. Furthermore being aminoguanidine a selective inhibitor of the inducible nitric oxide synthase (iNOS) (Misko et al. 1993;Corbett and McDaniel 1996), these results suggest that methanolic extracts reduced the activity of this enzyme, which is activated in macrophages that differentiate to an M1 phenotype and take part of the in ammatory process (Murray and Wynn 2011;Martinez and Gordon 2014). The methanolic extracts of Galphimia spp. did not show a cytotoxic effect over the RAW 264.7 cells, suggesting that the decrease in the nitrite production was due to an inhibition of its synthesis and not to cell death.
Regarding to the in vivo model it is well known that CD86 and MHC-II are molecules involved in T cell activation, that act as costimulatory signals and in the presentation of antigens, respectively (Ashley et al. 2012). In the evaluation of macrophages activation, for both methanolic extracts (QJ and MS) at the dose of 600 mg/kg the level of expression of CD86 did not show signi cant differences with regard to the mice without treatment or to those which only received ISS, hence the methanolic extracts of the population QJ and MS were decreasing to basal levels the expression of CD86 in the membrane of the macrophages (Fig. 4D).
The analysis of dendritic cells activation was performed through the detection of the constitutive membrane receptor CD11b and the costimulatory receptor CD86. CD11b regulates cell adhesion and migration to mediate the in ammatory response (Tan et al. 2000). The statistical analysis by one-way ANOVA showed that the percentage of dendritic cells CD11b+/CD86 decreased signi cantly in mice that were treated with the methanolic extracts of both populations (QJ and MS) for the dose of 600 mg/kg (Fig. 5B). Nevertheless, it is not possible to clearly associate this phenomenon with an anti-in ammatory effect, since signi cant differences between the mice LPS-stimulated, the mice without any treatment, or the mice that only received ISS, is not present. However, the signi cant reduction of the MFI of CD86 in mice which received both doses (200 and 600 mg/kg) of the two methanolic extracts (QJ and MS), in comparison with the mice that only received LPS, could be correlated with an anti-in ammatory effect. The methanolic extracts inhibited the expression of CD86, being greater to the higher doses in both populations; even for the population of QJ was signi cantly different in comparison with the lower doses of QJ and MS.
The reduction in the activation of macrophages and dendritic cells as a consequence of the actions of the evaluated methanolic extracts (QJ and MS) at both doses (200 and 600 mg/kg) indicates that the LPS-induced in ammatory process was being controlled. This inhibitory effect is better with the higher doses, suggesting that higher proportion of the metabolites with an anti-in ammatory effect present in the methanolic extracts, is required in order to obtain a more effective results. Likewise in the in vitro assay, in most of the evaluations performed, no signi cant differences were found among the galphimine-producer population (QJ) and the non-galphimine-producer population (MS); consequently the anti-in ammatory effect in Galphimia spp. populations is a complex response probably due to the presence of several metabolites.
The effect of Galphimia spp. in reducing microglia/macrophages activation is consistent with the macrophages activation in the periphery. Microglia are involved in the immune defense of the brain and also contribute maintaining homeostasis. These cells could change their phenotype to an activated form, when the homeostasis in central nervous system is disrupted by different kind of damages. The uncontrolled activation of microglia induces the production of pro-in ammatory cytokines and cytotoxic mediators, which are factors implicated in neuropathological conditions (Liu et al. 2011;Meneses et al. 2017;Kabba et al. 2018;Vainchtein and Molofsky 2020). The results of this work suggest that the methanolic extract of MS population control the LPS-induced excessive activation of microglia.
In this study, we con rmed the inhibition in the expression of Iba1 protein, in mice that received the methanolic extract of MS population, suggesting a decreased of microglia/macrophages activation. Nevertheless, an inhibitory effect of the methanolic extract of QJ population was not observed, neither in Iba1 or GFAP expression. Similarly to the evaluations previously shown, no signi cance differences were found among the galphimine-producer population (QJ) and the non-galphimine-producing population (MS); consequently the anti-in ammatory effect it is not potentiated by galphimines. Furthermore, in the neuroin ammation model, no anti-in ammatory effect was demonstrated for the methanolic extracts of the galphimine-producer population (QJ).

Conclusion
In this work we studied nine natural population of Galphimia spp., taking into account an integrative approach involving in vitro and in vivo anti-in ammatory analyses. It was possible to demonstrate the anti-in ammatory activity of crude extracts from all the collected populations of Galphimia spp. in the in vitro model, as well as from two selected populations in two in vivo models, which were used by the rst time with the genus Galphimia. The results obtained in this study support the potential use of methanolic extracts of Galphimia spp., as an alternative to treat central nervous system in ammation. The current research enabled to enhance the knowledge concerning Galphimia species in the light of its pharmacological activity. For further investigations it is necessary to identify the presence of diverse secondary metabolites with anti-in ammatory activity in the methanolic extracts of the natural populations of Galphimia spp., in order to develop a comprehensive analysis on the important antiin ammatory properties of crude extracts from plants of Galphimia species.  Seventy two hours later mice were sacri ced and the expression of ionized calcium binding adaptor molecule 1 (Iba1) and glial brillary acidic protein (GFAP) was studied by immunostaining.    MS200 y MS600: methanolic extracts of the population of Santa Catarina, Morelos, 200 and 600 mg/kg, respectively; QJ200 y QJ600: methanolic extracts of the population of Jalpan de Serra, Querétaro, 200 y 600 mg/kg, respectively. The percentage and MFI values correspond to the mean ± standard deviation. Signi cance was determined by one-way ANOVA (p<0.05, compared to LPS).

Figure 6
Glial brillary acidic protein (GFAP) and ionized calcium binding adaptor molecule (Iba1) expression in the four studied groups (mice that received an i.p. administration of ISS, LPS, or methanolic extracts of MS and QJ populations) of the neuroin ammation model experiment. Representative immuno uorescence of 50 μm coronal sections of mouse brain of the different groups stained with anti-Iba 1 (red) and anti-GFAP antibodies (green) and 4',6-diamidino-2-phenylindole (DAPI) (blue nuclei). Scale bar 200 μm