Effects of Oxalis barrelieri L. (Oxalidaceae) aqueous extract on diarrhea induced by Shigella dysenteriae type 1 in rats

Abstract Aim Oxalis barrelieri is a medicinal plant commonly used in Cameroon, for the treatment of many diarrheal diseases. The antibacterial properties of O barrelieri aqueous extract (WOb) against Shigella dysenteriae type 1 were investigated in vitro and in vivo. Methods Antibacterial activity was evaluated in vitro by disc diffusion method and by macrodilution method. S dysenteriae type 1 at a dose of 1.2 × 109 CFU was administrated orally to rats to induce shigellosis. For 6 consecutive days, diarrheic rats were treated with O barrelieri aqueous extract (50 and 100 mg/kg BW) or norfloxacin (20 mg/kg BW). The diarrheal stool weight and S dysenteriae type 1 density were assessed during the treatment period, and death rate recorded. Nitric oxide production in blood and in colonic homogenate and blood parameters were assessed, and the histological section of the colon was performed in the survivors. Results The minimal inhibitory concentration and minimal bactericidal concentration of WOb were, respectively, 6 mg/mL and 25 mg/mL. The mean minimal bactericidal concentration/minimal inhibitory concentration ratio for WOb against S dysenteriae type 1 was high (˃4); WOb could be classified as a bacteriostatic drug. WOb significantly (P < .01) reduced bacterial density and diarrheal stool weight. WOb decreased nitric oxide production (P < .01) in the large intestine and protected the mucosa of the colon from bacterial destruction. Conclusion The results suggest that O barrelieri aqueous extract possesses bacteriostatic and antidiarrheal activities and reduces damages caused to intestinal mucosa barrier by pathogenic mechanisms of Shigella. This extract could be used as an alternative therapeutic for infectious diarrhea.

osmotic substances and/or laxatives. Infectious diarrhea can be due to virus, bacteria, and/or parasites. 4 In the world and particularly in developing countries, infectious diarrhea still remains one of the leading causes of infant mortality. 9 A multicentric study from 6 Asian countries estimated Shigella as the causative agent in 5% of the diarrheal cases. 10 As little as 10 to 100 Shigella can cause shigellosis in human, 11 whereas 1.2 × 10 9 Shigella can cause dysenteric diarrhea in rats. 12 Shigellosis occurs worldwide, in sporadic, endemic, epidemic, and pandemic forms. 13 Most of the cases are children <5 years of age. The annual number of shigellosis episodes throughout the world is estimated to be 164.7 million, with 69% of all episodes and 61% of all deaths attributable to shigellosis involving children <5 years of age. 10 The economic impact of diarrhea and its treatment are of considerable importance. 14,15 The annual treatment costs for diarrhea are very high and ranged from US$ 907 116 to US$ 1 851 280 for ambulatory clinical consultations and from US$ 701 833 to US$ 4 581 213 for hospitalizations. 16 The re-emergence of Shigella dysenteriae type 1 (Sd1) with added resistance to ciprofloxacin, which has epidemic potential, has also been reported. 10 Many synthetic drugs such as diphenoxylate and loperamide are available for the treatment of diarrhea, but they have toxic side effects. 17 Therefore, the search for new, safe, more effective, and less toxic molecules has continued to be an important area of research in pharmacology. Since antiquity, diarrhea has been treated with medicinal plants in traditional medicine. 17 Oxalis barrelieri L. is traditionally used by Cameroon inhabitants (Central Africa) for diarrhea care. The O barrelieri aqueous extract, at doses of 50 and 100 mg/kg BW, showed significant antidiarrheal activities in rats treated with castor oil. 18 Phytochemical studies of O barrelieri aqueous extract revealed the presence of compounds such as phenols, terpenoids, anthocyanidins, anthraquinones, coumarins, and saponins. 18 This work has been initiated to evaluate the antidiarrheal properties of O barrelieri aqueous extract on infectious diarrhea, by using Sd1induced diarrhea in rats as a model. HNC. Whole plants were washed thoroughly with water, shade-dried, and ground. The powder (407 g) was macerated in distilled water (5 L) for 3 days in a percolator. After 3 days of maceration, the mixture was filtered by opening the tap of the percolator and was concentrated by vacuum distillation at 50°C to yield 128 g (31.45%) of brown extract. A stock solution (10 mg/mL) was prepared by dissolving 10 g of dry extract in distilled water to obtain 1000 mL of solution.

| Experimental animals
Prior to the study, male and female Wistar albino rats (60-98 g), approximately 6 weeks old, were selected and allowed to acclimatize for 1 week to our laboratory environment (22°C-25°C and 12 h light/12 h of darkness). In vivo experiments on rats were performed according to the European Union guidelines on animal care (CEE Council 86/609) that was adopted by the Ministry of Scientific Research and Innovation of Cameroon. 19 Animals housed in metabolic cages (1 animal/cage) were fed a diet consisting of carbohydrates (50%-55%), fats (15%-20%), and proteins (25%-30%). 20

| Bacterial strain
Clinical isolates of Sd1 from patients with severe infections were provided by the Centre Pasteur of Yaoundé, Cameroon.

| In vitro antibacterial susceptibility 2.4.1 | Disc diffusion method
Minimal inhibitory concentration (MIC) was determined with adapted E-test according to disc diffusion method. 21 Sterile 6-mm Ø filter paper discs (Schleicher & Schul, no. 2668, Dassel, Germany) were impregnated with 50 μL of O barrelieri aqueous extract at various concentrations. Sd1 inoculum of 0.5 mL (5 × 10 5 CFU) was flooded in Salmonella Shigella (SS) and in Mueller Hinton agar (Oxoid) and incubated for 15 minutes at 37°C. After incubation, the impregnated paper discs were placed on the plates according to decreasing extract concentrations (50 000-24 μg/mL) ( Figure 1). The petri dishes were incubated at 37°C for 24 hours. After 24-hour incubation at 37°C, the growth inhibition and the MIC (first disc that did not present growth inhibition) were observed on the petri dishes. The test was performed under sterile conditions in duplicate and repeated 3 independent times.

| Macrodilution method
This method has been adopted from NCCLS M26-A, 22 with modifications. 23,24 The Sd1 strains were adjusted to achieve a turbidity equivalent to a 0.5 McFarland (1 × 10 8 CFU/mL) and diluted (1:1000) 25 in brain heart infusion (Oxoid). A dilution series of the extract, ranging from 50 000 to 24 μg/mL, were prepared and then transferred to the broth in 14 tubes. A 1.0-mL extract was pipetted into tubes by twofold dilutions. Freshly grown bacteria of 1.0 mL were added to the tubes in a density of 10 5 CFU/mL (final concentration/ tube). The tubes were incubated overnight at 37°C. CFU was determined by diluting each well in tenfold dilutions. From each dilution, aliquots were transferred to agar plates and incubated FIGURE 1 Realization of the adapted E-test method with the Oxalis barrelieri aqueous extract on Salmonella Shigella agar inoculated with Shigella dysenteriae type 1 overnight. On the following day, the number of colonies was evaluated, and the initial CFU/tube retrospectively calculated by the formula:

| Sd1-induce diarrhea 2.5.1 | Diarrhea induction and treatment
Rats were housed separately in metabolic cages. Before diarrhea induction, we checked that our animals were not carrying Shigella. In To exclude food involvement in induction of diarrhea, Group 6 (normal control [NC]) consisting of 5 normal rats received food and water, but no bacterial inoculums or drug. This group was treated only with distilled water (10 mL/kg BW) for 6 consecutive days.
The number of deaths was recorded during treatment. The stools were collected daily using a sterile stool pot of the metabolic cage. The weight and quality of feces were examined daily for 6 consecutive days of treatment. Sd1 counts in feces were done before induction of diarrhea and daily for 6 consecutive days after onset of diarrhea. For this purpose, 0.5 g of stool was dissolved in 4.5 mL of sterile saline solution, serial dilutions were made, and 0.5 mL of each dilution was inoculated on the SS agar plate and incubated for 24 hours at 37°C. 30 After incubation, the number of Sd1 was determined. 33 After 6 days of treatment, all survival animals were sacrificed and their blood and their colon were collected for blood cells count using manual method 34 and/or for NO test using the modified Griess method. 35 Colon fragments were fixed in 10% buffered formalin for histopathological examination. 30

| NO dosage
Nitric oxide concentration was evaluated in serum and in colon homogenates. 30 To obtain Griess solution, 0.25 mL of Griess 1 (0.8 g sulfanilic acid + 250 mL acetic acid 30%) was added to 0.25 mL of Griess 2 (0.05 g of α-naphthylamine + 100 mL acetic acid 30%). A 0.5-mL serum or homogenate of the colon was added to 0.5 mL of Griess solution, and the mixture obtained was left for 20 minutes at room temperature. After 20 minutes, the optical density of each mixture was read using a spectrophotometer (T60-1611ESW) at 553 nm and recorded. 35

| Histopathological investigations
Histopathological investigations were done according to methods described in the literature. 38 For this purpose, colon fragments of each experimental rat were fixed in 10% buffered formalin in labeled flasks for histological examination. These colon fragments were embedded in paraffin wax, and 2-μm-thick sections were made with the microtome.
These preparations were mounted on glass slides that were then stained with hematoxylin and eosin and examined under a standard light microscope (MOTIC 1820 LED: SM7432-MC1ST-RPIWFM). 39

| Statistical analysis
The data are means ± standard error of the mean ( X ± SEM) expressed in tables, figures, and photos. These data were analyzed by 1-way analysis of variance followed by the Dunnett t test and the Tukey test using Computer Pad InStat 3.05 (Graph Pad software, USU).

| Susceptibility of Sd1 to O barrelieri aqueous extract
In vitro, the O barrelieri water extract showed an intersection point between the growth zone and the growth inhibition zone corresponding to the MIC. This MIC was about 6 mg/mL (Figure 2).

| MIC and MBC values of O barrelieri water extract by the graphic method
The O barrelieri water extract (WOb) showed in vitro inhibitory activity on Shigella growth. The MIC and MBC values were 6 mg/mL and 25 mg/mL, respectively ( Figure 3). The MBC/MIC ratio was 4.16.

| Antidiarrheal activities
Normal rats that were not inoculated with Shigella and were treated with distilled water showed no signs of diarrhea. A few hours (4 h) after Sd1 inoculum administration, the rats became calm, less mobile, curled up, and showed erect hairs. Twenty-six hours after inoculum administration, the rats emitted the first diarrheal stool and became more aggressive. During the treatment, rats recovered mobility progressively, and their aggressiveness decreased. Diarrheal stools were soft or liquid containing mucus or blood marks and attracted flies by their fetid odors. These symptoms disappeared after 3 days of treatment.
During treatment, no death was recorded in all treated groups and NC. However, we recorded 100% death in DC group ( Table 1). The first day of onset of diarrhea, stool weights were 3.06 ± 0.49 g, 2.78 ± 0.62 g, 2.60 ± 0.37 g, and 2.42 ± 0.37 g, respectively, for TD, Nor20, WOb50, and WOb100. These values increased in untreated group over time, decreased significantly (P < .01) from the second day in norfloxacin-treated group (Nor20) and from the third day in extract-treated groups ( Figure 4). Diarrhea was eliminated by the sixth day of treatment.

| Effect of O barrelieri on NO production
In DC rats, NO production in colon was markedly high compared to NC: 418.15 ± 48.65μM against 57.34 ± 3.76μM (P < .01), 2 days after the onset of diarrhea. After 6 days of treatment, NO production in colon was significantly reduced (P < .01) by 20 mg/kg BW norfloxacin and by 50 and 100 mg/kg BW aqueous extract of O barrelieri after the sixth day of treatment ( Figure 6A). The blood level of NO was higher in DC than normal rats: 132.42 ± 17.26μM against 60.59 ± 6.61μM.

| Effects of O barrelieri water extract on the microhistology of colon in diarrheal rats
The control rat colon ( Figure 7A) showed normal mucosa. In untreated diarrheal rats, colon histology showed destruction of the mucosa ( Figure 7B). However, in rats treated with norfloxacin 20 mg/kg ( Figure 7C), with 50 mg/kg ( Figure 7D) or 100 mg/kg ( Figure 7E) O barrelieri aqueous extract, the microhistology of the colon showed a normal mucosa.

| DISCUSSION
The purpose of this study was to provide scientific support Picralima nitida, 41 1.17 mg/mL with aqueous extract of Mallotus oppositifolium, 32 and 3.5 mg/mL with aqueous ethanol extract of Euphorbia prostrata ait. 31 These results are probably due to the fact that this crude extract might contain less active compounds against this bacterial strain. For aqueous extract of O barrelieri, the ratio MBC/MIC was higher than 4, and this could thus indicate a bacteriostatic activity. 29 This bacteriostatic property could be confirmed by the number of Sd1 counted in the stool. The bacterial population decreased moderately in WOb-treated rats compared to norfloxacin-treated rats. Diarrheal rats developed soft stools, glary or mucus-linked lumpy stools, and feces with a fetid odor that evidenced the presence of pus that are typical signs of infectious or "invasive" diarrhea. 30,42 In untreated diarrheic rats, destruction of the colonic mucosa and significant NO production were due to the pathogenic effects of Shigella. However, the protection of the colonic mucosa and the signifi-