In vitro antibacterial and antidiarraheic activity of root bark extract of Anogeissus leiocarpa ( Combretaceae ) during an experimental bacterial diarrhea induced by Escherichia coli extended-spectrum β-lactamases ( ESBL ) in albino Wistar rats

1 Biochemical Pharmacodynamy Laboratory, Training and Research Unit of Biosciences, University Félix HouphouëtBoigny, P. O. Box 582 Abidjan 22, Abidjan, Côte d’Ivoire. 2 Department of Biochemistry-Genetics, Training and Research Unit of Biological Sciences, University Peleforo Gon Coulibaly, P. O. Box 1328 Korhogo, Korhogo, Côte d’Ivoire, 3 Department of Clinical and Basic Biochemistry, Pasteur Institute of Côte d’Ivoire, P. O. Box 490 Abidjan 01, Abidjan, Côte d’Ivoire.


INTRODUCTION
Diarrheal diseases, the third leading cause of death in the world, after respiratory diseases and AIDS (OMS, 2015) continue to be one of the leading causes of morbidity and mortality.Globally, they constitute the second leading cause of infant mortality in children less than five years (Bhutta et al., 2013), with about 2.5 billion diarrhea episodes and 2.2 million deaths all years (Bahmani et al., 2015).Of the 2.2 million annual deaths, 37% of cases occur in sub-Saharan Africa (OMS, 2016).In Côte d'Ivoire, diarrhea represents 14% of the reasons for consultation in maternal and child health centers (Kouakou, 2012) and 15% of deaths of children less than 5 years (Liu et al., 2014).The main agents involved are viruses, protozoa and bacteria among which Escherichia coli is one of the most important etiologic agents (Asadi et al., 2010).
Given the magnitude of this situation, World Health Organization (WHO) has adopted a treatment based on oral rehydration (OR) and oral rehydration solution (OSR).However, the annual rates of use of OR and OSR, respectively of 0.39 and 1.02% are still very low (Birger et al., 2006).In addition to this program, antibiotics, although used, are becoming increasingly ineffective because of microbial resistance and toxicity problems.They are also still beyond the reach of population grants in developing countries because of their high cost.In this situation, it is necessary and urgent to offer to disadvantaged populations more accessible new therapeutic solutions taking into account their culture and their purchasing power.
To meet this challenge, exploring the rich Ivorian medicinal flora with nearly 800 species of medicinal plants (Aké-Assi, 1991) could offer alternative therapeutic solutions.It is in this context, that the present study focused on Anogeissus leiocarpa (DC) Guill.& Perr, a plant of the Ivorian medicinal flora.Phytochemical studies carried out on A. leiocarpa root barks revealed the presence of phenolic compounds (flavonoids, tannins, leucoanthocyanins and polyphenols), saponins and sterols (Moronkola and Kunle, 2014;Gbadamosi and Ogunsuyi, 2014).These compounds are known to have antimicrobial activity (Mann et al., 2014;Adamu et al., 2017).
This study focused on the in vitro and in vivo investigation of the antibacterial potential of Combretaceae, a selected medicinal plant of Côte d'Ivoire flora by using infected rats with E. coli extendedspectrum β-lactamases (ESBL).

METHODOLOGY Plant material
The root barks of A. leiocarpa were used.These organs were harvested in January 2013 in Kouto (Bagoué region), a town located at 725 km north of Abidjan.This plant has been authenticated by Professor Aké-Assi of the National Floristic Center of Félix Houphouët-Boigny University and compared to the voucher specimen No. CNF 14798.

Animals
Albinos white rats, male and female of Wistar strain aged 2 to 3 months and weighing between 180 to 200 g were used.

Preparation of the total aqueous extract and organic fractions of A. leiocarpa
Total aqueous extract was prepared according to Guede-Guina (1993) and ethyl acetate, dichloromethane and ethanol fractions according to Manga et al. (2013).

Preparation of the bacterial inoculum
A volume of either 0.01, 0.1 or 1 ml of opalescent pre-culture broth was collected for Pseudomonas, enterobacteria and Staphylococci, respectively, and then diluted in a tube containing 10 ml of physiological saline to constitute dilution inoculum 10°.

Preparation of concentration ranges
A concentration range of 500 to 7.81 mg/ml similar to that of Mann et al. (2014) was prepared by the double dilution method in 7 test tubes.These tubes were then sterilized by autoclaving at 121°C for 15 min (Bolou et al., 2011) and stored in a refrigerator at +4°C.

Preparation of culture media
Culture media were prepared according to manufacturers' instructions (Liofilchem ® and Cultimed ® ).

Antibacterial sensitivity test
The agar well diffusion method was used (Irshad et al., 2012) for assessing the in vitro antibacterial activity of the prepared extract.Cefotaxime (CTX, 30 μg) for enterobacteria and gentamycin (GEN10 μg) for other bacteria served as positive controls.Only concentrations of extracts or fractions with an inhibition zone *Corresponding author.E-mail: kmania1@yahoo.fr.Tel: +22507784098/+22540455823. Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License diameter (IZD) greater than 7 mm were reported in results

Minimum inhibitory concentration (MIC)
The incorporation of the plant extracts into Muller-Hinton (MH) agar was done using the double dilution method in agar slant tubes as described by Ouattara et al. (2013).Nine experimental tubes whose concentration varies to double dilution from 50 to 0.195 mg/ml and 2 control tubes, the growth control tube (TC) and the sterility control tube (TS) are prepared.The slope of the experimental tubes and that of the TC tube was seeded.The tubes were incubated at 37°C for 24 h.The MIC was the concentration of the first tube from which no microbial visible growth (Khaleel et al., 2016).

Minimum bactericidal concentration (MBC)
MBC is the lowest concentration of substance that leaves at most 0.01% of surviving germs.Using a loop calibrated at 2 μl, the contents of the tubes in which no haze was observed were seeded on MH (Box B) in parallel streaks 5 cm in length at the surface, starting with by the MIC tube.After 24 h incubation in an oven at 37°C, the numbers of colonies on the streaks of Box B with those of Box A were compared.In practice, the CMB corresponds to the concentration of the experimental tube whose number of colonies present on the streak is less than or equal to the number of colonies present on the streak of the dilution 10 -4 .

Experimental design
Forty-two rats of both sexes previously deparasized through oral administration of 10 mg/kg body weight of tetracycline for 3 days (Ricicová et al., 2010) were used.They were divided into 7 Lots of six rats each.Rats from 6 Lots received, orally, 2 ml of the infective dose (Eman et al., 2008) evaluated at 2 × 10 8 CFU/ml with a sterile disposable needle-less syringe.The 7th batch is uninfected.Treatment started 48 h after the induction of diarrhea following the appearance of diarrheal faeces.For treatment, the infected rats received orally concentrations of the ethanolic fraction of A. leocarpa and ciprofloxacin (the reference antibiotic) according to the prescription of Venkatesan et al. (2005).Thus, daily and for 14 days, Lot 1 (uninfected and untreated) received 1 ml of distilled water; Lot 2 (infected and untreated) received 1 ml of distilled water; Lot 3 (infected and treated) received 2 ml of ETHA at dose of 69.44 mg/kg body weight corresponding to 1 × MIC; Lot 4 (infected and treated) received 2 ml at a dose of 271.87 mg/kg body weight (5 × CMI); Lot 4 (infected and treated) received 2 ml of ETHA at a dose of 631.3 mg/kg body weight (10 × CMI) ; Lot 6 (infected and treated) received 2 ml of ETHA at a dose of 2222.22 mg/kg body weight (40 × CMI) and Lot 7 (infected and treated) received 1 ml of ciprofloxacin at dose of 5 mg/kg body weight w.

Enumeration of faecal E. coli
Rats faeces were collected in sterile polyethylene sterile bags of 500 ml (LMR ® , France) placed under the cage of the rats, twice daily for 2 weeks.One (1) g of faeces removed with a spatula was transferred to 9 ml of BMH, then vortexed for 10 s and incubated at 37°C for 3 to 4 h to obtain dilution inoculum10°.From this dilution 10°, dilutions ranging from 10 -1 to 10 -5 were prepared.This serial dilution were cultured (0.1 ml) in duplicate on Mac Conkey agar (two plates per dilution).After incubation at 37°C for 18 to 24 h, typical smooth pink color colonies were counted on two successive Adrien et al. 465 dilutions.Plates inoculated with a sample dilution that yields between 10 and 300 colonies per plate were read (ISO 7218, 2007).The E. coli faeces load (UFC per g of faeces) was calculated according to the formula proposed by ISO 7218 ( 2007): E. coli faeces load (UFC/g of faeces) where ΣC: sum of colonies counted on the two Petri dishes retained, V: volume of seeded inoculum on each Petri dish, and d: dilution corresponding to the first retained Petri dish, with the least diluted inoculum.

Evaluation of the water content of faeces
Fresh faeces was collected after 6 h and weighed individually to determine the wet weight (WW).Faecal samples were dried in a conventional oven at a temperature of 70°C.After 24 h, dried faeces were weighed again for determining their dry weight (DW) (Navarro et al., 2006).The difference between the WW and the DW helped to determine the water content of faeces (Navarro et al., 2006) expressed as a percentage, by the following relation: Water content (%) 100

Mortality rate and pathological manifestations
The mortality rate of the rats in the different groups is calculated as the number of dead rats during the experiment compared to the total number of rats used in each Lot (Eman and Hoda, 2008).Faeces were checked continuously for color and consistency, the development of diarrhea in infected Lots, as well as any change in activity and behavior were recorded weekly throughout the experiment.

Statistical analysis
All data during this experience were expressed in three replicates.
The results are expressed as mean ± standard error of means.They were determined by Dunnett's test using GraphPad Prism 7.0 Statistic Software and were considered significant at p-value less than 0.05.

Ethical approval
The experiments were conducted according to the ethical guidelines of Ethics Committee of Pasteur Institute of Côte d'Ivoire (Charter of Ethics of the Pasteur Institute, Text of September, 2012).

Antimicrobial activity of aqueous extract and organic fractions of A. leiocarpa
The effects of total aqueous extract and organic A. leiocarpa fractions obtained by agar well diffusion method on bacterial growth are shown in Table 1.The total aqueous extract and the dichloromethane fraction showed no activity on all the bacteria tested with inhibition zones diameters less than 8 mm.Both ethanol (ETHA) and ethylacetate (EAA) fractions were active against all infectious bacteria tested with an IZD values ranging from 8 to 16 mm and 8 to 15 mm, respectively.Diameter of zone of inhibition decreased when reducing gradually the extract or fraction concentration excepted for EDMA and ETAA.On a general note ETHA exhibited higher sensitivity than EAA.Only S. aureus Meti-R and Streptococcus spp.displayed the highest sensitivity with an IZD up to 16.64 at 250 and 500 mg/ml, highest zone of inhibition observed with ETHA compared to EAA.With the exception of E. coli ESBL which is the main resistant strain at 125 mg/ml, the IZD of most bacteria tested with ETHA are larger than those of EAA.In addition, EAA at the concentration of 125 mg/ml revealed 3 resistant strains which are E. coli ATCC, E. coli ESBL and P. aeruginosa with an IZD below 8 mm.The inhibition zone of total aqueous extract and dichloromethane fraction at any concentration is below 8 mm.For standard drugs, extreme sensitivity was observed with an IZD that ranged from 10 to 30 mm.

Antibacterial parameters of the total aqueous extract and organic fractions of A. leiocarpa
MIC, MBC and MBC/MIC ration values are shown in Table 2.The MICs values of all strains studied with both ETAA and EDMA extracts were above 50 mg/ml, thus their MBC and MBC/MIC ratio were not determined.The MICs of A. leiocarpa ranged from 6.25 to 12.5 mg/ml in ETHA fraction.For EAA fraction, it was 6.25 mg/ml for S. Typhi 43PI16 and S. Typhi 1586, but was up to 25 mg/ml for E. Coli ATCC, ESBL, and P. aeruginosa but it was up to 50 mg/ml for Shigella dysenteriae and K. oxytoca.The MBC/MIC ratio of EAA and ETHA fractions observed were strictly inferior to 4 corroborating the bactericidal effect against all bacteria tested.

Infection rate, mortality in rats and faeces appearance
The antibacterial activity in vivo was evaluated by increasing the concentration of ETHA shown in Table 3.
These results indicated antibacterial activity in vivo after 7 days of treatment especially from Lot 3 (ETHA 69.44 mg/kg body weight) to Lot 7 (ETHA 2222.22 mg/kg body weight) was correlated to a gradual decrease of mortality rate from 30 to 0%.However, apart from Lot 1 as control and untreated rats in Lot 2 which scored 40 and 10% mortality rate at 7 and 14 days, respectively, there was no mortality rate

Effect of ETHA on E. coli enumeration in rats diarrheal faeces
The effect of ethanolic fraction on bacterial load in diarrheic faeces is as shown in Figure 1.It can be seen that after the 2 days of rats infection, the E.
coli colony forming units per g of faeces (UFC) counted decrease from Lots 3 to 7 when the concentration of ETHA was increased progressively.Outstandingly, the UFC of Lot 6 is close to that of the Lot 7, the standard drug that drops rapidly to zero after 6 days of treatment but the UFC of Lot 2 remained higher than those with treatment during the 13 days of study.

Effect of ETHA on water content
From the Figure 2 that shows the percent of facael water content, a significant decrease of water lost during the ETHA treatment was observed and it is strongly correlated to the diminishing feacal E. coli load in Figure 1.Only the rats treated with ciprofloxacin or cefotaxime showed no statistically significant difference (p > 0.05) in water content when compared with Lot 1 which represent not uninfected and untreated rats.

Effect of ETHA on the volume of diarrheal faeces
To gain additional information on the antibacterial activity of ETHA, the effect of ETHA on the volume of diarrheic faeces investigation was perfomed on both volume and weight of the rats' faeces (Figures 3 and 4).There was a significant difference (p < 0.01) on both volume and weight of diarrheic faeces.Also, there was no significant (p > 0.05) decrease in infected and treated rats from 69.99 to 2222.22 mg/kg body weight when compared with uninfected and untreated rats.Notably, it is observed that the weight and volume estimated for the Lot 6 is almost similar to that of the drug control (Lot 7).

DISCUSSION
The purpose of this study was to summarize the investigation of the antimicrobial properties of both the aqueous extract and organic fractions (ethyl acetate, dichloromethane and ethanol) of A. leiocarpa using antibacterial activity in vitro and in vivo, respectively.This work made clear the advantages of the combination of these two experiments to obtain interesting information not only to gain insights into sensitivity or inhibitory measurement but also to obtain real time data from the direct effect of the fraction test on selected laboratory animal (the Wistar albino rats).
The results showed that microorganisms tested were susceptible to plant extracts and the highest inhibitory activity was observed for ethanol and ethyl acetate fractions which presented antibacterial activity against all bacteria examined.
These observations were consistent with previous study realized by Biyiti et al. (2004) who indicated that by a diffusion method, bacteria were sensible for IZD of 9 to 14 mm, very sensitive for IZD of 15 to 19 mm, highly sensitive for IZD larger than 20 mm whereas not sensible for IZD less than 8 mm, which is in accordance with the present results.Strains of E. coli ATCC 25922, E. coli ESBL, P. aeruginosa 131813, S. Typhi 43PI16, S. Typhi 1586, S. dysenteriae 1079PI15 and K. oxytoca were therefore sensitive to ethanolic fraction of A. leiocarpa.S. aureus Meti-R 1532C/10 and Streptococcus spp.were very sensitive to the ethanolic and ethyl acetate fractions of A. leiocarpa according to their inhibition diameters (14.57± 0.20 to 16.64 ± 0.56 mm).Similar results were obtained by Ichor et al. (2011) and Mann (2012), who obtained diameters of inhibition with strains of S. Typhi, E. coli, Shigella spp.and P. aeruginosa, ranging from 9 to 17 mm in the presence of the methanolic leaves extract of A. leiocarpa.In addition, Mann et al. (2008) showed that the ethanolic leaves extract of A. leiocarpa inhibit the in vitro growth of strains of Pseudomonas MDR, S. aureus Meti-R and E. coli.
MICs determined using the broth dilution method were correlated to those obtained by the diffusion method.The MIC value gave a measure of the antibacterial performance of antibiotics but it appeared that the MICs values of plant extracts and essential oils were not standardized.Also, it was important to note that there were no consensus on the inhibition concentration for natural products, hence the consistency of the antimicrobial activity results are based on the growth inhibition zones observed and the ratio of MBC/MIC estimated.The results of this study showed that, ethanolic fraction (ETHA) was the most active fraction which displayed the lowest MICs values, ranging from 3.12 to 12.5 mg/ml, while the ethyl acetate fractions (EAA) gave the highest MICs that ranged from 6.25 to 25 mg/ml apart from S. dysenteria and K. oxytoca but no antibacterial activity were recorded with ETAA and EDMA.These results were similar to those of Timothy et al. (2015) and  Ali et al. (2017) with MICs ranging from 6.3 to 44.6 mg/ml and 5 to 20 mg/ml, respectively with ethanolic and ethyl acetatefraction extract of A. leiocarpa barks.Although antibacterial activity was detected in both EAA and ETHA fractions against almost tested bacteria strains, the MBC values (minimum bactericidal concentration) also showed a similar pattern of activity from that of MIC.The extract action was bactericidal when the ratio of the MBC/MIC is ˂4 and bacterioatatic for MBC/MIC >4 (Berché et al., 1991).
Based on results scored, the EAA and ETHA extracts on sensitive bacteria were especially bactericidal against both Gram positive and Gram negative bacteria.This result could be explained by the fact that most antimicrobial active components were less polar compounds that were not water soluble and so the organic solvent extracts showed a more potent activity.Contrary to the use of dichloromethane which was an organic solvent, no bacterial activity has also been observed.That was in accordance with previous work of Mabiki et al. (2013) that showed for the most of part of terpenes/terpenoids compounds in dichloromethane extract.It was known that because of this chemical diversity, terpenes/terpenoids have great industrial uses as flavors, fragrances (Schwab et al., 2008), high grade lubricants, biofuels, agricultural chemicals and in the near future will play a more significant role in medicines (Niehaus et al., 2011).
On the whole, Gram-positive bacteria (S. aureus Meti-R 1532C/10 and Streptococcus spp.) were more sensitive than Gram-negative bacteria.Bari et al. (2010) corroborated the high resistance of some Gram-negative bacteria compared to Gram-positive bacteria.In fact, Gram-negative staining bacteria had efflux pumps that prevented the intracellular accumulation of antibacterial agents (Demetrio et al., 2015).It was noted that ethanolic fraction of A. leiocarpa demonstrated in vitro the most active antibacterial activity during this study.In order to confirm this interesting observation, this in vivo therapeutic activity was evaluated using an experimental bacterial diarrhea induced by the E. coli ESBL, a strain resistant to β-lactams.
Diarrhea could be measured by several parameters such as the water content, weight, volume, bacterial load in the diarrhea feces and death of the specimen.For this instance, an infective dose of E. coli of 2 × 10 8 CFU/ml similar to that used by Mushtaq et al. (2005) was used to induce experimental E. coli diarrhea in rats.This dose brings about a colonization of the gastrointestinal tract in 24 h, and was responsible for dehydration, loss of appetite, appearance of demolded, and soft, liquid or semi-liquid light brown faeces (Forrester, 2002).
Mortality rate of 50% was observed in the untreated and infected Lot but no death was noted for both Lots treated with Ciprofloxacin and the ETHA fraction of 631.3 and 2222.22 mg/kg body weight, respectively.In fact, a significant reduction (p < 0.001 and p < 0.01) of rats' death was observed from 30 to 0% when increasing the dose of ETHA and a decrease of the rats weight and volume of diarrheal faeces indicating a progressive eradication of the E. coli inducted infection compared to infected and untreated Lots.These results illustrated the therapeutic action of the ETHA fraction.
These observation was corroborated by an important decreasing of faecal E. coli ESBL load during the experiment that may be ascribed to the healing power of the extract.After the treatment, the bacterial load of the infected and treated Lots decreased significantly (p < 0.001) as during the treatment and dose-dependent manner compared to the untreated infected Lot.Similar results were obtained by Niehaus et al. (2011) and Ibrahim and Sarhan (2015) when enumerating E. coli from diarrheal rat faeces.These results highlighted the therapeutic potentialities of the ethanolic fraction of A. leiocapa.
The antibacterial effect of A. leiocarpa extracts could be attributed to the bioactive compounds present in the extracts such as phenolic compounds (flavonoids, tannins, leucoanthocyanins and polyphenols), and saponins and sterols (Moronkola and Kunle, 2014).It has been suggested that bioactive compounds act through two main mechanisms of action.The first is related to their hydrophobic property, which facilitates their adhesion to the surface of bacteria, causing their instability (Jongbloed et al., 2007).The second was the inactivation of different bacterial molecules such as enzymes or receptors, following the binding of bioactive compounds (Pandey and Kumar, 2013).Some of these compounds could lead to cell membrane perturbations and exert a βlactam action on the transpeptidation of the cell wall.Other compounds could interact with the lipid bilayers in cell membranes, leading to the separation of the two membranes, thus leading to cellular swelling and cell death (Tshingani et al., 2017).
The reduction in the bacterial burden of faeces in infected and untreated rats may be explained by the protective action of their immune system against the pathogens (Lunga et al., 2014).The in vitro and in vivo analyses of antibacterial activity revealed the antibacterial and therapeutic potential of A. leiocarpa.

Conclusion
This study showed that root barks of A. leiocarpa extracts possessed antibacterial activity by inhibiting in vitro bacterial growth.Its therapeutic activity was carried out against E. coli in experimentally infected rats by the significant reduction of defecation frequency, water content of faeces, weight and volume of diarrheic faeces, and E. coli load of faeces.The present results demonstrated that A. leiocarpa should be used in the traditional therapeutic arsenal against resistant infectious germs.

Figure 3 .
Figure3.Volume of faeces of rats treated and untreated during experiment.Values are means±SEM.Each Lot includes 6 animals (n=6/Lot).***p<0.001:significant difference compared to the uninfected and untreated Lot, ns: no significant at p<0.05 when compared with the uninfected and untreated Lot (Lot1).Lot 1: no infected and no treated, Lot 2: infected and untreated, Lot 3: infected and treated with ETHA at dose of 69.44 mg/kg bw corresponding to 1 × MIC, Lot 4: infected and treated with ETHA at dose of 271.87 mg/kg bw corresponding to 5 × MIC, Lot 5: infected and treated with ETHA at dose of 631.3 mg/kg bw corresponding to 10 × MIC, Lot 6: infected and treated with ETHA at dose of 2222.22 mg/kg bw corresponding to 40 × MIC, Lot 7: infected and treated with ciprofloxacin at dose of 5 mg/kg bw.

Figure 4 .
Figure 4. Weight of faeces of rats treated and untreated during experiment.Values are means±SEM.Each Lot includes 6 animals (n=6/Lot).***p<0.001and **p<0.01:significant difference compared to the uninfected and untreated Lot, ns: no significant at p<0.05 when compared with the uninfected and untreated Lot (Lot1).Lot 1: no infected and no treated, Lot 2: infected and untreated, Lot 3: infected and treated with ETHA at dose of 69.44 mg/kg bw corresponding to 1 × MIC, Lot 4: infected and treated with ETHA at dose of 271.87 mg/kg bw corresponding to 5 × MIC, Lot 5: infected and treated with ETHA at dose of 631.3 mg/kg bw corresponding to 10 × MIC, Lot 6: infected and treated with ETHA at dose of 2222.22 mg/kg bw corresponding to 40 × MIC, Lot 7: infected and treated with ciprofloxacin at dose of 5 mg/kg bw.

Table 1 .
Sensitivity of bacterial strains tested to aqueous extract and organic fractions of Anogeissus leiocarpa.

Table 2 .
Antibacterial parameters of the total aqueous extract and organic fractions of A. leiocarpa.

Table 3 .
Infection and mortality rate during experiment.Lot 3: infected and treated with ETHA at dose of 69.44 mg/kg bw corresponding to 1 × MIC; Lot 4: infected and treated with ETHA at dose of 271.87 mg/kg bw corresponding to 5 × MIC; Lot 5: infected and treated with ETHA at dose of 631. 3 mg/kg bw corresponding to 10 × MIC; Lot 6: infected and treated with ETHA at dose of 2222.22 mg/kg bw corresponding to 40 × MIC; Lot 7: infected and treated with ciprofloxacin at dose of 5 mg/kg bw.
MR: Mortality rate, IR: infection rate; Lot 1: no infected and no treated; Lot 2: infected and untreated;