Antibacterial and antibiofilm activities of a traditional herbal formula against respiratory infection causing bacteria

The plants, Althaea officinalis, Tilia cordata and Psidium guaja have been used traditionally to treat respiratory infection symptoms. Flowers of A. officinalis and leaves of T. cordata and P. guaja have been used to treat cough, sore throat, catarrh, oral and pharyngeal mucosa irritation. Therefore, this study was designed to examine the antibacterial and antibiofilm effects of these plants individually as well as in combination, as a formula against respiratory infections causing pathogens. The tested pathogens were Extended Spectrum Beta-Lactamase producing Escherichia coli (ESBL), Beta-Lactamase producing Escherichia coli (BL), Beta-Lactamase producing Klebsiella pneumoniae (BL), Beta-Lactamase producing Pseudomonas aeruginosa (BL), Enterobacter cloacae, and Beta-Lactamase producing Staphylococcus aureus (BL). The tested plants were extracted using ethanol and then fractionated using different polarity solvents (hexane, ethyl acetate and water). Disc diffusion and microdilution (Minimum Inhibitory Concentration) methods were used to evaluate the antibacterial activity while the antibiofilm activity was tested using crystal violet assay. The results showed that A. officinalis and T. cordata extracts and fractions exhibited weak antibacterial activity having MIC values ranged from 6.25 to 12.5 mg/mL. P. guaja exhibited moderate antibacterial activity with MIC values ranged from 6.25 to 1.56 mg/mL. Combination between these plants extracts and fractions in equal proportion provides stronger antibacterial (with MIC values ranged from 6.25 to 0.8 mg/mL) and antibiofilm activities (MBIC50 was 0.2 mg/mL). Therefore, this study provides a valuable scientific knowledge to support the use of plants in combination rather than individually.


Introduction
Respiratory infections are considered as the most common infection worldwide. Globally, more than 50 million deaths are reported every year due to infections related to respiratory system. 1,2 The respiratory infections could be caused by several pathogenic agents including viruses and bacteria. 3,4 In case of bacterial infections, most common causative agents for pharyngitis and tonsillitis are Group A Beta Hemolytic Streptococci, Corynebacterium diphtheria and Neisseria gonorrhoeae. Epiglottitis and bronchitis are frequently caused by Haemophilus influenzae type b, Corynebacterium diphtheria, Streptococcus pneumoniae and Mycoplasma pneumoniae. Pneumonia is caused by such agents as Streptococcus pneumoniae, Staphylococcus aureus, Streptococcus pyogenes, Haemophilus influenzae, Klebsiella pneumoniae, Escherichia coli, Pseudomonas aeruginosa and Mycobacterium tuberculosis. 5  respiratory infections is noteworthy. Due to this problem, in 2017, the World Health Organization issued a list of pathogens that required immediate action to develop new antibiotic therapy. 6 In addition to carbapenem-resistant Acinetobacter baumannii and carbapenemresistant Pseudomonas aeruginosa issued in this list, carbapenem-and third-generation cephalosporin-resistant have been considered. They later include Klebsiella pneumoniae; methicillin resistant and vancomycin resistant Staphylococcus aureus; penicillin nonsusceptible Streptococcus pneumoniae; and ampicillin-resistant Haemophilus influenzae. Therefore, finding new effective antimicrobial agents to treat respiratory infections is highly required. In Jordan, recent studies conducted in two districts, reported that the most common infections in elderly patients were urinary tract infection (53.75%), and respiratory tract infection (23.75%). 7,8 An increasing interest in plants traditional medicine has emerged. Traditional medicine is a fundamental source for treating diseases worldwide. 9,10 In developing countries, more than 70% of the population still rely on traditional medicine. 11 In Jordan, report showed that more than 65% of the populace think that traditional medicine is effective in curing diseases. 12 However, many reports had documented the traditional use of medicinal plants including traditional healers as key sources for these traditional knowledge. 10,13 Usually, individual or combined extracts of plant(s) are recommended to treat diseases. Frequently, combined formula of several traditional medicinal plants is used potently for medication with lower toxicity. 14 In Jordan, the flowers from A. officinalis, leaves from T. cordata and P. guaja are recommended frequently by traditional healers to treat ISSN 2616-0692 (Electronic) 528 © 2020 the authors. This work is licensed under the Creative Commons Attribution 4.0 International License respiratory infections. The traditional uses of these plants are broad and some of these uses are associated with the respiratory system infections.
A. officinalis (English name: Marshmallow Plant; Local name: Khetmeah) is a perennial species that belongs to Malvaceae family. It is native throughout Europe, Western Asia, and North Africa. Traditionally, decoction of the dried leaves and flowers is used to treat dry cough, oral and pharyngeal mucosa irritation and as expectorant. 15,16 Several secondary metabolites were isolated from A. officinalis including coumarins, scopoletin, hypolaetin-8-glucoside, Isoquercitrin, kaempferol, caffeic, p coumaric acid, ferulic acid, phydroxybenzoic acid, salicylic acid, p-hydroxyphenylacetic aicd and vanillic acid. 16 T. cordata (English name: lime tree; local name: Zezafoon) is deciduous tree that belongs to Malvaceae family. Traditionally, leaves is prepared as a herbal tea to relief the symptoms of common cold, coughs and catarrh. 17 Lime tree possesses a diverse groups of secondary metabolites including caffeic, p-coumaric, chlorogenic acids, kaempferol, quercetin, myricetin, hyperoside, quercitrin, isoquercitrin, anethole, citral, citronellol, eugenol, limonene, menthone, nerol, a-pinene, terpineol, fenchone, a-and b-thujone, and farnesol. 18 P. guaja (English name: Guajá; local name: Guava) is belonging to Myrtaceae family. It is a fruit tree that distributed all over the world. Decoction of leaves of P. guaja is used orally or as gargle to treat flu, sore throats and cough. [19][20][21] The leaves of P. guaja is rich in menthol, α-pinene, β-bisabolene, β-pinene, β-copanene, limonene, terpenyl acetate, isopropyl alcohol, caryophyllene, longicyclene, cineol, caryophyllene oxide, humulene, farnesene, selinene, curcumene and cardinene. 22 Moreover, many pharmacological investigations have confirmed that the extracts and isolated compounds from A. officinalis, T. cordata and P. guaja possess broad biological activities in which this supports their broad and extensive traditional uses. Pharmacological investigations showed that A. officinalis possess anti-complement, anti-inflammatory, antitussive, antioxidant and hypoglycemic effects while T. cordata possess anti-fungal, anti-viral, anti-inflammatory and Antioxidant activities. 16,23 Gutiérrez et al., 21 and Anand et al., 24 reported that P. guaja are effective as antioxidant, hepatoprotection, anti-allergy, antimicrobial, antigenotoxic, antiplasmodial, cytotoxic, antispasmodic, cardioactive, anticough, antidiabetic, antiinflamatory and antinociceptive agent. Therefore, this study was designed to examine the inhibitory effect of the flowers from A. officinalis, leaves from T. cordata and P. guaja individually as well as in combination as a formula against pathogens that caused respiratory infections. The ability of these plants extracts to inhibit P. aeruginosa and E. coli biofilms was also evaluated.

Plant materials and extraction
The flowers from A. officinalis, leaves from T. cordata and leaves from P. guaja were purchased from herbal markets (Mutah, Alkarak, Jordan). Voucher specimens (No. M120, M121 and M122, respectively) were deposited in the department of medical laboratory sciences, Mutah University, Jordan. The plants parts were dried in shade for 10 days then they were cleaned and ground into fin powder. From the powder plant, 250 g was weighed into flask containing 600 mL of 95% ethanol and kept at 20-25˚C for 24 h. The crude ethanol extracts for each plant were collected, filtrated using Millipore filter syringe (0.45 µm) and stored at 4°C.

Liquid-liquid fractionation
Liquid-liquid fractionation of the crude ethanol extracts for each plant were performed according to Bibi et al. 25 with some modification. Briefly, each extract (10 g) was suspended in 250 mL water and partitioned sequentially with two organic solvents (hexane and ethyl acetate, 200 mL, each) using separating funnel. Solvents of the ethanol extract and those of three fractions (hexane, ethyl acetate and water) were removed, filtrated and appropriately concentrated using rotary evaporator. Finally, all extracts and fractions (hexane, ethyl acetate and water) were stored as aliquots at 4°C.

Preparation of test extracts
Stock solutions were prepared by dissolving 100 mg from the test extracts and fractions (ethanol, hexane, ethyl acetate and water) separately in 1000 µL dimethyl sulfoxide (DMSO) and sequentially filtrated through 0.22 µm filter syringe. In addition, similar stock solutions were prepared for the herbal formula that composed of a mixture of extracts (ethanol, hexane, ethyl acetate and water) in equal proportion 1:1:1. All samples were stored at -4°C.

Antibacterial activity of plant extracts
Five Gram negative strains including Extended Spectrum Beta-Lactamase producing Escherichia coli (ESBL), Beta-Lactamase producing Escherichia coli (BL), Beta-Lactamase producing Klebsiella pneumoniae (BL), Beta-Lactamase producing Pseudomonas aeruginosa (BL) and Enterobacter cloacae, and one Gram positive strain; Beta-Lactamase producing Staphylococcus aureus (BL) were used in this study. These strains were provided by Al Bashir hospital (Amman, Jordan). In addition, two reference strains: E. coli ATCC 25922 and P. aeruginosa ATCC 10145 provided by Department of Biology, Mu'tah University, Al-Karak, Jordan were used. The disc diffusion method was performed as previously described [26][27][28] with some modifications. Briefly, 24 h bacterial suspension adjusted to 0.5 McFarland's standard (1.5x10 8 CFU/mL) was prepared. Then, Mueller-Hinton agar plates were seeded with 100 µL of the suspended bacteria and spreading was performed using sterile cotton swab. A sterile blank discs (6 mm in diameter) containing 0.5 mg of plant extracts, negative control (DMSO) or Ciprofloxacin (5 µg) were transferred into the inoculated plates. All the plates were incubated at 37°C for 24 h. After which, the inhibition zone formed was measured as millimeter diameter. Each test was performed in triplicate.

Minimum inhibitory concentration (MIC) of plant extracts
Microdilution method was used to estimate the Minimum inhibitory concentration (MIC) of each extract as previously described 29 with some modifications. A 96-well plate was used to prepare two-fold dilution of the tested extracts. The final concentrations of the extracts were 0.2, 0.4, 0.8, 1.56, 3.13, 6.25, 12.5 and 25.0 mg/mL. Then, 10 µL of bacterial suspension adjusted to 0.5 McFarland's standard (1.5x10 8 CFU/mL) was inoculated into each well. The same test was carried out with DMSO as a control. The inoculated plates were incubated at 37°C for 24 h. Each test was performed in triplicate. The lowest concentration of the tested extracts needed to inhibit the visible growth of the tested microbes after 24 h was considered as the MIC values.

Anti-Biofilm activity and minimum biofilm inhibitory concentration (MBIC)
The anti-biofilm activity against E. coli (ESBL), and P. aeruginosa (BL), was evaluated using crystal violet assay. 30 Sub-MIC concentrations equal to 0.2, 0.4, 0.8 mg/mL were prepared using 96well plate. Then 10 µL of bacterial suspension containing 1.5x10 8 CFU/mL (0.5 McFarland's standard) was inoculated into each well. The same test was carried out with DMSO as a control. Each test was performed in triplicate. After 24 h incubation at 37°C, the growth medium was discarded, and the plates were washed with distilled water and stained with 200 µL of 0.4% crystal violet. After 20 min, the stain was removed, and the excess stain was rinsed off with tap water (three times, 200 µL, each) before adding 200 µL of 95% (v/v) ethanol to solubilize the crystal violet. From the dissolved crystal violet into ethanol, 150 µL from each well were transferred to new 96 well plate for spectrophotometric measurement (OD590nm) in an ELISA reader. The lowest concentration of the tested extracts required to prevent the formation of ≥50% of the biofilm was considered as minimum biofilm inhibitory concentration (MBIC50). 31

Results and Discussion
Plants extraction yield In this study, the ethanol crude extracts were further fractionated using different polarity solvents including hexane, ethyl acetate and water. Yield of the tested plants extracts, and fractions are shown in Table 1. In general, the yield of hexane fractions was lower than the yields of water and ethyl acetate fractions. The ethyl acetate fractions of A. officinalis and P. guaja have the highest yield (38.14 and 58.90%, respectively), whereas T. cordata water fraction gave the highest yield (52.27%). The fractionation procedure was performed to obtain fractions containing compounds distributed according to their polarity. Variable polarity solvents were used including hexane, ethyl acetate and water. None polar compositions was extracted using hexane while medium and high polar compositions were extracted using ethyl acetate and water. 32 The high yield in A. officinalis and P. guaja extract using ethyl acetate and water indicated that the components present in these species are medium to high polar components.
Antibacterial activity of A. officinalis, T. cordata and P. guaja The antibacterial activity was evaluated using disc diffusion method and microdilution method. In general, P. guaja exhibited stronger antibacterial activity than A. officinalis and T. cordata (Table 2). No remarkable antibacterial activity was observed for A. officinalis and T. cordata. In particular, weak antibacterial activity of A. officinalis extract and fractions was reported with maximum inhibition zone of 8.33 mm (Table 2). Similar result was indicated for T. cordata extract and fractions with maximum inhibition of 8.67 mm ( Table 2). The antibacterial activity of A. officinalis ethanol extract and fractions was observed against P. aeruginosa (ESBL) and K. pneumonia (BL) while T. cordata ethanol extract and fractions showed no antibacterial activity (0.0 mm) against all strains tested except E. coli (BL). Our results regarding the antibacterial activity of A. officinalis and T. cordata are in parallel to that reported previously. Naovi 33 showed that the flower, leaf, root and seed extracts of A. officinalis at 10.0 mg/mL exhibited no antibacterial activity against Corynebacterium diphtheriae, Diplococcus pneumoniae, Staphylococcus aureus, Streptococcus pyogenes and Streptococcus viridans. Ozturk and Ercisli 34 showed that the aqueous extracts from aerial parts of A. officinalis had no antibacterial effects, whereas the methanol extracts exhibited moderate antibacterial activity against Acidovorax facilis, Bacillus sp., Enterobacter hormachei, and Kocuria rosea with maximum inhibition zone of 12 mm. Regarding the antibacterial activity of T. cordata, Fitsiou et al., 35 showed that T. cordata essential oil exhibited no antibacterial activity against Staphylococcus aureus ATCC 25923, Sarcina lutea ATCC 9341, Bacillus cereus ATCC 14579, Escherichia coli ATCC 25922. At 50 µg/mL, T. cordata hydroethanol extract exhibited weak antibacterial activity against methicillin-resistant Staphylococcus aureus (MRSA), Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, and Mycobacterium intracellulare with less than 37% inhibition activity. 36 P. guaja ethanol extract showed stronger antibacterial activity. All strains tested exhibited similar susceptibility with inhibition zone ranged from 9.83 to 11.5 mm. As shown in Table 2, all fractions of P. guaja exhibited antibacterial activity against the tested strains except hexane fraction. Ethyl acetate and water fractions appear to possess similar inhibitory effect and all strains were equally susceptible to the tested fractions. Maximum activity of P. guaja was observed when ethyl acetate fraction was tested against E. coli (BL) with inhibition zone of 13.17 mm. In other study, water, methanol and chloroform extracts of P. guaja leaves were reported to inhibit the growth of Staphylococcus aureus and beta-streptococcus group A. 37 It was reported that, the hexane, methanol, ethanol, and water extracts of P. guaja leaves exhibited antibacterial activity against gram positive bacteria B. cereus and S. aureus (mean zones of inhibition of 8.27 and 12.3 mm, and 6.11 and 11.0 mm, respectively) but had no activity against gram negative bacteria. 38 Minimum Inhibitory Concentration (MIC) of the plant extract against the tested isolates is shown in Table 3. The MIC ranged from 0.80 to 12.50 mg/mL. The MIC values of A. officinalis and T. cordata extracts and fractions were in the range between 6.25 to 12.50 mg/mL indicating weak antibacterial activity (Table 3). Previous reports showed that A. officinalis and T. cordata possess bacteriostatic and bactericidal activities. 34,39 Rezaei et al., 39 showed that hydroalchoholic extract of A. officinalis possesses bacteriostatic and bactericidal activities against Staphylococcus aureus at 330 and 660 μg/mL, respectively, whereas the extract was inactive against Gram negative bacteria including Listeria sp, Pseudomonas, and Escherichia coli. Ozturk and Ercisli,34 showed that, the aqueous extracts from aerial parts of A. officinalis had no antibacterial effects; whereas, the methanol extracts exhibited significant antibacterial activity against Acidovorax facilis, Bacillus sp., Enterobacter hormachei, and Kocuria rosea with range of MIC values (62.50 to 500 μg/mL). MIC and MBC of A. officinalis ethanol extract against P. aeruginosa was found 62.5 mg/L 40 . The ethanol extracts of T. cordata produced bacteriostatic activity against gram negative bacteria including Listeria ivanovii, Serratia rubidaea, Listeria innocua with range of MIC values (100 to 400 μg/mL) however, against E. coli, P. aeruginosa and the gram positive; Enterococcus raffinosus, Lactobacillus rhamnosus, S. epidermis, Brochothrix thermosphacta and Paenobacillus larvae, the MIC was more than 1000 μg/mL. 41 Lower MIC values were observed for P. guaja (1.56 to 6.25 mg/mL). The MIC values of P. guaja extracts and fractions were 1.56 mg/mL against all strains tested. The exception of this is the MIC values against P. aeruginosa (3.13 mg/mL) and E. cloacea (6.25 mg/mL). Previous reports showed that P. guaja extracts possess antibacterial activity with MIC values ranging from 150 μg/mL to 4 mg/mL. [42][43][44] Ethanol and water extracts of P. guaja were reported with bacteriostatic activity against L. monocytogenes, S. aureus, V. parahaemolyticus, and A. faecalis with MIC values ranged from 0.1 to 0.4 mg/mL and 0.2-0.7 mg/mL, respectively. 45 The variation in MIC values revealed in other reports for the same plant extracts may be due to the variation in their chemical composition, solvents and methods of extractions as well as the pathogenic strains used. 27,46 Antibacterial activity of the herbal formula comprising A. officinalis, T. cordata and P. guaja The antibacterial activity of the herbal formula comprising A. officinalis, T. cordata and P. guaja in equal proportion was evaluated using disc diffusion method ( Table 2). The antibacterial activity of the water fraction was higher than the antibacterial activity of the ethanol extract. In addition, the antibacterial activity of A. officinalis, T. cordata and P. guaja extracts and fractions combined in equal proportion increased compared to their activities when they were evaluated individually. The inhibition zones observed for the combined ethanol extracts ranged from 9.33 to 16.17 mm while the zone of inhibition ranged from 0.0 to 11.5 mm when these extracts were evaluated individually. The antibacterial activity for the combined fractions ranged from 9.0 to 19.17 mm. Maximum inhibitory activity was observed when water fractions were combined generating the range of inhibition zones between 11.83 and 19.17 mm. Combined water fractions was the most active against E. coli (BL), E. coli (ESBL) and E. coli ATCC 25922 leading to inhibition zones of 19.17, 19.0 and 18.67 mm, respectively. E. cloacae was the most resistant strain to these combinations with moderate diameter of inhibition zones ranges between 9.0 and 11.83 mm.  Interestingly, the combined extracts and fractions showed stronger antibacterial activity. However, this activity is closer to the activity of the P. guaja extracts and fractions suggesting that P. guaja is the major active components in this formula. The sub-MIC values for the combined extracts and fractions reported in this study indicated that combining two or more extracts may increase the antibacterial activity. The reason might be related to the synergistic interaction between the formula components. 49 In fact, traditional healers use combination between plants to increase the therapeutic effects of a single plant. 50

cordata and P. guaja extracts and fractions
The antibiofilm activity of ethanol extracts and fractions individually and in combination was evaluated using crystal violet assay at sub MIC concentrations (0.2, 0.4, 0.8 mg/mL) against P. aeruginosa ( Figure 1) and E. coli (Figure 2). The results showed dose-dependent inhibition activity for all extracts and fractions tested. At the highest concentration tested (0.8 mg/mL), hexane fraction of A. officinalis, ethanol and water fractions of T. cordata and P. guaja completely inhibited the formation of P. aeruginosa biofilm (>90%). Hexane, ethyl acetate and water fractions of P. guaja caused complete inhibition of E. coli biofilm. Regarding the combined extracts and fractions, all combined extracts and fractions completely inhibited the formation of P. aeruginosa and E. coli. Minimum Biofilm Inhibitory Concentration (MBIC50) for the tested extracts and fractions and in combination was determined ( Table 4). The lowest MBIC50 reported was 0.2 mg/mL for P. guaja and the herbal formula extracts and fractions indicating potent antibiofilm activity. Higher MBIC50 was observed for T. cordata and A. officinalis extracts and fractions (0.4 mg/mL, each). In another study, A. officinalis extracts at concentrations equal to MIC (62.5 mg/L) or higher inhibited P. aeruginosa biofilm by 87.3 and 93.7%, respectively. 40 There were no available data about the antibiofilm activity of T. cordata and P. guaja and the formula.