Anti-infective efficacy of Psidium guajava L. leaves against certain pathogenic bacteria

Introduction

Given the heavy global burden of infectious diseases, it is imperative to discover novel pharmaceutical assets for combating antimicrobial resistance, with particular focus on antibiotic-resistant bacterial pathogens recently listed by the World Health Organization as of high/critical priority (Tacconelli et al., 2018). Since the antibiotic pipeline lacks new mechanisms against resistant bacteria, particularly gram-negative bacteria (see here for more information), it is necessary to look for new antibiotics as well as non-antibiotic approaches to tackle bacterial infections.

A reverse pharmacology approach (Raut et al., 2017) of investigating plant extracts, particularly those employed in documented or folklore traditional medicine, for their potential anti-pathogenic efficacy may pave the way for discovery and development of novel antimicrobial molecules/ formulations. We undertook the current study to investigate anti-infective potential of one such plant extract, Psidium guajava L. (common name- guava; Family- Myrtaceae) leaf extract, against five different pathogenic bacteria. This plant has traditionally been used for treatment of various gastrointestinal problems including diarrhea and dysentery (Birdi et al., 2010), which are caused usually due to microbial infections. Validation of such traditional medicinal practices through modern scientific approach is necessary for their wider acceptance in the community, and for building public confidence in them (Kothari, 2018).

Methods

Microwave Assisted Extraction (MAE) (Kothari et al., 2009)

1 g of leaf powder was soaked into 16 mL of water or 50% ethanol, and subjected to microwave heating (Electrolux EM30EC90SS) at 720 W. Total extraction duration was 140 s, of which first heating was for 40 s, and subsequent two heating cycles of 10 s each. Intermittent cooling period between any two heating cycles was kept 40 s. Liquid volume at the end of extraction was 4 mL.

Vacuum Assisted Extraction (VAE)

1 g of dry leaf powder was mixed with 16 mL of water. Vacuum pump (MEDICA INSTRUMENT Mfg. Co.) was attached to the vessel containing plant material and solvent, and the working pressure was set at 7.36 psi (15 In. Hg). Total duration of heating was 20 min, of which for 15 min the system was at 65°C (at which boiling started). Extraction was stopped when liquid volume was reduced to 4 mL.

Extraction performed by methods described above, was followed by macro-filtration using nylon strainer followed by centrifugation (at 10,000 rpm for 15 min; Remi BZCI-8729), and filtration with Whatman paper # 1 (Axiva, Haryana). After this filtration, solvent was evaporated from the extract. For bioassay, extracts was reconstituted in absolute DMSO (Merck, Mumbai). Reconstituted extracts were collected in sterile flat bottom glass vials (15 mL, Merck, Mumbai) covered with aluminum foil, and protected from light to avoid photo-oxidation of light-sensitive compounds. The internal surface of vial cap was also wrapped with aluminum foil to avoid leaching of vial cap material (Houghton & Raman, 1998). Reconstituted extract was stored under refrigeration for further use. Extraction efficiency was calculated as percentage weight of the starting dried plant material.

Extraction efficiency obtained with these methods was 6.30%, 5.80%, and 6.0% respectively. All the extracts were reconstituted in dimethylsulfoxide (DMSO, Merck) upon drying, and stored under refrigeration (4-8º C) till further use.

In vivo efficacy of these water extracts against Pseudomonas aeruginosa, and Staphylococcus aureus was tested in the nematode host Caenorhabditis elegans, wherein the extract prepared by MAE had better anti-infective activity. Therefore, the extract prepared by MAE was compared with its hydroalcoholic extract prepared using the same method. Extraction efficiency obtained for the latter case was 2.0%.

Results

GLE prepared by three different methods were compared, at three different concentrations, for their anti-infective activity against P. aeruginosa and S. aureus (Figure 1; Dataset 1: Underlying data). At 50 µg/mL, GLE prepared by MAE proved superior to that prepared by decoction or VAE method, with respect to its ability to attenuate P. aeruginosa’s virulence towards C. elegans. At 0.5 µg/mL, extract prepared by decoction method registered least activity against this bacterium. At the same concentration, against S. aureus, extract prepared by VAE displayed the least activity. Based on these results, we concluded MAE as a better extraction method, and then extracted guava leaves using this method in water as well as water:alcohol (1:1) mixture. Both of these extracts prepared using MAE were then assayed for their anti-infective potential against five different pathogenic bacteria.

Figure 1.

Comparison of in vivo anti-infective efficacy of P. guajava leaf extracts prepared by three different extraction methods, against P. aeruginosa (A–C), and S. aureus (D–F). Catechin (50 μg/mL) and gentamicin (0.1 μg/mL) employed as positive controls conferred 100% and 80% protection on the worm population, respectively. DMSO present in the ‘vehicle control’ at 0.5%v/v did not affect virulence of the bacterium towards C. elegans. DMSO (0.5%v/v) and GLE at tested concentrations showed no toxicity towards C. elegans. MAE: Microwave Assisted Extraction; VAE: Vacuum Assisted Extraction; GLE: Guava Leaf Extract.

Both water as well as the hydroalcoholic extract of guava leaves could attenuate virulence of all the test pathogens (except S. pyogenes) towards C. elegans (Figure 2; Dataset 1: Underlying data). Both these extracts exhibited statistically similar anti-pathogenic efficacy against all susceptible bacteria, but the hydroalcoholic extract exhibited 10-15% better activity against S. aureus than the water extract. Despite the lowest extraction yield among all extracts reported in this study, the hydroalcoholic GLE was found to possess the highest (at par with water extract against all gram-negative pathogens) anti-pathogenic activity. Critical importance of choice of most appropriate extraction method and solvent for preparation of bioactive extracts has earlier been also emphasized by us (Gupta et al., 2012; Kothari et al., 2012), and others (Ngo et al., 2017; Sasidharan et al., 2011).

Figure 2. Comparison of the in vivo anti-infective potential of water extract and hydroalcoholic extract of P. guajava leaf extracts prepared by Microwave Assisted Extraction method, against five different pathogenic bacteria.

Figures A–E shows data against P. aeruginosa, S. aureus, S. marcescens, C. violaceum and S. pyogenes respectively. Catechin (50 μg/mL) employed as a positive control conferred 100% protection on worm population against all the pathogenic bacteria except S. pyogenes. Against S. pyogenes, catechin could not offer any protection to host worms. Gentamicin (0.1 μg/mL) allowed survival of worm population to the extent of 80% in face of P. aeruginosa, S. aureus, or S. pyogenes challenge; and 100% against the remaining two pathogens. DMSO present in the ‘vehicle control’ at 0.5%v/v did not affect virulence of the bacteria towards C. elegans. DMSO (0.5%v/v) and GLE at tested concentrations showed no toxicity towards C. elegans.

To have some insight into the mode of action of GLE, we incubated all the five test bacteria with GLE to investigate whether it affects bacterial growth and/or quorum-sensing (QS) regulated pigment production (a marker trait). Bacterial cell density and pigment production were quantified as earlier described by us (Joshi et al., 2016; Patel et al., 2018). Pigment production in all the four pigmented bacteria was modulated at ≥1 concentration(s) of the GLE tested. (Figure 3; Dataset 1: Underlying data). Since this extract did not inhibit bacterial growth heavily, it can be expected to exert lesser selection pressure (as opposed to potent bactericidal agents) on susceptible bacterial populations, and may not induce rapid development of resistant phenotypes. Ability of GLE to interfere with bacterial QS is an important observation, as QS in recent years has emerged as a potential target for novel anti-pathogenic agents (Fong et al., 2018). These ‘pathoblockers’ may attenuate virulence of the target pathogens without necessarily killing them (Kamal et al., 2017).

Figure 3. Effect of hydroalcoholic extract of P. guajava leaves prepared by Microwave Assisted Extraction method on bacterial growth and QS-regulated pigment production.

(A) P. aeruginosa (B) S. aureus (C) S. marcescens (D) C. violaceum (E) S. pyogenes. Bacterial cell density and pigment production were quantified as earlier described by us (Joshi et al., 2016). Bacterial growth was measured as OD₇₆₄ for the four pigmented bacteria, while for S. pyogenes OD₆₆₀ was used. OD of pyoverdine was measured at 405 nm, and that of pyocyanin at 520 nm; Pyoverdine Unit was calculated as the ratio OD₄₀₅/OD₇₆₄ (an indication of pyoverdine production per unit of growth); Pyocyanin Unit was calculated as the ratio OD₅₂₀/OD₇₆₄ (an indication of pyocyanin production per unit of growth. OD of staphyloxanthin was measured at 450 nm, and Staphyloxanthin Unit was calculated as the ratio OD₄₅₀/OD₇₆₄ (an indication of staphyloxanthin production per unit of growth). OD of prodigiosin was measured at 535 nm, and Prodigiosin Unit was calculated as the ratio OD₅₃₅/OD₇₆₄ (an indication of prodigiosin production per unit of growth). OD of violacein was measured at 585 nm, and Violacein Unit was calculated as the ratio OD₅₈₅/OD₇₆₄ (an indication of violacein production per unit of growth). QS: Quorum sensing

Conclusion

Results of the present study validate the traditional use of guava leaves for medicinal purposes and suggests one of the possible mechanisms through which it exerts its anti-infective activity, i.e. its ability to interfere with the bacterial QS machinery. Further investigation regarding GLE’s effect on pathogenic bacteria at the whole transcriptome level is warranted to unravel the molecular mechanisms underlying its anti-pathogenic efficacy.

Data availability