It has been a while since the study of natural products against microorganisms has gained interest by researchers and pharmaceutical industry. The bacterial resistance and side effects of the antimicrobial drugs available in the market have contributed to this fact (Ramakrishna  et   al.,   2011).  Moreover,  the  ecological awareness, traditional knowledge appreciation, and lack of access of marginalized communities to pharmaceutical medicines make the rational search for new compounds still important (Rates, 2001; Halberstein, 2005; Pelkonen et al., 2014).

Tooth  decay  has  been  considered  one  of  the  most prevalent disease in the world (Petersen, 2003)being S. mutans one of the greatest risk factor for its development (Nishikawara, 2006).Trying to solve the oral diseases problem, antibacterial effect study of natural products or their derivatives has been conducted (Kouidhi et al., 2015).

Two main techniques have been used in natural products against bacteria research: The agar diffusion (Kirby Bauer) technique, considered a qualitative method, and the broth dilution technique which is defined as quantitative or semi-quantitative method depending on the objective of the studies (Valgas et al., 2007). Both techniques have been used to pre-select plant species, initial concentration to be used, extraction solvents, or isolated molecules giving support to more specific methods related to mode of action of antimicrobial substances (Balouiri et al., 2016). Although the broth dilution is considered the gold test for antimicrobial resistance profile (Campana et al., 2011), agar diffusion has an important role in detection of subpopulation of bacteria with resistance to antimicrobials (Matthew, 2015) and it is considered an adequate and simple tool to evaluate bacterial resistance in clinical diagnosis (Schumacher et al., 2018). Therefore, the purpose of this paper is to review and compare different methodologies used to verify the antimicrobial effect of medicinal plants and their derivatives on S. mutans.

 

The study is characterized by a systematic review including peer reviewed articles published in international journals using as index tool the “Periodicos CAPES” (http://www.periodicos.capes.gov.br/index.php) which gives integrated access to international indexers such as PubMed, Web of Science, LILACS, SciELO, and SCOPUS. The criteria used to select the articles included: (1) “Plant extract,” “Streptococcus mutans,” and “medicinal plants” as key words; (2) English, Spanish, and Portuguese languages; (3) published between January, 1^st 2006 and December, 31^th 2016. Following selection, articles were classified by plant taxonomical  classification  (family  and  species), methodology approach (agar diffusion or broth dilution techniques), extract or substance unit measure (mg, %, etc.) used to standardize samples, solvents used for plant substances extraction such as apolar (hexane, benzene, chlorophorm, butane, and petroleum ether), aprotic polar (dichloromethane, ethyl acetate, and acetone/propane) and protic polar (ethanol, butane, methanol, and water), and isolated substances with anti-S. mutans activity. The criterion used to consider a “good result” for agar diffusion technique was an inhibition halo of ³ 18 mm since from the articles selected this diametre seemed to reflect sweetable concentrations in broth dilution. In the case of broth dilution, the minimal inhibitory concentration (MIC) values ≤100 µg/mL (Cos et al., 2006) and the ratio between minimal bacterial concentrations (MBC) and MIC (MBC/MIC) ≥16 were considered “good results.” MBC is ≥16 times the MIC value means that the microorganism is tolerant following criteria analyzed by Sherris (1986). Articles with description of antibacterial techniques were described but no results were excluded from the analysis. Data systemizing and graph building were done using Microsoft Excel 2010 software and results were expressed in absolute values and frequencies.

 

Through refined search using the keywords “plant extract,” “Streptococcus mutans,” and “medicinal plants,” a total of 129 available articles were found. According to selection criteria, 28 articles were selected to this review giving 135 plant species classified in 60 botanical families tested against S. mutans. Asteraceae, Fabaceae, Lamiaceae, and Anacardiaceae families were the most frequent (Figure 1 and Table 1) and 30 plant families had relevant results by agar diffusion, MBC/MIC, or both (Table 2). Among them, Rhus standley (Anacardiaceae), Amphipterygium adstringens (Anacardiaceae), Aloe vera (Asphodelaceae), Mikamia glomerate (Asteraceae), Tagetes lucida (Asteraceae), Bixa orellana (Bixaceae), Bursera simaruba (Burseraceae), Drymariagracilis (Carophyllaceae), Cnidoscolus multilobus (Euphorbiaceae), Glyycyrrhiza uralensis (Fabaceae), Liquidambar macrophylla (Hamamelidaceae), Cinamomum vera (Lauraceae), C. zeylanicum (Lauraceae), Persea americana      (Lauraceae),    Eysenhardti    apolystachya (Leguminosae), Haematoxylon brasiletto (Leguminosae), Cedrela odorata (Meliceae), Myrcia riadubia (Myrtaceae), Syzygium aromaticum (Myrtaceae), Argemone mexicana (Papaveraceae), Piper sanctum (Piperaceae), Punica granatum (Punicaceae), and Datura stramonium (Solanaceae) results have met international journals standard requirements. It is worth noting that some “good results” reported for agar diffusion was not correlated to a “good result” in MBC/MIC.

 

 

 

 

Natural products scientific experts have been standardizing parametres to stablish sweatable antimicrobial techniques for plants extracts or substances investigation. In this case, MIC and MBC have been the most recommended strategies (Ríos and Recio, 2005; Cos et al., 2006). From the literature review, only 19 articles met the established criteria here. Results reported involved 135 plants using qualitative techniques such as agar diffusion (20%), quantitative techniques (MBC and/or MIC) (43.69%), and a combination of both (36.28%) (Figure 2). The fact that almost 80% of the results were from quantitative techniques indicates the tendency to improve the quality of results following high scientific impact journals requirements. Regarding the results reported by several of studies analyzed, a combination of both seemed to be the best choice. In this case, in the researchers’ opinion, researchers can associate a technique still used in clinical diagnosis (Kirby-Bauer) to a more accurate quantitative procedure: The qualitative method to screen plant extract and the quantitative method to establish extract concentrations.

 

 

A point of concern on medicinal plant folk usage validation is the preparation and solubilization and/or dilution of plant material. The expression of units in the analyzed results in the articles was quite varied: mg/mL (54.52%) followed by percentage (23.67%). It is important to note that 15.95% used mg/disco, 4.78% of articles did not mention any amount of measurement, and 1.06% used proportion criterion (volume/volume or weight/volume). When the best results (≥18.5 mm inhibition zone) was considered, the ones expressed in mg/mL (63.33%) were the most promising when compared to results expressed in percentage (13.33%). Furthermore, analyzing the diferences in agar diffusion technique, the more frequent was mg/mL usage on DV (62.43%) and percentage unit on WV (45.08%) (Table 3). Hence, taking into consideration these aspects, two points which deserve attention were considered: (1) the standardization of international units requested by journal editors (Cos et al., 2006)recognizing that mg/mL is still the most accurate measurement for in vitro tests instead of percentage, and (2) percentage seems to better represent the folk method or it is the best choice for some plant material processing (that is, resin, pasty, or gelatinous substances) and it should not be neglected.

 

 

Taking into consideration the best plants results against S. mutans by agar diffusion technique, three vegetal products were distinguishable in their results (³ 30 mm inhibition zone): Aloe vera gel, Garcinialancifolia fruit juice, and Allium sativum bulbo juice. The Aloe vera gel at 50 and 100% showed inhibition zone of 30 and 54 mm, respectively, and antibacterial action was confirmed by promising results showed in MIC technique (12.5 μg/mL) (Fani and Kohanteb, 2012).The G. lancifolia fruit juice at 5 mg/mL gave 47 mm inhibition zone (Policegoudra et al., 2012)and the bulbo juice from A. sativum at 100 mg/mL showed 30 mm of inhibition zone (OHara et al., 2008). It is important to note that in the case of A. sativum (Jain et al., 2015), although the anti-S. mutans potential has been also confirmed by MIC (6,250 μg/mL) and MBC (12,500 μg/mL), the concentration showed in these two quantitative techniques are not in the parameters established by high impact publications since they use (Cos et al., 2006) reference which, unfortunately, makes uncertain the validation of these species.

From the articles selected (n=43) using broth dilution as technique for searching natural products with anti- S. mutans activity, few (n=17) showed MIC and MBC in association. A ratio between MBC/MIC (r) was obtained in order to evaluate the quality of antibacterial results (Table 4). A ratio ≥16, indicating bactericidal tolerance, was found only for three species ethanolic extracts: Cnidoscolus multilobus, Tournefortia hartwegiana, and Coreopsis mutica. However, since MIC was considered “good results,” in the researchers’ opinion, it would be interesting to isolate molecules with antimicrobial action from these species. Microbial tolerance may be influenced by bactericidal activity such as antagonism of molecules, technical factors, or microorganism characteristics (Sherris, 1986; Traczewski et al., 2009).

 

 

Also, considering ideal MIC values, it was observed that for MIC ≥ 100 μg/mL, MBC values were 3 times higher; while for MIC ≤ 100 μg/mL, MBC values were next to 6 times higher then MIC. From the study point of view, the MIC values were  higher  than  clinical  valuable concentrations  suggested  in  the  literature  (Cos  et  al., 2006)to reach a bactericidal effect and it means that the natural products tested had a tendency to be bacteriostatic against S. mutans.

Following the same criteria cited earlier (MIC £ 100 μg/mL and r=1) and establishing the best products reported in literature, Glycyrrhiza uralensis deglycyrrhizinated licorice root extract (Ahn et al., 2012), Mikania glomerate ent-Kaurenoic acid-rich extract (Moreira et al., 2016b), Ipomoea alba chloroform fraction (Silva et al., 2014), and Pistacia atlantica aqueous extract (Roozegar et al., 2016)showed the best results. Considering isolated substances and criteria established here, only saponin class showed important results. Saponins isolated from the seeds of Madhuca longifolia and Bauhinia purpurea were tested against two S. mutans strains and promising results were found for both (Jyothi and  Seshagiri, 2012)(Table 5).

 

 

The polarity of plant metabolites is also a point of concern since it can interfere on substances diffusion and/or  solubilization. Non-polar  or other samples difficult to diffuse in the medium, should be avoided in diffusion methods (Valgas et al., 2007). In this review the most frequent category used was polar solvent but it is worth noting that a combination of solvent or crude extracts obtained directly from the plant without solvent addition was also described. By agar diffusion technique, from non-polar solvent extraction, only two showed ³ 18 mm inhibition zone: The hexanic extract from bark of Cinamonum verum (OHara et al., 2008)and the hexanic extract from bark/leaves of Schinuster ebinthifolius (Pereira et al., 2011).

Both articles reported the use of dimetylsulfoxide as solvent to solubilize the powdered extracts. Thus, although both hexanic extracts were active at higher than clinically acceptable concentrations, the results published by Ohara et al. (2008)  and Pereira et al. (2011) suggested that solvent such as dimetylsulfoxide can help some molecules to diffuse into agar medium and make possible the use of agar diffusion technique for apolar substances as suggested by Valgas et al. (2007).

The antibacterial property of a variety of natural products is documented; nonetheless, the great variability of secondary metabolites composition makes the studies in this area always laborious.For anti-S. mutans, it seems there is a restriction in this variety of substances. From the peer reviewed sources used in this work, the only pure substance report was saponins extracted from four species: Madhuca longifolia, Bauhinia purpurea, Celastrus paniculatus, and Semecarpus anacardium

by Jyothi and Seshagiri (2012)with promising results only for saponins isolated from M. longifolia. The antimicrobial broad spectrum of saponins has been reported (Avato et al., 2006; Qin et al., 2016), however the amphiphilic characteristics of this class, with the diversity which it carries, plays an important role in the antimicrobial property itself or in its solubilization in aqueous based medium used in the techniques employed were considered.

 

 

 

 

 

 

 

 

 

The present systematic review showed a  comprehensive examination of medicinal plants under the perspective of anti-S. mutans activity. Results showed that agar diffusion technique is still widely used for medicinal plants antimicrobial activity screening being either well-variant or disc-variant worthwhile as screening tests. The remarkable results from A. vera, G. lancifolia and A. sativum by agar diffusion or from G. uralensis, M. glomerata, I. alba, and P. atlantica by MIC/MBC show that although taxonomic criterion may be considered a leader for antimicrobial activity, ethnobotanical criterion should also be considered an excellent guide for in vitro studies. The unit of measurement used (µg/mL) is scientifically considered more accurate; however, it is worth noting that percentage as unit of measurement warrants validation since it mimics the popular usage of plants. Finally, the standardization of antimicrobial protocols for medicinal plants antimicrobial tests is needed in order to obtain more accurate results and make the comparision between natural products and controls easier.

 

The authors have not declared any conflict of interests.

 

We would like to thank the Universidade Estadual de Santa Cruz for structural support.