INTRODUCTION

Vaginal infections cause more than 10 million healthcare visits annually, mainly due to bacterial vaginosis, candidiasis, and trichomoniasis ^{1 , 2}. Bacterial vaginosis (BV) is the most common vaginal infection among reproductive-aged women, and it is characterized by the reduction or replacement of lactobacilli and an increase of opportunistic anaerobic bacteria ^(3-5). Members of the genus Lactobacillus, such as L. crispatus, L. jensenii and L. iners, are distinctive markers of a healthy vaginal microbiome, while Gardnerella vaginalis, Prevotella, Atopobium, Mobiluncus and Peptostreptococcus are frequent BV-associated anaerobic bacteria ^{3 , 6}.

Oral metronidazole or vaginal clindamycin are routinely used as first-line treatment options for BV, however, several pathogenic bacteria, such as Mobiluncus and Atopobium species, are commonly resistant to those antimicrobials ^(3,7-10). The intermittent and long-term use of metronidazole or clindamycin, and infections with resistant bacteria, often result in high recurrence rates of BV within the first 12 months of treatment ^(10-12). Additionally, the increase of antimicrobial resistance in anaerobic bacteria after treatment has been also reported ^{2 , 7 , 8 , 13}.

Despite the increased antimicrobial resistance and high relapse rates, the development of new drugs has been scarce, so reliance on available treatments remains necessary ^{11 , 14}. Plant-derived compounds represent a potential source of therapeutic options for a variety of bacterial infections due to their reduced cost and toxicity, as well as their reduced risk of side effects ^{15 , 16}. Bixa orellana L. is known in traditional medicine for its pharmacological properties, antimicrobial activity, and reduced toxicity ^(17-19). Thus, B. orellana constitutes a source of molecules with promising therapeutic potential that could substitute synthetic drugs used to treat BV given its antimicrobial effectiveness ¹⁹. The aim of the study was to evaluate the in vitro antimicrobial activity of B. orellana against BV-associated anaerobic bacteria and Lactobacillus spp. strains.

MATERIALS AND METHODS

RESULTS

The 40% DMSO, regardless of the volume used, had no bactericidal effect neither inhibited the bacterial growth of the ATCC reference strains used. Figure 1 (left side of the plates) shows the results for some ATCC reference strains.

Figure 1 Antimicrobial susceptibility testing using 40% DMSO and B. orellana extract against anaerobic bacteria; a) Prevotella bivia ATCC 29303, b) Veillonella parvula ATCC 10790, c) Atopobium vaginae ATCC BAA-55, d) Mobiluncus curtisii ATCC 43063, e) Peptoestreptococcus anaerobius ATCC 27337, and f) Peptococcus niger ATCC 27731. Sterile paper disks and wells on the left side of the plates were used for testing the activity of 40% DMSO, and those on the right side were used for testing the extract. 

Antimicrobial activity against ATCC reference strains

The inhibition zone was directly related to the volume of extract used (Table 1). The extract exhibited high antimicrobial activity against all ATCC reference strains, except for L. crispatus ATCC 33197, P. bivia ATCC 29303, and V. parvula ATCC 10790 (Table 1). Interestingly, the growth of L. crispatus ATCC 33197 was not inhibited regardless of the volume used, and the growth of V. parvula ATCC 10790 was only inhibited by 100 µL of the extract.

The MIC and MBC for ATCC stains are reported in Table 1. G. vaginalis ATCC 14018 was the most susceptible bacteria, with MIC and MBC of 1.0 mg/mL and 4.0 mg/mL, respectively. P. bivia ATCC 29303, V. parvula ATCC 10790 and L. crispatus ATCC 33197 were the most resistant bacteria, displaying higher MIC and MBC values compared to other ATCC strains.

Antimicrobial activity against clinical isolates

Among G. vaginalis clinical isolates, the most susceptible was strain M10 (MIC and MBC: 1.0 mg/mL), followed by strains MD (MIC: 1.0 mg/mL; MBC: 4.0 mg/mL) and M23 (MIC and MBC: 2.0 mg/mL) while the other eight isolates displayed MIC and MBC values of 2.0 mg/mL and 4.0 mg/mL, respectively (Table 2). Overall, the MIC₅₀ and MIC₉₀ were 2 mg/mL, and the MBC₅₀ and MBC₉₀ were 4.0 mg/mL.

Lactobacillus spp. isolates grew abundantly over the range of 1.0 to 16.0 mg/mL of the extract. However, there was no evidence of bacterial growth at 32.0 mg/mL. Consequently, the MIC value for all clinical isolates of Lactobacillus spp. was 32.0 mg/mL (Table 2), and the MIC₅₀ and MIC₉₀ were 32.0 mg/mL. Regarding the MBC, 54.5% (6/11) of the Lactobacillus spp. isolates (M7, M24, M25, M33, MA and MB; Table 2) showed a value of 32.0 mg/mL, while other clinical isolates showed values above 32.0 mg/mL. The MBC₅₀ and the MBC₉₀ for Lactobacillus spp. were 32.0 mg/mL and >32.0 mg/mL, respectively.

The extract exhibited a significant and differential antimicrobial activity between G. vaginalis and Lactobacillus spp. strains (p<0.001). Specifically, G. vaginalis isolates were inhibited at low concentrations of the extract (MIC: 2.0 mg/mL, MBC: 4.0 mg/mL), while Lactobacillus spp. isolates were inhibited at higher concentrations (MIC: 32.0 mg/mL, MBC: 3.0 mg/mL).

DISCUSSION

In recent years, the use of medicinal plants or herbs has been promoted by scientific research due to their pharmacological activity, low toxicity, and inexpensive accessibility ^{15 , 16 , 19 , 25}. Plants possess and produce a wide variety of secondary metabolites such as tannins, terpenoids, flavonoids, glycosides, saponins, anthranoids, quinones and coumarins, to which antimicrobial properties have been attributed ²⁷. Several studies suggest that B. orellana possesses various properties, including antimicrobial, antifungal, antioxidant, anti-inflammatory, and analgesic activity ^{(19, 28)}. The antimicrobial activity of B. orellana against several microorganisms has been previously described ^{19 , 28}, however, its antimicrobial activity on microaerophilic or anaerobic bacteria responsible for BV has been scarcely described.

Our results suggest that, in the absence of bacteriostatic or bactericidal activity of the 40% DMSO, the observed antimicrobial activity can be attributed to the extract. This finding is consistent with several other studies that suggest that B. orellana leaves extracts, obtained by methanol- or ethanol-based techniques, have antimicrobial activity against reference ATCC strains and clinical isolates ^{19 , 29 , 30}. Therefore, based on our results and the results reported by other studies ^{19 , 28 - 30}, it is reasonable to consider that the evaluated extract possesses antimicrobial activity.

BV is a dysbiosis characterized by drastic changes in the biota of the vaginal tract that result in a replacement of the lactobacillus-predominant vaginal flora by anaerobic bacteria ^{3 , 6}. G. vaginalis is one of the most frequent anaerobic bacteria causing BV, which symbiotically can form a polymicrobial biofilm with several BV-associated anaerobic bacteria such as A. vaginae and Prevotella spp ^{2 , 6 , 31}. Other anaerobic bacteria commonly detected in women with BV are Peptoestreptococcus and Mobiluncus ^{6 , 32 , 33}. Based on the observed low MIC and MBC values, it can be inferred that the evaluated extract has a high antimicrobial activity against most of the ATCC reference strains used here and also against clinical isolates of G. vaginalis. However, it is important to note that P. bivia and V. parvula, both ATCC strains, displayed low susceptibility to the extract. To the best of our knowledge, the in vitro activity of the B. orellana extract against G. vaginalis and other anaerobic pathogens has been scarcely described. Further studies are needed to validate our findings in a wide variety of clinical isolates.

Vaginal lactobacilli regulate pH and protect the mucosa against the establishment of pathogenic microorganisms ^{3 , 33}. In this study, L. crispatus ATCC 33197 and clinical isolates of Lactobacillus spp. were used as references of the Lactobacillus complex found in the healthy human vagina ^{( 6 , 32}. Interestingly, the extract inhibited the growth of lactobacilli at concentrations of 32.0 mg/mL. However, our results differ from those described elsewhere. Galindo-Cuspinera et al. ³⁴ suggested that the extract obtained from fruits and seeds of B. orellana exhibited limited antimicrobial activity against L. lactis ATCC 11454 and L. casei ATCC 39539, and no activity against L. plantarum ATCC 700210. Similarly, Ogunshe et al. ³⁵ suggested that B. orellana had limited in vitro activity against various vaginal Lactobacillus spp. strains. Overall, these discrepant results could be explained by; i) the use of different extraction methods (i.e., type of extract and solvent) resulting in variable concentration of phytochemicals, ii) the use of different parts of B. orellana (i.e., leaves, seeds, pods or fruits) for extract preparation, iii) the geographical location and exposure to climatic and environmental factors affecting B. orellana, and iv) by intrinsic differences between Lactobacillus spp. strains used in other studies ^{( 19}.

Overall, our findings are consistent with several other studies suggesting that B. orellana extract possesses antimicrobial activity ^{(17-19,28,30)}. An important major strength of this in vitro study is the evaluation of the antimicrobial activity of the extract against anaerobic bacteria representative of those causing BV, and against representative bacteria of the healthy vaginal biota. Our findings may support the scale-up and innovation of research for the development of potential therapeutic options. However, it is also important to consider the limitations of this in vitro study. Multiple phytochemicals are found in B. orellana, including flavonoids, polyphenols, tannins, quinones, terpenoids and alkaloids, which exert antimicrobial activity through complex processes ¹⁹. In this preliminary study we have not characterized the phytochemical compounds or their concentrations; therefore, we could not establish nor isolate the individual antimicrobial activity of each component of the extract against the studied bacterial isolates. Furthermore, it is reasonable to assume that the antimicrobial activity described here may not be extrapolated to all B. orellana species given possible intrinsic differences among them and differences in plant cultivation.

In summary, our in vitro experiments suggest that the extract of B. orellana could be a source of antimicrobial compound or compounds with high selective antimicrobial activity against most BV-associated anaerobic bacteria, and with reduced activity against protective bacteria found in the healthy vaginal mucosa. Further research is needed to identify the compound or compounds with antimicrobial activity, in order to validate the preliminary differential activity described in our study by using a more rigorous and complex study design.