Lactobacillus crispatus-loaded electrospun fibers yield viable and metabolically active bacteria that kill Gardnerella in vitro

Abstract

Bacterial vaginosis (BV) is a common condition that affects one-third of women worldwide. BV is characterized by low levels of healthy lactobacilli and an overgrowth of common anaerobes such as Gardnerella. Antibiotics for BV are administered orally or vaginally; however, approximately half of those treated will experience recurrence within 6 months. Lactobacillus crispatus present at high levels has been associated with positive health outcomes. To address the high recurrence rates following BV treatment, beneficial bacteria have been considered as an alternative or adjunct modality. This study aimed to establish proof-of-concept for a new long-acting delivery vehicle for L. crispatus. Here, it is shown that polyethylene oxide (PEO) fibers loaded with L. crispatus can be electrospun with poly(lactic-co-glycolic acid) (PLGA) fibers (ratio 1:1), and that this construct later releases L. crispatus as metabolically viable bacteria capable of lactic acid production and anti-Gardnerella activity. Probiotic-containing fibers were serially cultured in MRS (deMan, Rogosa, Sharpe) broth with daily media replacement and found to yield viable L. crispatus for at least 7 days. Lactic acid levels and corresponding pH values generally corresponded with levels of L. crispatus cultured from the fibers and strongly support the conclusion that fibers yield viable L. crispatus that is metabolically active. Cultures of L. crispatus-loaded fibers limited the growth of Gardnerella in a dilution-dependent manner during in vitro assays in the presence of cultured vaginal epithelial cells, demonstrating bactericidal potential. Exposure of VK2/E6E7 cells to L. crispatus-loaded fibers resulted in minimal loss of viability relative to untreated cells. Altogether, these data provide proof-of-concept for electrospun fibers as a candidate delivery vehicle for application of vaginal probiotics in a long-acting form.

Keywords: Bacterial vaginosis; Electrospinning; Gardnerella; Lactobacillus crispatus; Nanofibers; Poly(lactic-co-glycolic acid); Polyethylene oxide.

Figure 1.

Scanning electron microscopy (SEM) images of electrospun nanofibers. Left: blank fibers; right: L. crispatus-loaded mesh fibers composed of 1:1 PEO:PLGA. Bacterial incorporation is shown by presence of rod-shaped bacteria, highlighted in yellow. Scale bar = 5 μm. Both fibers were imaged on the side external to the mandrel, enabling comparison of uniform morphologies.

Figure 2.

CFU of L. crispatus cultured from mesh electrospun fibers daily over one week. (A) The probiotic-containing mesh fiber demonstrated cumulative viable recovery of L. crispatus (9.65×10⁶ CFU/mg) by 24 hr. (B) L. crispatus could be cultured at a similar or higher level from probiotic-loaded fibers daily for six consecutive days. In contrast, no recovery was obtained from blank fibers (open circles). The mean ± standard deviation of CFUs derived from cultured fibers are shown for three independent fiber batches, with each batch having three technical replicates.

Figure 3.

pH and lactic acid in supernatant from fibers cultured in MRS is shown as the mean ± standard deviation of three independent fiber batches, with each batch having three technical replicates. (A) The pH of fiber cultures reached its lowest point by 4d and the remaining fiber material continued to yield acidified cultures of Lactobacillus until at least day 7. (B) Amount of lactic acid was proportional to the observed recovery of L. crispatus upon culture of fibers in MRS medium.

Figure 4.

Gardnerella viability when exposed to electrospun-fiber-derived L. crispatus. Effect of 24 hr and 7 day culture supernatants (at varying dilutions resulting from incubation of fibers in MRS media) after 24 hr incubation in the presence of VK2 cells (A) defensively primed or (B) not defensively primed for protection against infection (as explained in Methods). The log (CFU/mL) values were determined from mean ± standard deviation of cultures from three independent L. crispatus-loaded fibers. (****p ≤ 0.0001)

Figure 5.

Evaluation of cytotoxicity of L. crispatus-loaded fibers on vaginal keratinocytes (VK2/E6E7 cells). (A) Negligible cytotoxicity (cell viability as percent of untreated cells) was observed in VK2/E6E7 cells using MTT assay. (B) No significant release of LDH was observed from VK2/E6E7 cells exposed to blank or L. crispatus-loaded fibers for 24 hr, relative to untreated cells. In contrast, cells treated with staurosporine showed significantly elevated LDH compared to untreated cells. Data represent mean ± standard deviation (n=5). Statistical significance between experimental groups, as calculated by one-way ANOVA (Tukey test), is represented by ****p ≤ 0.0001.