A comparative study of two Lactobacillus fermentum strains that show opposing effects on intestinal barrier integrity : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy, Massey University, Manawatu, New Zealand
The Lactobacillus species can exert health promoting effects in the gastrointestinal tract (GIT) of humans through several mechanisms, which include pathogen inhibition, maintenance of microbial balance, immunomodulation and enhancement of the GIT barrier function. However, different strains of lactobacilli can evoke different responses in the host and not all strains of the same species can be considered health promoting. Two strains identified as Lactobacillus fermentum, namely AGR1485 and AGR1487, isolated from human oral cavities, exhibit opposing effects on intestinal barrier integrity. Studies have shown that AGR1485 maintains trans-epithelial electric resistance (TEER), a measure of GIT barrier integrity, across Caco-2 cell monolayers, while AGR1487 decreases TEER by 12 hours.
This work aimed to test the hypotheses that the varying effects shown by these two L. fermentum strains are related to phenotypic differences between the two strains and are mediated by the interaction of secreted and/or cell-associated bacterial components with the GIT epithelial layer. Differences in metabolic events that occur during the various phases of growth in bacteria can impact not only cellular structure and secreted molecules, but may also affect their interactions with the intestinal epithelial cells. TEER assays were conducted to investigate if variation in bacterial secreted molecules and cell wall components associated with various phases of microbial growth can affect Caco-2 cell TEER. The effect on Caco-2 cell TEER caused by both strains was independent of bacterial growth phase. To test the hypothesis that it is the bacterial structural and/or secreted components that influence
Caco-2 cell TEER, assays were conducted with live versus UV-killed bacteria on Caco-2 cells. Results showed that for both strains of L. fermentum, dead bacteria have similar effects on Caco-2 cell TEER as live bacteria, implying that direct bacterial contact with Caco-2 cells is necessary for the effects. Analogous to TEER assays, live AGR1487 increased mannitol permeability while UV-killed AGR1487 did not, implying that AGR1487 uses both cell surface structures and/or metabolites through distinct mechanisms to modulate host barrier properties. Subsequent experiments conducted using secreted metabolites from bacteria, Caco-2 cells and bacteria-Caco-2 cell interactions indicated that they have no effect on Caco-2 cell TEER, strengthening the assumption that bacterial cell surface-associated components are involved in mediating these effects.
The bacterial cells were subjected to ultrasonication followed by ultracentrifugation to isolate the bacterial cell wall extract. TEER assays conducted with the cell wall extracts from both strains resulted in decreasing Caco-2 cell TEER, although at high concentrations, further strengthening the role of bacterial cell surface components in influencing barrier integrity of the Caco-2 cells. To narrow down proteinaceous components of the cell wall extracts from both the strains that influence Caco-2 cell TEER, they were fractionated through size exclusion chromatography and the effects of these cell wall fractions on Caco-2 cell TEER were studied. One fraction of AGR1487 CW appeared to decrease Caco-2 cell TEER, although at a high concentration. However, the results could not be repeated when the same fraction was applied at concentrations that the proteins comprising this fraction would be found in
live AGR1487. Even the high concentration tested previously did not decrease Caco-2 cell TEER and the discrepancy in results remains unexplained.
The results reported in this dissertation have added to the knowledge that the two strains of L. fermentum AGR1485 and AGR1487 show differences in their genome size and in their phenotypic characteristics. In addition, these bacteria utilise both cell surface and/or secreted metabolites through multiple mechanisms to modulate host response. In the future, identification of specific bacterial effector molecules that influence host response will be a major step towards understanding strain-specific characteristics shown by Lactobacilli.