Cyclopropane fatty acid synthase mutants of probiotic human-derived Lactobacillus reuteri are defective in TNF inhibition

Abstract

Although commensal microbes have been shown to modulate host immune responses, many of the bacterial factors that mediate immune regulation remain unidentified. Select strains of human-derived Lactobacillus reuteri synthesize immunomodulins that potently inhibit production of the inflammatory cytokine TNF. In this study, genetic and genomic approaches were used to identify and investigate L. reuteri genes required or human TNF immunomodulatory activity. Analysis of membrane fatty acids from multiple L. reuteri strains cultured in MRS medium showed that only TNF inhibitory strains produced the cyclopropane fatty acid (CFA) lactobacillic acid. The enzyme cyclopropane fatty acid synthase is required for synthesis of CFAs such as lactobacillic acid, therefore the cfa gene was inactivated and supernatants from the cfa mutant strain were assayed for TNF inhibitory activity. We found that supernatants from the wild-type strain, but not the cfa mutant, suppressed TNF production by activated THP-1 human monocytoid cells Although this suggested a direct role for lactobacillic acid in immunomodulation, purified lactobacillic acid did not suppress TNF at physiologically relevant concentrations. We further analyzed TNF inhibitory and TNF non-inhibitory strains under different growth conditions and found that lactobacillic acid production did not correlate with TNF inhibition. These results indicate that cfa indirectly contributed to L. reuter immunomodulatory activity and suggest that other mechanisms, such as decreased membrane fluidity or altered expression of immunomodulins, result in the loss of TNF inhibitory activity. By increasing our understanding of immunomodulation by probiotic species, beneficial microbes can be rationally selected to alleviate intestinal inflammation.

Figures

Figure 1

Synthesis of cyclopropane fatty acids in bacteria. Cyclopropane fatty acid synthase converts oleic acid and vaccenic acid into dihydrosterculic acid and lactobacillic acid, respectively.

Figure 2

FAME analysis of TNF inhibitory and TNF non-inhibitory strains. Five L. reuteri strains that differ in their ability to downregulate human TNF were grown to stationary phase in MRS and FAME analysis was performed. ATCC 55730 and CF48-3A are TNF non-inhibitory strains while ATCC PTA 6475, ATCC PTA 4659 and ATCC PTA 5289 are TNF inhibitory strains. The level of each fatty acid produced in the strains is indicated as a percentage of the total membrane fatty acids on the Y-axis. The identity of each fatty acid is indicated by color in the figure. * denotes lactobacillic acid.

Figure 3

The appearance of lactobacillic acid in late stationary phase cultures of L. reuteri ATCC PTA 6475 correlates with the appearance of TNF inhibitory activity by cell-free supernatants. Cultures of L. reuteri ATCC PTA 6475 were grown in MRS broth overnight at 37°C. Samples were collected for FAME analysis at different stages of growth. Experiments were performed in triplicate; error bars represent standard deviations. The relative quantities of each fatty acid produced in the strains are indicated as a percentage of the total membrane fatty acids on the Y-axis. The identity of each fatty acid is indicated by color in the figure.

Figure 4

L. reuteri ATCC PTA 6475 deficient in cyclopropane fatty acid production is unable to suppress human TNF production by activated monocytes. Cell-free supernatants from L. reuteri strains grown in (A) MRS medium or (B) LDMIII medium were tested for the ability to inhibit activated THP-1 human monocytoid cells from producing TNF. TNF production (pg/ml) was monitored by ELISA and the data were normalized to the medium control, which was set at 100%. Strains tested were L. reuteri ATCC PTA 6475, PRB173 (Δcfa) and L. reuteri DSM 17938. At least six independent experiments were performed and error bars indicate the standard deviation of the measurements. Using one-way ANOVA analysis, the relative quantities of TNF produced by wild-type L. reuteri ATCC PTA 6475 (in both MRS medium and LDMIII medium) were significantly different from that of medium control, PRB173 and L. reuteri DSM 17938 (p < 0.001). PRB173 and L. reuteri DSM 17938 were not significantly different than the medium control.

Figure 5

FAME analysis of TNF stimulatory and TNF inhibitory L. reuteri strains grown in LDMIII medium. L. reuteri cells that produced the cell free supernatants used in Figure 4B were subjected to FAME analysis. The percent of each fatty acid produced is indicated on the Y-axis. L. reuteri strains ATCC 55730, ATCC PTA 6475 and PRB173 (Δcfa) are depicted on the X-axis.