The Small RNA MicC Downregulates hilD Translation To Control the Salmonella Pathogenicity Island 1 Type III Secretion System in Salmonella enterica Serovar Typhimurium

Abstract

Salmonella enterica serovar Typhimurium invades the intestinal epithelium and induces inflammatory diarrhea using the Salmonella pathogenicity island 1 (SPI1) type III secretion system (T3SS). Expression of the SPI1 T3SS is controlled by three AraC-like regulators, HilD, HilC, and RtsA, which form a feed-forward regulatory loop that leads to activation of hilA, encoding the main transcriptional regulator of the T3SS structural genes. This complex system is affected by numerous regulatory proteins and environmental signals, many of which act at the level of hilD mRNA translation or HilD protein function. Here, we show that the sRNA MicC blocks translation of the hilD mRNA by base pairing near the ribosome binding site. MicC does not induce degradation of the hilD message. Our data indicate that micC is transcriptionally activated by SlyA, and SlyA feeds into the SPI1 regulatory network solely through MicC. Transcription of micC is negatively regulated by the OmpR/EnvZ two-component system, but this regulation is dependent on SlyA. OmpR/EnvZ control SPI1 expression partially through MicC but also affect expression through other pathways, including an EnvZ-dependent, OmpR-independent mechanism. MicC-mediated regulation plays a role during infection, as evidenced by an SPI1 T3SS-dependent increase in Salmonella fitness in the intestine in the micC deletion mutant. These results further elucidate the complex regulatory network controlling SPI1 expression and add to the list of sRNAs that control this primary virulence factor. IMPORTANCE The Salmonella pathogenicity island 1 (SPI1) type III secretion system (T3SS) is the primary virulence factor required for causing intestinal disease and initiating systemic infection. The system is regulated in response to a large variety of environmental and physiological factors such that the T3SS is expressed at only the appropriate time and place in the host during infection. Here, we show how the sRNA MicC affects expression of the system. This work adds to our detailed mechanistic studies aimed at a complete understanding of the regulatory circuit.

Keywords: EnvZ; MicC; OmpR; SPI1; Salmonella; SlyA.

FIG 1

Simplified model of the SPI1 T3SS regulatory circuit. Arrows indicate positive regulation, blunt ends indicate negative regulation, blue lines indicate transcriptional regulation, and green lines indicate posttranscriptional regulation.

FIG 2

The conserved small RNA MicC negatively regulates the SPI1 T3SS by repressing hilD translation in Salmonella. (A) Alignment of the MicC sequences from various Enterobacteriaceae. The asterisks indicate sequence identity. Sequences corresponding to the regions of MicC that base pair to ompC, ompD, and hilD are underlined. (B) β-Gal (β-galactosidase) activity in Salmonella (SM) strains harboring a hilD′-′lacZ translational fusion or a hilA′-lacZ⁺ transcriptional fusion, or an E. coli (EC) strain harboring a hilD′-′lacZ translational fusion under the control of an arabinose-inducible promoter. Each background contains the pBR-plac vector or pMicC plasmid. β-Gal activity units are defined as (μmol of ONP formed min⁻¹) × 10⁶/(OD₆₀₀ × ml of cell suspension). Results are shown as the median with interquartile range, and asterisks indicate significant differences between the data sets (n ≥ 4, P < 0.05, using a Mann-Whitney test). The strains used were JS749, JS892, and JMS6500, with plasmids pBR-plac vector or pMicC.

FIG 3

MicC negatively regulates hilD translation by base pairing near the RBS of the hilD mRNA. (A) Predicted base-pairing interaction between MicC and hilD mRNA. The RBS is underlined; boxes represent the nucleotides changed in the complementary mutations, with changes indicated in bold. (B) β-Gal activity in the Salmonella (SM) strain harboring the wild-type hilD′-′lacZ translational fusion with vector pBR-plac, wild-type pMicC, or mutated pMicC-mt plasmid. (C) β-Gal activity in E. coli (EC) strains harboring either the wild-type or mutated hilD′-′lacZ translational fusion with empty vector, wild-type pMicC, or mutated pMicC-mt plasmid. Results are shown as the median with interquartile range. Asterisks indicate significant differences between the data sets (n ≥ 4, P < 0.05, using a Mann-Whitney test). The strains used were JS892, JMS6500, and JMS6510, with plasmids pBR-plac vector, pMicC, or pMicC-mt.

FIG 4

MicC does not regulate HilC, RtsA, or HilA. (A and B) Relative β-gal activity in Salmonella hilC or hilA transcriptional fusion strains that are hilD⁺ (A) or hilD-null (B). (C) Relative β-gal activity in E. coli hilC′-′lacZ, rtsA′-′lacZ, or hilA′-′lacZ translational fusion strains. All strains include either pBR-plac vector or pMicC plasmid. Results are normalized to each strain containing the vector and are shown as the median with interquartile range. Asterisks indicate significant differences between the data sets (n = 4, P < 0.05, using a Mann-Whitney test). The strains used were JS2187, JS2196, JS2551, JS2552, JMS6503, JMS6504, and JMS6505, with plasmids pBR-plac vector or pMicC.

FIG 5

MicC inhibits translation of the hilD mRNA but does not induce degradation. (A and B) β-Gal activity in Salmonella hfq (A) or rne131 (B) strains harboring the wild-type hilD′-′lacZ translational fusion with vector pBR-plac or wild-type pMicC. Results are shown as the median with interquartile range, and asterisks indicate significant differences between the data sets (n = 4, P < 0.05, using a Mann-Whitney test). (C) Northern analysis of hilD mRNA levels. Salmonella strains were grown in SPI1 inducing conditions for 3 h. RNA was isolated at various time points after addition of rifampicin (Rif). The blot was probed for hilD (top panel), and after stripping, probed again for 5S RNA (bottom panel). (D) The intensities of the full-length hilD mRNA bands were quantified and normalized to the 5S bands. The vector or pMicC bands at the zero time point were considered to be 100%, and mRNA decay curves were created based on band intensities. The strains used were 14028, JS2118, and JS2119, with plasmids pBR-plac vector or pMicC.

FIG 6

SlyA and EnvZ/OmpR regulate micC and the SPI1 T3SS. β-Gal activity in Salmonella strains with a micC′-lacZ⁺ transcriptional fusion (A), hilD′-′lacZ translational fusion (B), or hilA′-lacZ⁺ transcriptional fusion (C) in backgrounds containing the indicated mutations. The results are shown as the median with interquartile range, and asterisks indicate significant differences between the data sets (n ≥ 4, P < 0.05, using a Kruskal-Wallis test followed by post hoc Dunn’s multiple comparisons). The strains used were JS749, JS892, JS2523, JS2524, JS2525, JS2526, JS2527, JS2528, JS2529, JS2530, JS2531, JS2532, JS2533, JS2534, JS2535, JS2536, JS2537, JS2538, JS2539, JS2540, JS2541, JS2542, JS2543, JS2544, JS2545, JS2546, JS2547, JS2548, JS2549, and JS2550.

FIG 7

Deletion of MicC provides a fitness advantage in vivo. Competitive index (CI) for in vitro and in vivo infections comparing the following strains: ΔmicC to wild type (WT) after oral infection (A), ΔmicC to WT after intraperitoneal (IP) infection (B), or ΔmicC Δspi1 to Δspi1 after oral infection (C). Upper small intestine (contains duodenum and jejunum), lower small intestine (contains ileum), and spleen were harvested after oral infections, whereas only the spleen was harvested after IP infections. Each circle represents the CI from a single mouse. For in vitro competitions, n = 3; panel A, n = 4; panels B and C, n = 5. The horizontal bars indicate the median of each data set, and the asterisk indicates significant difference (P < 0.05) using a Mann-Whitney test. The strains used were JS135, JS2553, JS2554, and JS2555.