Comprehensive Identification of Meningococcal Genes and Small Noncoding RNAs Required for Host Cell Colonization

Abstract

Neisseria meningitidis is a leading cause of bacterial meningitis and septicemia, affecting infants and adults worldwide. N. meningitidis is also a common inhabitant of the human nasopharynx and, as such, is highly adapted to its niche. During bacteremia, N. meningitidis gains access to the blood compartment, where it adheres to endothelial cells of blood vessels and causes dramatic vascular damage. Colonization of the nasopharyngeal niche and communication with the different human cell types is a major issue of the N. meningitidis life cycle that is poorly understood. Here, highly saturated random transposon insertion libraries of N. meningitidis were engineered, and the fitness of mutations during routine growth and that of colonization of endothelial and epithelial cells in a flow device were assessed in a transposon insertion site sequencing (Tn-seq) analysis. This allowed the identification of genes essential for bacterial growth and genes specifically required for host cell colonization. In addition, after having identified the small noncoding RNAs (sRNAs) located in intergenic regions, the phenotypes associated with mutations in those sRNAs were defined. A total of 383 genes and 8 intergenic regions containing sRNA candidates were identified to be essential for growth, while 288 genes and 33 intergenic regions containing sRNA candidates were found to be specifically required for host cell colonization.

Importance: Meningococcal meningitis is a common cause of meningitis in infants and adults. Neisseria meningitidis (meningococcus) is also a commensal bacterium of the nasopharynx and is carried by 3 to 30% of healthy humans. Under some unknown circumstances, N. meningitidis is able to invade the bloodstream and cause either meningitis or a fatal septicemia known as purpura fulminans. The onset of symptoms is sudden, and death can follow within hours. Although many meningococcal virulence factors have been identified, the mechanisms that allow the bacterium to switch from the commensal to pathogen state remain unknown. Therefore, we used a Tn-seq strategy coupled to high-throughput DNA sequencing technologies to find genes for proteins used by N. meningitidis to specifically colonize epithelial cells and primary brain endothelial cells. We identified 383 genes and 8 intergenic regions containing sRNAs essential for growth and 288 genes and 33 intergenic regions containing sRNAs required specifically for host cell colonization.

FIG 1

Schematic representation of the Tn-seq screen for identification of genes involved in epithelial and endothelial cell colonization. (A) N. meningitidis mutant libraries were constructed via random insertion of a transposon carrying a kanamycin resistance cassette. Mutants were pooled from a 2-h outgrowth culture and stored at −80°C. (B) Aliquots of the Tn mutants bank cultures were thawed and grown until mid-log phase (the input pools); input pools were used to infect an epithelial or endothelial cell monolayer in parallel on a microslide (ibidi, Germany) and recovered 18 h later (output pools). The presence of each mutant was investigated in both input and output pools by Tn-seq analysis. (C) Density plot of log₂ fold changes of measured read counts versus expected read counts. The log₂-transformed ratio of measured over expected read counts of insertion sites was used to generate a kernel density plot with a Gaussian model with stepwise increasing bandwidth and 2,048 bins, until a single local minimum was found between the two major distributions. The local minimum was detected by calculating the first derivative of the density and by locating the position where it traversed from values below zero to values above zero. This fold change value corresponds to a value closest to the minimum between the peaks of essential and nonessential genes and was used as a cutoff to determine whether a mutant was growth defective or not. A second cutoff of −5 was chosen arbitrarily as a cutoff for gene essentiality, as this value represents the left-most part of the results with growth-defective mutants. At this cutoff, the possibility of a false positive is negligible.

FIG 2

Characterization of the transposon mutant libraries. (A) Distribution of unique TIS among the 3 input libraries along the sequenced genome of N. meningitidis Z2491. (B) The number of transposon-containing reads per ORF in the input pool, in relation to gene length, is shown for essential genes, growth-defective genes, and non-growth-defective genes. Essentiality was determined based on the ratio of the number of reads per gene divided by the expected number of reads per gene, which depends on the size of the gene and the number of transposon insertion sites. Thus, a large gene with few transposon insertion sites may not be essential, even though few reads are mapped within that gene. (C) Distribution of the CDS containing random Tn insertions on the genome map of N. meningitidis Z2491 and of the sRNAs identified by RNA-seq. In the outer circles, bars indicate N. meningitidis Z2491 CDS. In the inner circles, bars indicate the essential genes, the nonessential but growth-defective genes, and the non-growth-defective genes in routine cell culture medium (according to the color key shown in the figure); the bars on the innermost circle indicate the location of the 390 identified sRNAs.

FIG 3

Schematic representation of the main metabolic pathways of essential genes for routine growth of N. meningitidis. The main essential genes for growth on GCB agar plates and CCM are highlighted in red, and the genes that encode proteins that cause a severe growth defect are highlighted in orange. *1, essential genes petA, -B, and -C; *2, essential genes ccoO and -N; *3, essential genes atpA, -B, -D, -F, and -G and growth-defective genes atpE and -H.

FIG 4

Identification of conditional essential genes necessary for colonization of epithelial and endothelial cells. The Venn diagram shows the absolute number of necessary ORFs (log₂ FC < −1.4; adjusted P < 0.05) and beneficial ORFS (log₂ FC > 1.4; adjusted P value < 0.05). (ORFs were selected in both the epithelial cell and endothelial cell in vitro models.)

FIG 5

Schematic representation of the main metabolic pathways of conditional essential genes in N. meningitidis necessary for colonization of both epithelial and endothelial cells. The illustration shows the main metabolism pathways in proliferating cells, including glycolysis, TCA cycle, PP, KDPG pathways and the synthesis of nucleotides, lipids, and amino acids. Black bold arrows indicate the metabolism reorientation, whereas red bold arrows indicate the main biosynthesis pathways important during colonization.