Analysis of essential gene dynamics under antibiotic stress in Streptococcus sanguinis

Abstract

The paradoxical response of Streptococcus sanguinis to drugs prescribed for dental and clinical practices has complicated treatment guidelines and raised the need for further investigation. We conducted a high throughput study on concomitant transcriptome and proteome dynamics in a time course to assess S. sanguinis behaviour under a sub-inhibitory concentration of ampicillin. Temporal changes at the transcriptome and proteome level were monitored to cover essential genes and proteins over a physiological map of intricate pathways. Our findings revealed that translation was the functional category in S. sanguinis that was most enriched in essential proteins. Moreover, essential proteins in this category demonstrated the greatest conservation across 2774 bacterial proteomes, in comparison to other essential functional categories like cell wall biosynthesis and energy production. In comparison to non-essential proteins, essential proteins were less likely to contain 'degradation-prone' amino acids at their N-terminal position, suggesting a longer half-life. Despite the ampicillin-induced stress, the transcriptional up-regulation of amino acid-tRNA synthetases and proteomic elevation of amino acid biosynthesis enzymes favoured the enriched components of essential proteins revealing 'proteomic signatures' that can be used to bridge the genotype-phenotype gap of S. sanguinis under ampicillin stress. Furthermore, we identified a significant correlation between the levels of mRNA and protein for essential genes and detected essential protein-enriched pathways differentially regulated through a persistent stress response pattern at late time points. We propose that the current findings will help characterize a bacterial model to study the dynamics of essential genes and proteins under clinically relevant stress conditions.

Keywords: Streptococcus sanguinis; antibiotic stress; essential genes; proteomics; transcriptomics.

Fig. 1.

Bioinformatics analysis of functional categories and conservation of essential proteins. (a) Involvement of essential genes in S. sanguinis physiological pathways is shown in this bar chart. The number of essential genes (y-axis and number at the top of every chart bar) involved in the number of KEGG pathways (x-axis) is shown. Conservation of S. sanguinis essential (b) and non-essential (c) proteins across 2774 bacterial proteomes in relation to their COG annotations. Every dot on the scatter plots represents an (b) essential or (c) non-essential S. sanguinis protein. Proteins were clustered based on their functional categories as described by their COG annotations. Conservation of S. sanguinis proteins was determined by orthologues in 2774 bacterial proteomes. The average number of detected orthologues for every S. sanguinis protein in each COG category can be inferred from the y-axis projection of the mean (central horizontal bar) and the sem (vertical bar).

Fig. 2.

Differential expression of essential genes in antibiotic-treated S. sanguinis cells. (a) Circos plot representing the differential mRNA expression of essential genes at T₁₀, T₂₀, T₃₀ indicative of 10, 20 and 30 min respectively post-treatment with a sub-inhibitory dose of ampicillin in comparison to T₀ (untreated cells) in S. sanguinis SK36 strains. Green bars indicate a statistically significant up-regulation of gene transcription; red bars indicate a statistically significant down-regulation of gene transcription. Functional clustering was based on COG annotation, and further grouped into three essential functions as follows: G (green) for genetic information processing; C (blue) for cell wall biosynthesis; E (red) for energy production. (b) Bar chart showing the counts of up-regulated (green) and down-regulated (red) expression of essential genes at three time points.

Fig. 3.

Differential expression of essential proteins in antibiotic-treated S. sanguinis SK36 cells. (a) Circos plot representing the differential expression of essential proteins at T₁₀, T₂₀, T₃₀, indicative of 10, 20 and 30 min respectively post-treatment with a sub-inhibitory dose of ampicillin in comparison to T₀ (untreated cells) in S. sanguinis SK36 samples. (b) Bar chart showing the counts of up-regulated (green) and down-regulated (red) expression of essential proteins at three time points.

Fig. 4.

Determination of amino acids at the N-terminal positions in essential and non-essential proteins. (a) The localization of every amino acid in the predicted mature N-terminal position for essential and non-essential S. sanguinis proteins was counted and averaged using python scripts. For every amino acid, the difference between the composition percentage in essential versus non-essential proteins was tested for statistical significance. The six amino acids enclosed in a red square are the degradation-prone amino acids. EG, proteins encoded by essential genes; non EG, proteins encoded by non-essential genes. (b) Percentage of essential and non-essential proteins that possess a degradation-prone amino acid at their N-terminal position. *P-value<0.05; **P-value<0.001.

Fig. 5.

Amino acid composition of essential and non-essential proteins in S. sanguinis. Amino acid sequences were extracted from the NCBI database. Amino acid composition of essential and non-essential proteins was averaged from individual protein compositions.

Fig. 6.

Glycolysis pathway map showing differential expression of (a) essential genes and (b) essential proteins in S. sanguinis exposed to a sub-inhibitory dose of ampicillin for 20 min. The genes/proteins (circles) are size and colour-coded based on an intensity spectrum where a large green circle indicates up-regulation, a small red circle indicates down-regulation and a blank circle shows no significant detection. Non-essential genes/proteins are labelled with ‘P’ after gene/protein name.