Transcriptome change in Staphylococcus aureus in infecting mice

We performed in vivo RNA-sequencing analysis of Staphylococcus aureus in infected 24 mouse liver using the 2-step cell-crush method. We compared the transcriptome of S. 25 aureus at 6, 24, and 48 h post-infection (h.p.i) in mice and in culture medium. Genes 26 related to anaerobic respiration were highly upregulated at 24 and 48 h.p.i. The gene 27 expression patterns of virulence factors differed depending on the type of toxin. For 28 example, hemolysins, but not leukotoxins and serine proteases, were highly upregulated 29 at 6 h.p.i. Gene expression of metal transporters, such as iron transporters, gradually 30 increased at 24 and 48 h.p.i. We also analyzed the transcriptome of mouse liver infected 31 with S. aureus . Hypoxia response genes were upregulated at 24 and 48 h.p.i., and immune 32 response genes were upregulated from 6 h.p.i. These findings suggest that gene 33 expression of S. aureus in the host changes in response to changes in the host environment, 34 such as oxygenation status or immune system attacks during infection. improved vivo technique in the present we In the present study, S. aureus gene expression changes in mouse systemic were analyzed over time by in vivo RNA-Seq. The results suggested that S. aureus responds to changes in oxygenation and environmental influences associated with deterioration of the host's cardiovascular status as the infection progresses, and obtains energy through anaerobic respiration. Further, expression of metal transporters did not increase until 6 h.p.i., but increased remarkedly after that time. Our results also revealed the contribution of the manganese transporter mntABC and staphylopine to the pathogenicity of the broad metal transport system for the first time. As for the expression of pathogenic toxins, the timing of the upregulation differed depending on the early stage proteases to in in the host during regulating gene


Introduction 37
The rapid emergence of multi-antimicrobial resistant strains has created an urgent need 38 for the development of novel therapeutic agents. The number of recent discoveries of 39 therapeutically active antimicrobials with novel mechanisms such as texobactin 1 and 40 lysocin E 2 is limited, however, suggesting a depletion of excellent target molecules to 41 develop as novel antimicrobials. Antimicrobials are typically screened on the basis of 42 their antimicrobial activity in vitro. As recently pointed out 3,4 , however, pathogen 43 behavior exhibited during in vitro culture is much different from that in vivo in the host; 44 thus, it is important to identify antimicrobial targets expected to be more efficient in the 45 host.

Pathway analysis of the genes altered after infection with S. aureus 108
To elucidate the trend of S. aureus gene expression in the host environment, we performed 109 KEGG pathway enrichment analysis to compare with gene expression analysis in the 110 culture medium (Table 1). We found that expression of genes involved in carbon 111 metabolism (glycolysis) and TCA cycle pathways was significantly upregulated at 6 h.p.i 112 (Table 1), but not at 48 h.p.i. Expression of genes involved in beta-oxidation, responsible 113 for the production of acetyl-CoA from fatty acids, and the PTS system, required for 114 incorporation of phosphorylated saccharide, was significantly upregulated starting at 24 115 h.p.i. Expression of genes required for iron acquisition, such as biosynthesis of various 116 secondary metabolites-staphyloferrin A and B, was not upregulated at 6 h.p.i., but was 117 upregulated after 24 h.p.i. In addition, expression of ABC transporters, required for the 118 acquisition of nickel and manganese, was upregulated after 24 h.p.i. On the other hand, 119 expression of genes involved in terpenoid backbone biosynthesis, required for cell wall 120 synthesis, pigment, menaquinone, and peptidoglycan biosynthesis, was downregulated 121 from 24 h.p.i. As for the host side, we performed RNA-Seq analysis using RNA of liver 122 organs mixed in the sample that mapped uniquely on the mice genome (Supplementary 123 Dataset 2) and selected genes with a significant 5-fold difference (false discovery rate 124 [FDR] p-value <0.05) in liver infected with S. aureus compared with liver injected with 125 PBS to perform the GO term enrichment analysis (Reactome, Supplementary Dataset 3). 126 The results suggested that genes involved in the induction of innate immunity and 127 inflammation, and those related to metal sequestration were significantly upregulated 128 (Supplementary Table 3). On the other hand, reactomes related to ATP synthesis, such as 129 respiratory electron transport, ATP synthesis by chemiosmotic coupling, and heat 130 production by uncoupling proteins, were downregulated at all time-points 131 Table 3

Energy metabolism 146
We further analyzed the gene expression changes in each S. aureus pathway in mouse

Metal acquisition system 204
Iron acquisition is essential for pathogen growth in the host 18 . S. aureus has at least 5 iron 205 acquisition systems, and the genes involved in these systems are known to be upregulated 206 in the host, since the pathogen-infected host hides iron by increasing metal sequestration 207 proteins, as shown in Supplementary Table 2. Although the expression of iron acquisition 208 system genes was not upregulated at 6 h.p.i., it was highly increased at 24 and 48 h.p.i. 209 ( Figure 3A), a pattern that did not correspond to that of the host's metal sequestration 210 proteins, which were upregulated from 6 h.p.i. It might be that S. aureus obtained iron 211 from lysed hemocytes by hemolysins, which were highly upregulated at the initial 212 infection stage in the host, as described below.     S. aureus Newman strain was grown overnight on TSB medium at 37˚C. The full growth 314 was diluted 100-fold with 5 ml TSB and regrown, and then the cells were centrifuged and 315 suspended in PBS pH 7.2. The cells (5.6×10 7 CFU) were injected into C57BL/6J mice 316 via the tail vein. At 6, 24, and 48 h.pi, mice were killed to isolate liver. The organs were 317 immediately placed in liquid nitrogen and maintained at -80˚C until RNA extraction. One 318 kidney and a part of the liver were homogenized to calculate viable cell numbers in each 319 organ. Each experiment was conducted with 3 animals and data are represented as an 320 average.
All data were analyzed using CLC Genomics Workbench software, version 12 (CLC Bio, 358 Aarhus, Denmark). Reads were aligned to the Newman genome (Accession No. 359 NC_009641) and the mouse genome (Mus_musculus.GRCm38) allowing a minimum 360 length fraction of 0.95 and minimum similarity fraction of 0.95. Differential gene 361 expression analysis was performed using edgeR analysis 29 for a normalized dataset by 362 scaling using the default setting. Genes with an FDR p<0.05 using the Benjamini and 363 Hochberg's algorithm 30 were classified as having significantly different expression. 364 365

Data analysis 366
For the KEGG pathway analysis, we selected S. aureus genes whose expression levels 367 changed more than 2-fold in mouse liver compared with in vitro culture medium 368 conditions and whose FDR p-value was less than 0.05. R ver. 3.6.1, Bioconductor 3.10 369 package, and pathview 31 were used. GO term enrichment analysis was performed on the 370 site http://geneontology.org using the Reactome pathway. The genes selected for 371 expression were significantly (FDR p-value < 0.05) upregulated or downregulated 5-fold 372 in S. aureus-infected mice compared with PBS-treated mice and analyzed by the Fishers 373 exact test with FDR correction (FDR p-value < 0.05 was considered statistically 374 significant). 375 376

Construction of S. aureus mutants and complement strain 377
Single cross-over recombination; gene disruptions were performed as previously 378 described 32 . In summary, the internal regions within the open reading frames of the gene 379 aureus Newman 34 by phage transduction using phage 80α as previously described 35