RB50 ΔbtrA#1 ΔbtrA#2 RB50 ΔbtrA#1 ΔbtrA#2 time (h) O.D 600 0 5 10 15 20 25 0.1 1 10 Growth rate (h) RB50: 1.89±0.17 ΔbtrA: 6.78±1.35 bvgΔ54 RB50 ΔbtrA ΔbtrS n=3 btrA btrS bsp22 bopN bscN bteA fhaB cyaA prn flaA bipA relative fold difference 0.0 0.5 1.0 1.5 2 4 6 8 10 12 14 16 100 150 200 250 Pellet Supernatant ΔbteAHA RB50HA WD3HA ΔbteAHA RB50HA WD3HA ΔbteA RB50 WD3 ΔbteA RB50 WD3 BvgA BtrA-HA PRN Bsp22 Supernatant RB50 ΔbtrA ΔbtrS 25- 20- 15- 37- 10- 50- 75- 100- 150- 250- kDa CyaA Bsp22 Prn FhaB RB50 ΔbtrA ΔbtrS Pellet bvgΔ54 bvgΔ54 - Bsp22 - BopD - BopN BopB - * * * * LDH release (% max. lysis) Hela 0 20 40 60 80 100 RB50 ΔbscN ΔbteA ΔbtrA ΔbteAΔbtrA Bbr77 0 20 LDH release (% max. lysis) A549 40 60 80 100 p <0.0001 ns bsc locus bscP btrA btrW btrU btrV alr BB1640 BB1641 BB1643 btrS bscC bscW bscU bscT bscC bscR bscQ fhaB cyaA prn ptx bvgS bipA btrS bsc T3SS BvgS BvgA vrgs (e.g.flagella) FIGURE 3: btrA over expression; A, Colony morphology on Bordet-Gengou agar containing 7.5% defibrinated blood. B, Semi-quantitative RT-PCR of indicated gene transcripts. C, qRT-PCR of indicated gene transcripts. D, Expression and secretion of virulence gene products in pellet and culture supernatants. Coomassie brilliant blue protein stain after SDS-PAGE of O.D. normalized pellets and supernatants of indicated strains. Lower panels, immunoblots of the indicated virulence determinants. btrS over expression; E, Colony morphology on Bordet- Gengou agar containing 7.5% defibrinated blood. F, Semi-quantitative RT-PCR of indicated gene transcripts. G, qRT- PCR of indicated gene transcripts. H, Upper panel; Coomassie brilliant blue protein stain after SDS-PAGE of O.D. normalized pellets and supernatants of indicated strain. Lower panels, immunoblots of the indicated virulence determinants.xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx Microbiology, Immunology & Molecular Gene4cs Host Adaptation by Bordetella species: Differential control of Type Three Secretion System Umesh Ahuja and Jeff F. Miller ([email protected]) ABSTRACT Model of the Evolution of Mammalian Bordetellae Overexpression of btrA and btrS Confer Contras4ng Phenotypes BtrA is a T3SS Substrate That Regulates Mul4ple Virulence Genes BtrAS Selec4vely Regulate T3SS Expression in B. pertussis New Model of Virulence Gene Regulation RNA-Seq Reveals Novel Regulatory Network FIGURE 1: Minimum spanning tree of 3 closely related Bordetellae: A, The tree is based on the sequence of seven house keeping gene. The sequence types (STs) sharing the highest number of single locus variants were connected first. Each circle represents a ST, the size of which is related to the number of isolated. B, Model of the evolution of mammalian bordetellae. The bar on left indicates increasing degree of adaptation. Reference: Diavatopoulos, D.A., Cummings, C.A., Schouls, L.M., Brinig, M.M., Relman, D.A., and Mooi, F.R. (2005) Bordetella pertussis, the Causative Agent of Whooping Cough, Evolved from a Distinct, Human-Associated Lineage of B. bronchiseptica. PLoS Pathog 1: e45. B. bronchiseptica isolates from diverse hosts, including humans, display potent cytotoxicity against a broad range of mammalian cells which is dependent on BteA, a type three secretion system (T3SS) effector. In contrast, neither clinical nor laboratory isolates of B. pertussis display T3SS-dependent phenotypes in vitro, despite the fact that bsc T3SS genes are present, intact, and nearly identical to their B. bronchiseptica counterparts. Furthermore, bteA alleles are functionally interchangeable between these Bordetella subspecies. An hypothesis that could explain subspecies-specific differences in T3SS activity is that the regulatory network that controls T3SS has co-evolved with host specificity. Indeed, T3SS activity is known to be controlled at multiple levels. bteA and bsc apparatus genes are transcriptionally activated by an extracytoplasmic function (ECF) sigma factor, BtrS, which is encoded by a BvgAS-activated gene. We have discovered and characterized a regulatory node in B. bronchiseptica involving BtrA, a secreted antagonist of BtrS. RNA-seq analysis shows that BtrA negatively regulates expression of T3SS apparatus, effector, and regulatory loci, and positively controls hundreds of genes encoding toxins, adhesins, and products of unknown function. Deletion of btrA or over-expression of btrS leads to increased T3SS activity, while over-expression of btrA mimics the effects of ∆btrS allele on T3SS expression and phenotypes. Antagonism is mediated by a physical interaction between BtrA and BtrS which couples secretion to gene expression. Of particular relevance to human disease, deletion of btrA in B. pertussis derepressess bsc T3SS gene expression and confers readily detectable BteA-dependent cytotoxicity in vitro. On the basis of these and other results, we propose that the BtrA regulatory node determines, at least in part, subspecies-specific differences in T3SS expression between B. bronchiseptica and B. pertussis. The implications of these results for the design of improved vaccines against pertussis will be discussed. A B BtrA Physically Interacts with BtrS FIGURE 4: A, Schematic diagram of BtrA constructs used in this study and their effect on T3SS expression, FhaB expression and interaction with BtrS. B, Interactions between BtrA and BtrS in a yeast two-hybrid system.XXXXXXXXXXXXXXXXXXXXXXXXXXXXXxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx FIGURE 6: A, Immunoblot analysis of normalized pellet and culture supernatant of wild type B. pertussis and its btrA isogenic null mutant. B and C, Cytotoxicity of wild type B. pertussis and its isogenic mutant derivatives measured by LDH release. HeLa cells were infected with indicated bacteria at different MOIs for 6h and 24h. Results are average of 3 independent experiments. FIGURE 5: A and D, Heat map depicting the results of cluster analysis of the log 2 transformed RPKM values. B and E, The box plots showing the distribution of transcripts. C, Predicted regulatory pathways. F, Log 2 fold change gene expression of known virulence-related genes compared to the wild-type. G, Venn diagrams showing overlap of various overexpression gene clusters (a-c) with null-mutant gene clusters (1-4). xxxxxxxxxxxxxxxx A B C In view of our current findings we propose a new general model of virulence gene regulation in Bordetella that involves intricate interaction between BtrS and BtrA in fine-tuning the Bvg-driven response. BtrS act as a positive regulator of T3SS locus genes while BtrA acts as an antagonist of BtrS. The transcription of btrS is BvgAS-dependent while the activity of BtrS is regulated by sequestration by BtrA. The transcription of btrA is regulated by BtrS while the inhibitory intracellular levels of BtrA are controlled by its secretion through the T3SS pore. The secretion through the T3SS pore augurs further T3SS induction. The rate of BtrA production and secretion through T3SS is dependent on the level of BtrS activation. FIGURE 2: A, Genomic organization of btrA-btrS locus. B, T3SS-dependent secretion of BtrA. C, Colony morphology on Bordet- Gengou agar containing 7.5% defibrinated blood. D, Growth rate determination in Stainer-Scholte medium. E, Model of BvgAS-dependent virulence gene regulation. F, qRT-PCR of indicated virulence- related genes. G, Immunoblot analysis of indicated virulence gene products in culture supernatants and pellets. H, In vitro cytotoxicity of indicated strains towards HeLa and A549 cells.X wild type bvgSΔ54 ΔbtrS ΔbtrA ΔbtrAΔbtrS 5 95 Percentile Log 2 Transformed RPKM Expression (log 2 ) -2 0 2 4 6 8 10 12 n=89 btrS btrA BvgS BvgA Cluster 1 genes (e.g.flagella) Expression (log 2 ) -2 0 2 4 6 8 10 12 14 n=208 btrS btrA BvgS BvgA Cluster 2 genes (e.g.flagella) Expression (log 2 ) -2 0 2 4 6 8 10 12 14 n=168 btrS btrA BvgS BvgA Cluster 3 genes (e.g.fhaB/cyaA/prn) btrS btrA BvgAS Cluster 4 genes (e.g.T3SS) n=126 wild type bvgSΔ54 ΔbtrS ΔbtrA ΔbtrAΔbtrS Expression (log 2 ) -2 0 2 4 6 8 10 12 14 16 18 Cluster 1 Cluster 2 Cluster 3 Cluster 4a Cluster 4b vector pbtrS 5 95 Percentile Log 2 Transformed RPKM pbtrA Expression (log 2 ) Expression (log 2 ) Expression (log 2 ) vector pbtrS pbtrA -2 0 2 4 6 8 10 12 14 16 18 n=29 -2 0 2 4 6 8 10 12 n=61 -2 0 2 4 6 8 10 12 14 n=12 Cluster a Cluster b Cluster c Log 2 Fold Change (RPKM) cyaC cyaA cyaB cyaD cyaE sphB1 brkB prn fim3 ptxA ptxB ptxE ptxD ptxC ptlA ptlB ptlC ptlD ptlL ptlE ptlF ptlG ptlH fhaL bipA fhaS fimD fimC fimB fimA fhaB bvgA bvgS bvgR fimN fimX fim2 dnt pagP ΔbtrS ΔbtrA ΔbtrAΔbtrS bscF bscE bscD bcrD orf2 orf3 orf4 bopN bsp22 orf6 bcrH1 bopD bopB bcrH2 bcr4 bscI bscJ bscK bscL bscN bscO bscP bscQ bscR bscS bscT bscU bscW bscC btrA btrS btrU btrW btrV btcA bteA -8 -6 -4 -2 0 2 4 6 -4 -3 -2 -1 0 1 2 3 a 2 3 4 abc 1+2 3 4 b 1+2 3 4 c 3 4 D E F G Supernatants Pellets RB50 Bp536 Bp536ΔbtrA#1 Bp536ΔbtrA#2 Bp536 Bp536ΔbtrA#1 Bp536ΔbtrA#2 mid-log stationary RB50 Bp536 Bp536ΔbtrA#1 Bp536ΔbtrA#2 Bp536 Bp536ΔbtrA#1 Bp536ΔbtrA#2 mid-log stationary CyaA Bsp22 S4 subunit Prn FhaB A B C M1 L104 T164 T134 Q194 BtrAFL BtrAΔQ194-M226 BtrAΔT164-M226 BtrAΔT134-M226 BtrAΔL104-M226 M226 T124 S177 BtrAΔL101-V133 BtrAΔT134-D163 BtrAΔT164-M193 BtrAT124-S177 S81 BtrAS81-S177 + + + - - - + + + + + + + + + T3SS FhaB + nd nd - nd nd - nd nd - nd nd - nd nd BtrS interaction BtrAΔT164-M226 BtrAΔT134-M226 BtrAΔL104-M226 BtrAΔL101-V133 BtrAΔT134-D163 BtrAΔT164-M193 BtrAT124-S177 BtrAS81-S177 DDO DDO X/A QDO X/A BtrAFL BteA644 BtrS BtrS BtrS BtrS BtrS BtrAFL BtrS BteA644 BtrAFL BtrAFL BtrS BtrS BtrS BtrS BtrS BtrAΔQ194-M226 Bait Prey RB50/vector ΔbtrA/pbtr A ΔbtrA/vector RB50/pbtr A RB50/vector ΔbtrA/pbtr A ΔbtrA/vector RB50/pbtr A B A RB50/vector ΔbtrA/vector recA btrA + + + + m m m m gDNA bopN bsp22 btrS fhaB RB50/pbtrA ΔbtrA/pbtrA cyaA prn flaA C btrA btrS bsp22 bopN cyaA fhaB prn flaA relative-fold difference 0.0 0.5 1.0 2 4 6 8 10 12 14 16 18 20 22 24 150 300 450 RB50/vector RB50/pbtrA ΔbtrA/vector ΔbtrA/pbtrA RB50/vector RB50/pbtrS ΔbtrA/vector ΔbtrA/pbtrS ΔbtrA/vector RB50/vector 25- 20- 15- 37- 10- 50- 75- 100- 150- 250- kDa Bsp22 RB50/pbtrA ΔbtrA/pbtrA ΔbtrA/vector RB50/vector RB50/pbtrA ΔbtrA/pbtrA pellet supernatant FhaB CyaA Prn 25- 20- 15- 37- 10- 50- 75- 100- 150- 250- kDa D F E G H RB50/vector ΔbtrA/vector ΔbtrA/pbtrS RB50/pbtrS RB50/vector ΔbtrA/vector ΔbtrA/pbtrS RB50/pbtrS RB50/vector ΔbtrA/vector recA btrA + m gDNA bopN bsp22 btrS fhaB RB50/pbtrS ΔbtrA/pbtrS cyaA prn flaA + m + m + m relative-fold difference btrA btrS bsp22 bopN cyaA fhaB prn flaA 0.0 0.4 0.8 1 5 9 13 17 21 25 150 300 450 ΔbtrA/vector RB50/vector 25- 20- 15- 37- 10- 50- 75- 100- 150- 250- kDa CyaA Prn FhaB RB50/pbtrS ΔbtrA/pbtrS ΔbtrA/vector RB50/vector RB50/pbtrS ΔbtrA/pbtrS pellet supernatant Bsp22 25- 20- 15- 37- 10- 50- 75- 100- 150- 250- kDa BvgS BvgA [Adenylate cyclase/CyaA] CyaE CyaB BteA [Pertussis toxin] (Bp only) Capsule [Pertac4n] [FHA] TCT Type I CyaD [Type III] Vag8, BrkA Tcf Pep4doglycan layer OM IM FimC usher FimD Fim2 & 3 FimB Type IV vrg “Phosphorelay” bvgA bvgS bvgR + + FhaB FhaC Adapted from Locht et. al. Current Opinion in Microbiology 2001, 4:82–89 B A F G H A B C D E California NanoSystems Ins4tute