1 Global Analysis of Gene Expression Reveals Mrna Superinduction Is
Total Page:16
File Type:pdf, Size:1020Kb
1 2 3 4 5 Global analysis of gene expression reveals mRNA superinduction is required for the 6 inducible immune response to a bacterial pathogen 7 8 9 10 Kevin C. Barry1, Nicholas T. Ingolia2*, and Russell E. Vance1,3,4* 11 12 13 14 1Division of Immunology and Pathogenesis, Department of Molecular & Cell Biology 15 2Division of Biochemistry, Biophysics and Structural Biology, Department of Molecular 16 & Cell Biology 17 3Cancer Research Laboratory 18 4Howard Hughes Medical Institute 19 University of California, Berkeley, California 94720, USA. 20 21 22 23 *Correspondence: Russell E. Vance, Division of Immunology and Pathogenesis, 24 Department of Molecular and Cell Biology, University of California, Berkeley, CA 25 94720-3200. Phone: (510) 643-2795. Fax: (510) 642-1386. E-mail: [email protected] 26 or Nicholas T. Ingolia, Division of Biochemistry, Biophysics and Structural Biology 27 Department of Molecular and Cell Biology, University of California, Berkeley, CA 28 94720-3200. Phone: (510) 664-7071. Email: [email protected] 29 1 30 Abstract 31 The inducible innate immune response to infection requires a concerted process of gene 32 expression that is regulated at multiple levels. Most global analyses of the innate immune 33 response have focused on transcription induced by defined immunostimulatory ligands, 34 such as lipopolysaccharide. However, the response to pathogens involves additional 35 complexity, as pathogens interfere with virtually every step of gene expression. How 36 cells respond to pathogen-mediated disruption of gene expression to nevertheless initiate 37 protective responses remains unclear. We previously discovered that a pathogen- 38 mediated blockade of host protein synthesis provokes the production of specific pro- 39 inflammatory cytokines. It remains unclear how these cytokines are produced despite the 40 global pathogen-induced block of translation. We addressed this question by using 41 parallel RNAseq and ribosome profiling to characterize the response of macrophages to 42 infection with the intracellular bacterial pathogen Legionella pneumophila. Our results 43 reveal that mRNA superinduction is required for the inducible immune response to a 44 bacterial pathogen. 2 45 Introduction 46 Gene expression is a concerted process that is regulated at multiple steps, including 47 transcription, mRNA degradation, translation, and protein degradation. Most global 48 studies of gene expression have focused on the transcriptional response, but the relative 49 importance of transcription in determining protein levels remains debated (1-6). One 50 recent study analyzed the response of dendritic cells to lipopolysaccharide (LPS) and 51 found that changes in mRNA levels accounted for ~90% of observed alterations in 52 protein levels (7). However, the response to infection with a virulent pathogen is certainly 53 more complicated than the response to a purified immunostimulatory ligand such as LPS. 54 Indeed, pathogens have evolved to disrupt or manipulate almost every cellular process 55 involved in gene expression (8). An effective innate immune response to infection 56 therefore requires that host cells be able to induce appropriate responses in the face of 57 pathogen manipulation. 58 Inhibition of host protein synthesis is a common strategy used by many viral and 59 bacterial pathogens to disrupt host gene expression (9, 10). For example, the intracellular 60 bacterial pathogen Legionella pneumophila uses its Dot/Icm type IV secretion system 61 (T4SS) to translocate into host cells several effector proteins that block host protein 62 synthesis, including at least four effectors that target the elongation factor eEF1A (10-14). 63 Similarly, the bacterial pathogen Pseudomonas aeruginosa blocks host translation 64 elongation by secretion of exotoxin A (10, 15, 16). Interestingly, we previously 65 discovered that host cells respond to protein synthesis inhibition — whether by 66 Legionella, exotoxin A, or by pharmacological agents that block translation initiation or 67 elongation — by initiating a specific host response characterized by production of 68 specific pro-inflammatory cytokines, including interleukin-23 (Il23a), granulocyte 69 macrophage colony stimulating factor (Csf2) and interleukin-1α (Il1a) (11, 13). The 70 mechanism by which infected host cells are able to produce certain cytokines despite a 71 global (>90%) block in protein synthesis remains unclear, but at least two distinct models 72 have been proposed (9-11, 13, 15, 17-19). One model posits that the block in protein 73 synthesis leads to superinduction of cytokine mRNAs that is sufficient to overcome the 3 74 partial block in host protein synthesis (11, 13). Alternatively, it has been proposed that 75 host cells may circumvent the global block in protein synthesis by selective translation of 76 specific cytokine transcripts (15, 20). 77 To determine how host cells mount an inflammatory response when protein 78 synthesis is disrupted, we performed parallel RNAseq and ribosome profiling (21-23) of 79 Legionella-infected mouse primary bone marrow derived macrophages (BMMs). The 80 results reveal the relative contributions of translational regulation and mRNA induction 81 in controlling immune responses to pathogenic L. pneumophila, and support a model in 82 which the majority of gene induction in response to pathogenic infections occurs at the 83 level of mRNA induction. We were able to identify a subset of mRNAs that display 84 higher-than-average ribosome occupancy, but the elevated occupancy of these mRNAs 85 was observed in uninfected cells as well as in cells infected with L. pneumophila. We 86 propose that mRNA superinduction provides a robust mechanism for host cells to initiate 87 a response to infection despite pathogen-mediated disruption of host gene expression. 4 88 Results 89 mRNA superinduction mediates the host response to virulent L. pneumophila 90 The relative role of transcription versus translation in mediating the inducible response to 91 an infection with a virulent bacterial pathogen remains unclear. Thus, we performed 92 ribosome profiling (21-23) and total (rRNA-depleted) RNA sequencing of bone marrow- 93 derived macrophages (BMMs) infected with L. pneumophila. BMMs were infected with 94 a virulent ΔflaA strain, an avirulent T4SS-deficient ΔdotAΔflaA strain, or a Δ7ΔflaA 95 strain that lacks the 7 effectors associated with inhibition of host protein synthesis. RNA 96 was isolated at 6 hours post-infection, which was the earliest we could detect significant 97 L. pneumophila-induced translation inhibition without marked cytotoxicity (data not 98 shown). L. pneumophila strains on the ΔflaA background were used to reduce cell 99 cytotoxicity by avoiding the effects of NAIP5/NLRC4 inflammasome activation by 100 flagellin (24, 25) and we previously showed loss of flagellin does not affect blockade of 101 host translation or the transcriptional induction of inflammatory cytokines (11). Control 102 experiments demonstrated that ~90% of macrophages were infected with at least one 103 bacterium under our infection conditions (Figs. 1 S1A-B). 104 Lysates from infected macrophages were split and used to generate ribosome 105 profiling libraries and RNAseq libraries, thereby allowing us to compare directly the 106 mRNA levels and ribosome occupancy of those mRNAs from the same cells. As a 107 confirmation of the quality of the ribosome profiling libraries, ribosome footprints were 108 found to map preferentially to the exonic regions of infection-induced genes (Fig. 1), and 109 showed a strong bias towards 27-28 nucleotide fragment lengths (Fig. 1 S2), consistent 110 with the known size of ribosome-protected footprints. In accord with previous studies, 111 induction of ribosome footprints on Gem, Csf2, and Il23a required the 7-bacterial 112 effectors associated with the block in host protein synthesis, while induction of ribosome 113 footprints corresponding to Il1a and Il1b required the bacterial T4SS (Fig. 1A-F). 114 We first analyzed WT BMMs for T4SS-dependent gene induction, defined as the 115 ratio of normalized read counts in the virulent ΔflaA to avirulent ΔdotAΔflaA L. 116 pneumophila-infected conditions (see Methods). We found that the majority of T4SS- 5 117 dependent increases in ribosome footprints could be explained at the level of mRNA 118 induction, as there was nearly a perfect linear correlation between the extent of mRNA 119 induction and ribosome footprints for all T4SS-induced genes (Fig. 2A). This correlation 120 held for numerous known pathogen-induced mRNAs, including Il23a, Gem, Csf2, Il6, 121 Tnf, Cxcl1, Cxcl2, Dusp1, and Dusp2, as well as for the cytokines Il1a and Il1b that were 122 previously proposed to be preferentially translated (20). To confirm that cytokine protein 123 levels correlate with mRNA levels, we infected BMMs with ΔflaA or ΔdotAΔflaA L. 124 pneumophila and measured the levels of 42 cytokines or immune related proteins in the 125 supernatants or cell lysates of these BMMs at 6h, using commercially available bead 126 arrays. Of the cytokines assayed, 18 cytokines/proteins were measured above the limit of 127 detection in lysates, and 22 cytokines/proteins were measured above the limit of detection 128 in supernatants. The T4SS-dependent fold-induction of these protein levels was plotted 129 versus the T4SS-dependent fold-induction of mRNA levels (Fig. 2B-C). We observed a 130 robust correlation between the extent of mRNA induction and the extent of protein 131 induction, particularly