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 in lysates (Fig. 2B). The correlation seems to apply for the most

132 highly induced proteins/mRNAs (e.g., IL-10 (Il10) and GM-CSF (Csf2)) but also for

133 more modestly induced cytokines (Il1a, Il1b, Cxcl10). The less robust correlation

134 between mRNA levels and protein levels in the cell supernatant (Fig. 2C) may reflect

135 differing rates of secretion, accumulation in the supernatant over time, re-binding to cell

136 surface receptors, and stability in the supernatant. Taken together, these results suggest

137 that the inducible immune response to L. pneumophila is controlled primarily at the level

138 of mRNA superinduction (Fig. 2).

139

140 mRNA induction accounts for effector-induced gene expression

141 L. pneumophila uses multiple mechanisms to block host protein synthesis. It has been

142 shown that up to 7 bacterial effectors secreted into the host cytosol can block translation

143 (11, 13). Interestingly, the ∆7 strain that lacks these effectors is still able to partially

144 suppress host protein synthesis by a mechanism that remains to be fully characterized (11,

145 13, 26). It has been proposed that T4SS-competent L. pneumophila damages host cell

6 146 membranes, resulting in ubiquitylation-dependent down regulation of mTOR activity and

147 a block in cap-dependent translation (26). Consistent with its ability to partially suppress

148 protein synthesis, the ∆7 strain still provokes IL-1α production, though its ability to

149 stimulate Il23a and Csf2 expression is diminished (11, 13).

150 To determine the mechanism of effector-triggered cytokine induction, we

151 performed parallel RNAseq and ribosome profiling of BMMs infected either with ∆flaA

152 or ∆7∆flaA L. pneumophila. As expected, induction of Gem, Il23a, and Csf2 is highly

153 dependent on the 7 bacterial effectors, but again, similar to the total T4SS-dependent

154 gene induction (Fig. 2), the 7 effector-dependent induction of ribosome footprints on

155 these genes could be explained at the level of mRNA induction (Fig. 3A). While the 7

156 effector-dependent induction of the genes Dusp1, Dusp2, Cxcl1, Cxcl2, Tnf, Il1a, Il1b,

157 and Il6 was low, all changes in ribosome footprint reads could again be explained by

158 changes in mRNA levels (Fig. 3A). These data suggest that T4SS-dependent and 7

159 bacterial effector-dependent induction of inflammatory cytokines occurs by the induction

160 of mRNA transcripts rather than through a mechanism of selective ribosome loading of

161 mRNAs.

162

163 MyD88 signaling in response to L. pneumophila is required for mRNA induction

164 It was previously proposed that specific transcripts, such as Il1a and Il1b, can be

165 preferentially translated via a mechanism that requires signaling through the adaptor

166 protein MyD88 (20). Thus, we performed ribosome profiling and RNAseq on WT and

167 Myd88–/– BMMs infected with ΔflaA L. pneumophila. For all MyD88-induced genes,

168 including Il1a and Il1b, we observed a linear correlation between the induction of

169 ribosome footprints and RNAseq reads (Fig. 3B). This implies that MyD88-dependent

170 induction of Il1a and Il1b ribosome footprints is controlled primarily at the level of

171 mRNA induction, rather than at the level of selective ribosome loading of the mRNA. A

172 similar pattern was also observed for other MyD88-induced genes, including Cxcl1, Csf2,

173 Tnf, and Il6 (Fig. 3B). Taken together, our results argue that the ability of host cells to

174 overcome a pathogen-induced block in protein synthesis, and produce inflammatory

7 175 cytokines such as IL-1α and IL-1β, requires a T4SS- and MyD88-dependent increase in

176 mRNA levels rather than preferential loading of these cytokine mRNAs with ribosomes.

177

178 Ribosome occupancy of mRNAs varies but is independent of infection

179 The above analyses sought to determine whether T4SS-dependent or MyD88-dependent

180 gene induction was due to increased mRNA levels or increased ribosome loading of

181 mRNAs. However, the analyses did not reveal whether there is differential ribosome

182 occupancy of constitutively expressed (i.e., non-induced) mRNAs. We thus analyzed the

183 ratio of ribosome footprint reads to RNAseq reads for all (induced and non-induced)

184 transcripts in uninfected BMMs, and in BMMs infected with ΔflaA, ΔdotAΔflaA, or

185 Δ7ΔflaA L. pneumophila. This analysis revealed a wide range of ribosome occupancies

186 across different transcripts (Figs. 4A-D). As might be anticipated, many of the mRNAs

187 with the highest ribosome occupancy encoded abundant ‘housekeeping’ proteins,

188 including Acta1 and histone mRNAs (e.g. Hist1h2ba, H2afj, and Hist3h2ba) (Fig. 4A;

189 Table 1). Importantly, most mRNAs that exhibit increased ribosome occupancy in

190 uninfected BMMs also exhibit increased ribosome occupancy in ΔflaA, ΔdotAΔflaA, or

191 Δ7ΔflaA L. pneumophila-infected BMMs (Fig. 4B-D; Table 1), implying that the

192 increased ribosome occupancy of these mRNAs is constitutive and not induced in

193 response to infection. A few mRNAs of immunological interest, namely Lyz1, S100a11,

194 and Cxcl3 exhibited elevated ribosome occupancy in all infection conditions (Figs. 4B-D;

195 Table 1). In contrast, Ftl1 mRNA exhibited very low ribosome occupancy (Fig. 4A),

196 consistent with a previous report showing that Ftl1 translation can be strongly repressed

197 (27). Atf4 is another gene known to be regulated at the level of translation (28), and in

198 ΔflaA and ΔdotAΔflaA L. pneumophila-infected BMMs, Atf4 exhibited low ribosome

199 occupancy (Fig. 4B-C). Atf4 was not expressed at high enough levels to be called as

200 detected in uninfected or Δ7ΔflaA L. pneumophila-infected BMMs (Figs. 4A and D).

201 Taken together, our results reveal that several mRNAs exhibit constitutive increased or

202 decreased ribosome occupancy, as expected. Despite this, ribosome occupancy of

203 mRNAs was not markedly affected by L. pneumophila infection (Figs. 4A-D).

8 204

205 Global analysis of translation inhibition by L. pneumophila

206 A benefit of ribosome profiling is that it permits the mapping of ribosome footprints with

207 nucleotide resolution. Thus, to characterize the position of ribosomes on mRNAs after

208 infection with L. pneumophila, we generated metagene ribosome footprint profiles from

209 libraries generated from WT BMMs (Fig. 5). Metagene profiles were generated by

210 mapping the inferred A site position of ribosome footprint reads relative to the start (Fig.

211 5A, C, E, G, I) or stop (Fig. 5B, D, F, H, J) codon on a given transcript. Mapped reads

212 were then summed to produce a global view of the distribution of ribosomes across all

213 transcripts in our dataset. As expected from the known stepwise codon-by-codon

214 movement of the ribosome, all metagene plots demonstrated a characteristic 3-nucleotide

215 periodicity (Fig. 5). Interestingly, metagene ribosome profiles of uninfected (Fig. 5A-B),

216 ΔflaA (Fig. 5C-D), ΔdotAΔflaA (Fig. 5E-F), or Δ7ΔflaA (Fig. 5G-H) infected BMMs

217 appeared grossly similar to each other, even though global translation is blocked only in

218 the ΔflaA and Δ7ΔflaA L. pneumophila-infected conditions (11, 13, 20). This result can

219 be explained by the fact that ribosome metagene profiles do not distinguish whether

220 ribosome footprints arise from stalled or translating ribosomes, unless the stall occurs at a

221 characteristic distance from the start or stop codon. In fact, we did notice a slight increase

222 in the number of ribosomes found at the start site of the transcript in Δ7ΔflaA L.

223 pneumophila-infected BMMs as compared to other conditions (Fig. 5I). This may reflect

224 a selective block in translation initiation by this strain (see below). In addition, we noted

225 that in all conditions, ribosomes accumulated at the stop codon, suggesting that, in

226 BMMs, translation termination may be a limiting step in translation (Fig. 5J).

227

228 L. pneumophila blocks translation at the levels of initiation and elongation

229 To distinguish whether an observed ribosome footprint arises from a stalled or translating

230 ribosome, we performed ribosome run-off experiments. In these experiments, new

231 translation initiation was blocked by the drug harringtonine 120 seconds prior to cell lysis.

232 Harringtonine inhibits the first rounds of peptide bond formation following ribosome

9 233 subunit joining and results in accumulation of ribosomes at the translational start site and

234 run-off of elongating (but not stalled) ribosomes that have already cleared the start codon

235 (21, 23, 29-31). Importantly, cells experiencing a block in translation elongation will

236 exhibit less ribosome run-off after harringtonine treatment, and an increased number of

237 reads at the 5ʹ end of mRNAs after drug treatment (23), compared to cells in which

238 elongation is not blocked.

239 As expected, uninfected and ΔdotAΔflaA-infected BMMs show an increase in

240 ribosome footprints at the translation start site and a preferential loss of ribosome

241 footprints from the 5ʹ and 3ʹ end of mRNAs, consistent with the expected effects of

242 harringtonine and demonstrating clear ribosome run-off (Figs. 6A-B,E-F, Fig. 6 S1A-B).

243 By contrast, ΔflaA L. pneumophila infected BMMs treated with harringtonine exhibited

244 little ribosome run-off (Figs. 6C-D, Fig. 6 S1A-B), consistent with the expectation that

245 ΔflaA L. pneumophila blocks host translation elongation. The Δ7ΔflaA L. pneumophila

246 strain, lacking all known bacterial effectors that block host protein synthesis, nevertheless

247 shuts down host translation (11), yet we observed clear evidence of run-off of elongating

248 ribosomes from the 5ʹ and 3ʹ end of mRNAs following harringtonine treatment (Figs. 6G-

249 H, Fig. 6 S1A-B). These data suggest that the residual block in host protein synthesis

250 induced by Δ7ΔflaA L. pneumophila is at the level of translation initiation. Similar results

251 can be seen when analyzing longer stretches of coding sequences (Fig. 6 S1C-F).

252 It is important to note that there are a small proportion of uninfected bystander

253 cells assayed in our experiments. However, it is unlikely that these uninfected cells are

254 responsible for the ribosome run-off seen in Δ7ΔflaA L. pneumophila infected BMMs

255 because the conditions used in these experiments led to most (~90%) cells being infected

256 with L. pneumophila (Fig. 1 S1A-B). Furthermore, if our infection conditions resulted in

257 large numbers of uninfected cells, then a similar run-off should have been observed in the

258 ∆flaA-infected sample, which it was not. Thus, these results suggest that the 7 effectors

259 are required to block translation elongation, and that the residual translation inhibition

260 induced by Δ7ΔflaA L. pneumophila is at the level of translation initiation (Fig. 6).

261

10 262 Cytokine transcripts do not escape the pathogen-induced translation block

263 Although the above results demonstrate a global block in translation elongation in ∆flaA-

264 infected cells, it remains possible that specific transcripts escape this block. We therefore

265 analyzed our translation run-off datasets to assess translation elongation on a per-mRNA

266 basis. We plotted the number of ribosome footprint reads for each transcript in paired

267 untreated and harringtonine treated samples (Fig. 7A-D). In this analysis, we expect that

268 an mRNA with actively elongating ribosomes would show a reduction in the number of

269 5ʹ reads in the harringtonine treated sample, as ribosomes will run off the transcript,

270 compared to the untreated sample. In order to best measure run-off elongation and avoid

271 the expected but confounding effects of harringtonine-induced accumulation of footprints

272 at start codons (which were clearly observed; Fig. 6), we excluded the first 25 codons and

273 analyzed ribosome footprint occupancy over the next 300 codons. Consistent with our

274 previous analysis, we find that uninfected, ΔdotAΔflaA, and Δ7ΔflaA-infected BMMs

275 show a clear global signature of ribosome run-off, again suggesting that the block in host

276 protein synthesis induced by Δ7ΔflaA L. pneumophila infection is occurring at the level

277 of translation initiation (Fig. 7A-D). Importantly, in ΔflaA-infected BMMs there is no

278 evidence of ribosome run-off, consistent with ΔflaA L. pneumophila inducing a block in

279 host translation elongation (Fig. 7B). Interestingly, in all conditions tested, cytokine-

280 related genes fell well within the average of ribosome retention across all transcripts, and

281 if anything, were found to have reduced ribosome run-off compared to a typical gene (Fig.

282 7A-D). A similar trend was seen when we further examined ribosome run-off for specific

283 immune and housekeeping transcripts by plotting the cumulative read counts over the

284 length of the mRNA (Fig. 7 S1). These results imply that at this time point, cytokine

285 transcripts are not preferentially translated in response to pathogenic infection, but

286 instead are controlled at the level of mRNA induction (Fig. 7A-D).

11 287 Discussion

288 Inducible gene expression is of central importance for the immune response to infection.

289 A recent study showed that in response to innate immune stimulation with purified LPS,

290 dendritic cells almost entirely control the induction of genes at the level of transcription

291 (7). However, this conclusion may not apply to cells infected with a virulent pathogen

292 that manipulates gene expression. We thus investigated the relative contributions of

293 mRNA induction and translation during infection with an intracellular bacterial pathogen,

294 Legionella pneumophila, that blocks host protein synthesis.

295 Pathogen-induced blockade of host protein synthesis has been shown in a number

296 of infection models to be sensed by the host and induce an inflammatory response (11, 13,

297 15, 18, 19, 32). We previously identified IL-1α as a key inflammatory cytokine induced

298 preferentially in response to translation inhibition imposed by L. pneumophila (11).

299 However, the mechanism by which cytokine proteins are induced despite a pathogen-

300 induced translation blockade remains unclear. We previously provided evidence for a

301 model in which translation inhibition results in a failure to synthesize negative feedback

302 inhibitors of transcription, e.g., IκB or A20 (13). We proposed this results in a massive

303 and sustained production of cytokine transcripts, termed transcriptional superinduction,

304 that is sufficient to overcome the partial (~95%) block in translation and allow for

305 production of cytokine proteins (11, 13). Another report provided data suggesting that IL-

306 1 production is mediated by MyD88-enhanced protein synthesis, though alternative

307 explanations were also entertained (20). A third study proposed that virulent L.

308 pneumophila regulates cap-dependent translation initiation, via manipulation of the

309 mTOR signaling pathway, to regulate the protein levels of highly abundant transcripts in

310 infected macrophages (26). In our present study, we found that the induction of ribosome

311 footprints by L. pneumophila could be explained by an underlying induction of mRNAs.

312 We did not find evidence for selective ribosome loading of abundant cytokine mRNAs.

313 In addition, ribosome run-off experiments confirmed that cytokine mRNAs are not

314 selectively translated during infection (Fig. 7). Furthermore, we find that the role of

315 MyD88 signaling in gene expression appears to be primarily at the level of mRNA

12 316 induction and not translational regulation (Fig. 3). Thus we conclude that preferential

317 translation does not account for the majority of specific gene induction following

318 infection by L. pneumophila.

319 It remains possible that selective translation initiation mechanisms, e.g., via

320 uORFs, might also contribute modestly to the inducible immune response to L.

321 pneumophila, but these subtle effects were not evident in our global analysis. In any case,

322 it is difficult to explain how regulation of translation initiation could overcome a

323 downstream pathogen-induced block in translation elongation such as is observed during

324 L. pneumophila infection. It is also possible that post-translational mechanisms, which

325 are not addressable with the ribosomal profiling techniques used here, may regulate

326 protein production by infected cells. Indeed, inflammasome-dependent caspase-1

327 processing is known to be an important post-translational regulatory mechanism

328 controlling IL-1β production by infected cells (33). Lastly, our data do not specifically

329 address the mechanism of mRNA induction, though our prior work suggested mRNA

330 induction involves new transcription rather than increased mRNA stability (13).

331 Although wild-type L. pneumophila blocks translation elongation via translocated

332 effectors, we found that Δ7ΔflaA L. pneumophila lacking the effectors nevertheless

333 blocks protein synthesis at the level of translation initiation (Fig. 6). Thus, in contrast to a

334 previous study that used virus-based translation reporter experiments in L. pneumophila-

335 infected RAW macrophages (26), we were clearly able to dissociate the L. pneumophila-

336 induced block in host protein synthesis into two components: (1) an elongation block that

337 required the 7 translocated effectors, and (2) an initiation block that did not require the 7

338 effectors. In addition, our analysis represents an advance over prior studies because we

339 were able to analyze the translation of all endogenous transcripts simultaneously as

340 opposed to measuring translation only of a single exogenous reporter mRNA.

341 Intriguingly, in contrast to the effector-dependent block in translation that we show

342 occurs at the level of elongation, the majority of host-mediated regulation of translation

343 occurs at the level of translation initiation (34). Thus, while it is possible that a novel

344 bacterial effector that directly targets translation initiation could explain the residual

13 345 inhibition of translation by the Δ7ΔflaA L. pneumophila mutant, we favor the hypothesis

346 that the residual block in host protein synthesis may be a result of the host stress response

347 induced by pathogenic infection, consistent with numerous prior studies (9, 10, 19, 26, 35,

348 36). Indeed, T4SS-competent L. pneumophila has been suggested to induce membrane

349 damage that inhibits the mTOR pathway and blocks translation initiation (26). Further

350 studies will be required to identify the bacterial and host pathways required for the

351 residual translation inhibition caused by the Δ7ΔflaA L. pneumophila mutant.

352 The results presented here further support a role for translation inhibition as a

353 signal that the innate immune system uses to recognize and preferentially respond to

354 pathogens (13). Our work provides nucleotide-level analysis of the global block in host

355 protein synthesis induced by L. pneumophila, and demonstrates that L. pneumophila

356 infection results in inhibition of host protein synthesis both at the level of translation

357 initiation and elongation. Importantly, our results also provide insights into the molecular

358 mechanisms by which host cells are able to mount a protective immune response despite

359 a pathogen-induced block in protein synthesis. Using ribosome run-off assays in

360 combination with ribosome profiling and RNAseq, we find that mRNA superinduction,

361 rather than selective mRNA translation, is the strategy by which host cells produce

362 inflammatory cytokines in the face of pathogen-mediated translation inhibition. To be

363 effective, the strategy of mRNA superinduction requires that the magnitude of mRNA

364 superinduction exceeds the magnitude of the block in protein synthesis. Indeed, our data

365 suggest this is the case, as we observe >1000-fold induction of certain mRNAs, whereas

366 we previously estimated the block in protein synthesis to be ~95% (20-fold) (13). One

367 possible advantage of mRNA superinduction as a strategy for overcoming a pathogen-

368 mediated block of protein synthesis is that it does not require specific translation factors,

369 as was previously proposed might mediate selective mRNA translation in L.

370 pneumophila-infected cells (20). In addition, in mammalian cells, selective translation is

371 usually regulated at the level of translation initiation, a strategy that would be easily

372 defeated by pathogens such as L. pneumophila that block the downstream process of

373 translation elongation. Importantly, since numerous viral and bacterial pathogens and

14 374 toxins interfere with host protein synthesis, we propose that our results may provide

375 general insight into the inducible innate immune response to infection. 376

15 377 Materials and Methods

378 Ethics statement

379 These studies were carried out in strict accordance with the recommendations in the

380 Guide for the Care and Use of Laboratory Animals of the National Institutes of Health

381 under animal protocol AUP-2014-09-6665. The protocol was approved by the Animal

382 Care and Use Committee at the University of California, Berkeley.

383

384 Access to high throughput sequencing data

385 The data discussed in this publication have been deposited in NCBI's Gene Expression

386 Omnibus (37) and are accessible through GEO Series accession number GSE89184

387 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE89184).

388

389 Cell culture

390 Macrophages were derived from the bone marrow of C57BL/6J (Jackson Laboratory, Bar

391 Harbor, ME, USA) and Myd88 –/– (38) mice on the B6 background. Macrophages were

392 derived by 8 days of culture in RPMI supplemented with 10% serum, 100μM

393 streptomycin, 100U/mL penicillin, 2mM L-glutamine and 10% supernatant from 3T3-

394 macrophage-colony stimulating factor cells, with feeding on day 5. Cells were re-plated

395 in antibiotic free media 24hrs prior to infection with L. pneumophila.

396

397 Bacterial strains and infections

398 All L. pneumophila strains were derived from LP02, a streptomycin-resistant thymidine

399 auxotroph derived from L. pneumophila LP01. The ΔdotAΔflaA, ΔflaA, and Δ7ΔflaA

16 400 strains were generated on the LP02 background and have been described previously (11,

401 13, 25, 32). 2-fold dilutions of L. pneumophila strains used for infections were grown

402 overnight in liquid buffered-yeast-extract culture and, at the time of infection, cultures

403 with an optical density (600 nm) greater than 4.0 were selected. Bone marrow derived

404 macrophages were plated at a density of 1.56x105 cells per cm2 (1.5.x106 cells per well of

405 a 6-well plate) and infected at an MOI of 3 by centrifugation for 10min at 400 xg. After

406 one hour of infection media was changed. All in vitro L. pneumophila infections were

407 performed in the absence of thymidine to prevent bacterial replication which would

+ 408 otherwise differ between the ∆dotA and Dot strains. The lack of thymidine can result in

409 a loss of bacterial viability, though we attempted to mitigate this concern by examining

410 host gene expression at a relatively early 6h time point.

411

412 Library preparation and sequencing for ribosome profiling

413 Ribosome profiling experiments were undertaken as previously described (21). Bone

414 marrow derived macrophages were plated in tissue culture treated 6-well plates (1.5x106

415 BMMs/well) or 75cm2 flasks (1.2x107 BMMs per flask). At 6hrs post-infection BMMs

416 were lysed by flash freezing and thawed in the presence of lysis buffer (21). When used,

417 harringtonine (LKT Laboratories, Saint Paul, MN, USA) was added at a final

418 concentration of 2 μg/mL for 120 seconds at the end of the 6hr infection. 100 μg/mL of

419 cycloheximide (Sigma-Aldrich, St. Louis, MO, USA) was added to freeze ribosomes

420 after the 120-second harringtonine treatment. Following cycloheximide treatment cells

421 were immediately lysed. Clarified lysates were split and some was used to generate

422 ribosome footprints while some was used to isolate total RNA for RNA sequencing

17 423 (described below). All RNA and DNA gel extractions were performed overnight as

424 previously described (21). The Ribo-Zero Gold rRNA Removal Kit (Illumina, San Diego,

425 CA, USA) was used to remove rRNA from ribosome profiling samples before the

426 dephosphorylation and linker ligation steps (21). Final ribosome profiling libraries were

427 sequenced on a HiSeq2000 System (Illumina) with single read 50 (SR50) read lengths by

428 the Vincent J. Coates Genomics Sequencing Laboratory at UC, Berkeley. See

429 Supplemental Methods for more details on ribosome profiling analyses.

430

431 Generation of RNAseq libraries

432 300μL of clarified lysate was mixed with 900μL of Trizol LS (Thermo Fisher Scientific,

433 Waltham, MA, USA) and RNA was isolated following the manufacturer’s guidelines.

434 RNA integrity was measured utilizing the RNA Pico method on the Agilent 2100

435 Bioanalyzer at the University of California, Berkeley Functional Genomics Laboratory.

436 High quality RNA with a RNA integrity number (RIN) >8 (Agilent Technologies, Santa

437 Clara, CA, USA) was submitted to the QB3-Berkeley Functional Genomics Laboratory

438 and single read 100 base pair read length (SR100) sequencing libraries were generated.

439 Libraries were sequenced on a HiSeq2000 System (Illumina) by the Vincent J. Coates

440 Genomics Sequencing Laboratory at UC, Berkeley. See Supplemental Methods for more

441 details on RNAseq analyses.

442

443 Alignment of RNAseq reads and differential expression analysis

444 RNA sequencing reads were preprocessed using tools from the FASTX-Toolkit

445 (http://hannonlab.cshl.edu/fastx_toolkit/) by trimming the linker sequence from the 3ʹ end

18 446 of each read and in some cases removing 10-15 nucleotides from the 5ʹ of each read to

447 mitigate a region of overrepresented nucleotides. Alignment and differential expression

448 analysis of RNAseq reads were undertaken as previously described (39). Briefly, high

449 quality and preprocessed sequencing reads were aligned using the TopHat splicing-aware

450 short-read alignment program to a library of transcripts derived from the UCSC Known

451 Gene data set, and those with no acceptable transcript alignment were then aligned

452 against the Mus musculus genome (mm10).

453

454 Alignment of ribosome footprint sequences

455 Sequences were processed as described previously (21). Sequences were preprocessed by

456 trimming the linker sequence from the 3ʹ end of each sequencing read and removing the

457 first nucleotide from the 5ʹ end of each read. Reads were then aligned to a rRNA

458 reference using the Bowtie short-read alignment program. All sequences aligning the

459 rRNA reference were discarded. All non-rRNA sequencing reads were aligned using the

460 TopHat splicing-aware short-read alignment program to a library of transcripts derived

461 from the UCSC Known Genes data set, and those with no acceptable transcript alignment

462 were then aligned against the mouse genome (mm10). Perfect-match alignments were

463 extracted and these files were used for analyses. For most analyses, footprint alignments

464 were assigned to specific A site nucleotides by using the position and total length of each

465 alignment, calibrated from footprints at the beginning and the end of CDSes, as

466 previously described (21, 23).

467

468 Counting of ribosome profiling and RNAseq reads

19 469 Counting of reads was performed as previously described (22, 23). Reads were mapped

470 to coding sequences and counted, excluding reads that mapped to the first 15 codons or

471 the last 5 codons of a CDS due to accumulation of ribosomes (23). In order to analyze

472 gene-specific ribosome run-off (Fig. 7A-D), we counted reads mapping from codon 26 to

473 codon 325, i.e., a 300-codon window excluding the first 25 codons of a gene.

474

475 Ribosome occupancy analysis

476 For analyses of ribosome occupancy (Fig. 4) ribosome footprint and mRNAseq read

477 counts were calculated similarly. Read counts were normalized to CDS length, as longer

478 transcripts inherently have increased read counts, generating a read density (read density

479 = read count ÷ transcript length) for each gene. Read densities were further normalized to

480 the sum of read counts of 12 mitochondrial protein-coding genes (see below) as an

481 estimate of total cells in each condition, allowing for comparison among different

482 conditions and libraries (40). For each transcript in the dataset the average raw ribosome

483 footprint read counts for each infection conditions was calculated and transcripts with an

484 average ribosome footprint or RNAseq read count less than 100 were discarded.

485 Additionally, any transcript that had ribosome footprint reads but 0 RNAseq reads was

486 also discarded. Discarded transcripts were defined as undetectable.

487

488 MyD88-dependent gene induction analysis

489 Two experiments were used to generate two independent libraries consisting of B6 and

490 Myd88–/– BMMs infected with ΔflaA or ΔdotAΔflaA L. pneumophila. For each gene in

491 the dataset the average raw ribosome footprint read counts for ΔflaA L. pneumophila

20 492 infected B6 BMMs were sorted and genes with an average ribosome footprint or RNAseq

493 read count less than 100 were discarded. Additionally, any gene that had ribosome

494 footprint reads but no detectable RNAseq reads in B6 or Myd88–/– BMMs were discarded.

495 The sorted read counts were then normalized to ribosome footprint or RNAseq read

496 counts of 12 mitochondrial protein-coding genes (see below) as an estimate of total cells

497 in each condition. MyD88-dependent gene induction was calculated using the equation:

498 MyD88-dependent gene induction = average(normalized B6 read count) ÷

499 average(normalized Myd88–/– read count).

500

501 Type IV secretion system-dependent gene induction analysis

502 Four independent experiments were used to generate four collections of sequencing

503 libraries consisting of B6 BMMs infected with ΔflaA or ΔdotAΔflaA L. pneumophila.

504 For each gene in the dataset the average raw ribosome footprint read counts for ΔflaA L.

505 pneumophila infected B6 BMMs were sorted and genes with an average ribosome

506 footprint or RNAseq read count less than 100 were discarded. Additionally, any gene that

507 had ribosome footprint reads but no detectable RNAseq reads in ΔflaA or ΔdotAΔflaA

508 L. pneumophila infected B6 BMMs were discarded. The sorted read counts were then

509 normalized to ribosome footprint or RNAseq read counts of 12 mitochondrial protein-

510 coding genes (see below) as an estimate of total cells in each condition. T4SS-dependent

511 gene induction was calculated using the equation: T4SS-dependent gene induction =

512 average(normalized ΔflaA-infected read count) ÷ average(normalized ΔdotAΔflaA-

513 infected read count).

514

21 515 Analysis of cytokine and protein levels in infected BMM lysates and supernatants.

516 B6 BMMs were left uninfected or infected with ΔflaA or ΔdotAΔflaA L. pneumophila

517 at an MOI of 3 in duplicate, as described above. Media was changed 1 hour following

518 infection and at 6hrs post-infection supernatants were collected and BMMs washed with

519 PBS. BMMs were lysed in 400μL mammalian cell PE lysis buffer (G-Biosciences, St.

520 Louis, MO, USA) following the manufacturers instructions. Lysates and supernatants

521 were cleared by spinning at 20,000 x g for 30 minutes at 4°C. Cytokine and protein

522 levels were measured using a commercially available cytokine bead array (Rodent MAP

523 4.0-Mouse Sample Testing, Ampersand Biosciences, Saranac Lake, NY, USA) and total

524 protein levels were measure by bicinchoninic acid (BCA) assay (Ampersand Biosciences,

525 Saranac Lake, NY, USA). Protein and cytokine levels in each infection condition were

526 normalized to total protein levels. Infectivity was confirmed by staining for L.

527 pneumophila (see below). mRNA levels of cytokines were determine by counting

528 (counting method described above) previously acquired RNAseq data of B6 BMMs

529 infected with ΔflaA or ΔdotAΔflaA L. pneumophila at an MOI of 3 for 6 hrs. RNAseq

530 read counts were normalized to transcript length and the sum of RNAseq read counts of

531 12 mitochondrial protein-coding genes (see below) as an estimate of total cells in each

532 condition (RNAseq normalization described above). T4SS-dependent induction was

533 measured by taking the ratio of protein or mRNA levels in the ΔflaA infected condition

534 to protein or mRNA levels in the ΔdotAΔflaA infected condition: T4SS-dependent

535 induction = average(normalized ΔflaA mRNA or protein) ÷ average(normalized ΔdotA

22 536 ΔflaA mRNA or protein). T4SS-induction was averaged from two independent

537 experiments and plotted.

538

539 7-effector-dependent gene induction analysis

540 Two independent experiments were used to generate two collections of sequencing

541 libraries consisting of B6 BMMs infected with ΔflaA or Δ7ΔflaA L. pneumophila. For

542 each gene in the dataset the average raw ribosome footprint read counts for ΔflaA L.

543 pneumophila infected B6 BMMs were sorted and genes with an average ribosome

544 footprint or RNAseq read count less than 100 were discarded. Additionally, any gene that

545 had ribosome footprint reads but no detectable RNAseq reads in ΔflaA or Δ7ΔflaA L.

546 pneumophila infected B6 BMMs were discarded. The sorted read counts were then

547 normalized to ribosome footprint or RNAseq read counts of 12 mitochondrial protein-

548 coding genes (see below) as an estimate of total cells in each condition. 7 effector-

549 dependent gene induction was calculated using the equation: 7 effector-dependent gene

550 induction = average(normalized ΔflaA-infected read count) ÷ average(normalized Δ7Δ

551 flaA-infected read count).

552

553 Metagene profile analysis of ribosome profiling libraries

554 Metagene profiles were generated as previously described (22, 23). These metagene

555 profiles indicate the total number of ribosome footprints whose A site falls at the

556 indicated position relative to the start or stop codon of the coding sequence, and reflect a

557 simple, unweighted sum of the footprint profiles around the beginning and the end of

558 each protein-coding gene. The A site position was estimated for each footprint using a

23 559 length-dependent offset from the 5ʹ end of the fragment. The distance from this A site

560 position to the start or stop codon of the coding sequence was then computed, taking into

561 account the fact that translation initiation occurs with the second codon in the A site.

562

563 Analysis of ribosome run-off of individual genes

564 Cumulative ribosome occupancy profiles (Fig. 7 S2) were computed by taking the

565 cumulative sum of ribosome footprints mapping to each position in the gene, scaled by

566 the normalization factor derived from mitochondrial translation in that sample.

567

568 Mitochondrial genes used for library normalization

Size Gene ID Name (bp) ENSMUST00000082392 mt-Nd1 299 ENSMUST00000082396 mt-Nd2 326 ENSMUST00000082402 mt-Co1 495 ENSMUST00000082405 mt-Co2 208 ENSMUST00000082407 mt-Atp8 48 ENSMUST00000082408 mt-Atp6 207 ENSMUST00000082409 mt-Co3 241 ENSMUST00000082411 mt-Nd3 96 ENSMUST00000082414 mt-Nd4 439 ENSMUST00000082418 mt-Nd5 588 ENSMUST00000082419 mt-Nd6 153 ENSMUST00000082421 mt-Cytb 361 569

570

571 Quantification of infectivity

572 WT Bone marrow derived macrophages were plated on a sterile #1.5 coverslip by placing

573 the coverslip in a tissue culture treated 6-well plates and adding 1.5x106 BMMs/well in

24 574 antibiotic free media 24 hrs prior to infection. 2-fold dilutions of L. pneumophila strains

575 used for infections were grown overnight in liquid buffered-yeast-extract culture and, at

576 the time of infection, cultures with an optical density (600 nm) greater than 4.0 were

577 selected. BMMs were infected at an MOI of 3 by centrifugation for 10min at 400 xg.

578 Media was changed after one hour of infection. At 6 hrs post-infection coverslips were

579 collected, washed in PBS, and placed in fixative solution (100 uM sodium periodate, 75

580 uM Lysine, 2.9 uM NaH2PO4, 3.2% sucrose, and 4% paraformaldehyde) for 1 hr at 37°C.

581 Following fixation BMMs were blocked in 2% goat serum in PBS. To stain extracellular

582 L. pneumophila, blocked BMMs were incubated with a rabbit anti-Legionella antibody

583 (RRID: AB_231859; Fitzgerald Industries International, North Acton, MA, USA 20-

584 LR45), washed in PBS, and stained with a goat-anti-rabbit IgG secondary antibody

585 conjugated to Cascade Blue (RRID: AB_2536453; ThermoFisher Scientific, Waltham,

586 MA, USA, C-2764). In some experiments mammalian cell membrane was labeled with

587 FITC-labeled wheat germ agglutinin (Sigma-Aldrich, St. Louis, MO, USA, L4895) prior

588 to permeabilization. BMMs were permeabilized by dipping coverslips into ice-cold

589 methanol. Permeabilized BMMs were blocked with 2% goat serum and stained with a

590 rabbit anti-Legionella antibody (Fitzgerald Industries International, North Acton, MA,

591 USA 20-LR45) followed by incubation with a goat-anti-rabbit IgG secondary antibody

592 conjugated to TexasRed (RRID: AB_2556776; ThermoFisher Scientific, Waltham, MA,

593 USA, T-2767) to mark all (intracellular and extracellular) L. pneumophila. Coverslips

594 were mounted in vectashield antifade mounting medium (Vector Laboratories,

595 Burlingame, CA, USA, H-1000) and visualized on a Nikon TE2000 inverted microscope.

25 596 All antibody stains were incubated for 30 min at 37°C and all blocking steps were

597 incubated for 60 min at 37°C.

598 Importantly, the staining method described above results in intracellular bacteria

599 staining positive for TexasRed while extracellular bacteria are double positive for

600 Cascade Blue and TexasRed. Quantification of infectivity was undertaken by two

601 methods using the differential staining of intracellular and extracellular L pneumophila.

602 In experiments where differential contrast (DIC) microscopy, Cascade Blue, and

603 TexasRed were visualized counting of intracellular bacteria in BMMs was done by hand

604 using the image analysis software ImageJ (RRID:SCR_003070; Rasband, W.S., ImageJ,

605 U. S. National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/,

606 1997-2016) and the Cell Counter plugin (https://imagej.nih.gov/ij/plugins/cell-

607 counter.html). Uninfected BMMs were classified as BMMs that were not associated with

608 L. pneumophila or only associated with extracellular (Cascade Blue + TexasRed double

609 positive) bacteria. Infected BMMs were classified as macrophages containing at least one

610 intracellular L. pneumophila (Texas Red only), independent of the number of

611 extracellular bacteria associated with the BMM. In experiments where the cell membrane

612 of BMMs was labeled with FITC-conjugated wheat germ agglutinin along with DIC,

613 Cascade Blue, and TexasRed, analysis of infectivity was undertaken using the imaging

614 software Imaris (RRID:SCR_007370; Bitplane, Zurich, Switzerland). Using Imaris,

615 surfaces of BMMs were drawn on the FITC-conjugated wheat germ agglutinin channel to

616 mark individual BMMs. All extracellular bacteria were removed from analysis by

617 generating a new channel that subtracted the Cascade Blue channel from the TexasRed

618 channel, e.g. Intracellular Channel = TexasRed Channel – (Scaling Value x Cascade Blue

26 619 Channel). The scaling value was calculated by measuring the average pixel intensities in

620 each channel for double positive bacteria. As an example, if the TexasRed channel had an

621 average pixel intensity of 350 and the Cascade Blue channel was 3500 then the equation

622 would be: Intracellular Channel = TexasRed Channel – (0.1 x Cascade Blue Channel).

623 The outcome of this calculation is the generation of a channel that removes the TexasRed

624 signal of extracellular bacteria, thus allowing for analysis of bacteria that are only

625 intracellular. Lastly, using the Sortomato utility

626 (http://open.bitplane.com/tabid/235/Default.aspx?id=90) in Imaris, new cell surfaces

627 were drawn for cells that contained a signal in the new Intracellular channel (with double

628 positive bacteria removed), marking cells infected with an intracellular bacterium.

629 Surfaces were also drawn for cells that did not have a signal in the Intracellular channel,

630 marking uninfected BMMs or BMMs only associated with extracellular L. pneumophila.

631 Results were checked by eye to confirm that all surfaces accurately marked uninfected

632 and infected BMMs; the surfaces generated by Sortomato were used to quantify

633 infectivity.

27 634 Acknowledgments

635 We thank members of the Vance and Barton Laboratories for discussion and Justin De

636 Leon, Edward Roberts, and Matthew Krummel for technical assistance. We also thank

637 Justin De Leon, Mary Fontana, Shelley Starck, and Jeannette Tenthorey for critical

638 reading and discussion of the manuscript. This work was supported by the HHMI and

639 National Institutes of Health Grants AI063302, AI075039, AI080749, and CA195768.

640 The Vincent J. Coates Genomics Sequencing Laboratory at UC Berkeley is supported by

641 NIH S10 instrumentation Grants S10RR029667 and S10RR027303. R.E.V. is an HHMI

642 Investigator, and research in the Vance Laboratory was supported by investigatorships

643 from the Burroughs Wellcome Fund and the Cancer Research Institute.

644

645 Competing Interests

646 The other authors declare that no competing interests exist.

28 647 References

648 1. Li JJ, Bickel PJ, Biggin MD. System wide analyses have underestimated protein 649 abundances and the importance of transcription in mammals. PeerJ. 2014;2:e270. Epub 650 2014/04/02.

651 2. Schwanhausser B, Busse D, Li N, Dittmar G, Schuchhardt J, Wolf J, et al. Global 652 quantification of mammalian gene expression control. Nature. 2011;473(7347):337-42. 653 Epub 2011/05/20.

654 3. Breker M, Schuldiner M. The emergence of proteome-wide technologies: 655 systematic analysis of proteins comes of age. Nature reviews Molecular cell biology. 656 2014;15(7):453-64. Epub 2014/06/19.

657 4. Maier T, Guell M, Serrano L. Correlation of mRNA and protein in complex 658 biological samples. FEBS letters. 2009;583(24):3966-73. Epub 2009/10/24.

659 5. Vogel C, Marcotte EM. Insights into the regulation of protein abundance from 660 proteomic and transcriptomic analyses. Nature reviews Genetics. 2012;13(4):227-32. 661 Epub 2012/03/14.

662 6. de Sousa Abreu R, Penalva LO, Marcotte EM, Vogel C. Global signatures of 663 protein and mRNA expression levels. Molecular bioSystems. 2009;5(12):1512-26. Epub 664 2009/12/22.

665 7. Jovanovic M, Rooney MS, Mertins P, Przybylski D, Chevrier N, Satija R, et al. 666 Immunogenetics. Dynamic profiling of the protein life cycle in response to pathogens. 667 Science. 2015;347(6226):1259038. Epub 2015/03/07.

668 8. Finlay BB, McFadden G. Anti-immunology: evasion of the host immune system 669 by bacterial and viral pathogens. Cell. 2006;124(4):767-82. Epub 2006/02/25.

670 9. Mohr I, Sonenberg N. Host translation at the nexus of infection and immunity. 671 Cell Host Microbe. 2012;12(4):470-83. Epub 2012/10/23.

672 10. Lemaitre B, Girardin SE. Translation inhibition and metabolic stress pathways in 673 the host response to bacterial pathogens. Nat Rev Microbiol. 2013;11(6):365-9. Epub 674 2013/05/15.

675 11. Barry KC, Fontana MF, Portman JL, Dugan AS, Vance RE. IL-1alpha signaling 676 initiates the inflammatory response to virulent Legionella pneumophila in vivo. J 677 Immunol. 2013;190(12):6329-39. Epub 2013/05/21.

678 12. Belyi Y, Tabakova I, Stahl M, Aktories K. Lgt: a family of cytotoxic 679 glucosyltransferases produced by Legionella pneumophila. Journal of bacteriology. 680 2008;190(8):3026-35. Epub 2008/02/19.

29 681 13. Fontana MF, Banga S, Barry KC, Shen X, Tan Y, Luo ZQ, et al. Secreted 682 bacterial effectors that inhibit host protein synthesis are critical for induction of the innate 683 immune response to virulent Legionella pneumophila. PLoS Pathog. 2011;7(2):e1001289. 684 Epub 2011/03/11.

685 14. Shen X, Banga S, Liu Y, Xu L, Gao P, Shamovsky I, et al. Targeting eEF1A by a 686 Legionella pneumophila effector leads to inhibition of protein synthesis and induction of 687 host stress response. Cell Microbiol. 2009;11(6):911-26. Epub 2009/04/24.

688 15. Dunbar TL, Yan Z, Balla KM, Smelkinson MG, Troemel ER. C. elegans detects 689 pathogen-induced translational inhibition to activate immune signaling. Cell Host 690 Microbe. 2012;11(4):375-86. Epub 2012/04/24.

691 16. Iglewski BH, Liu PV, Kabat D. Mechanism of action of Pseudomonas aeruginosa 692 exotoxin Aiadenosine diphosphate-ribosylation of mammalian elongation factor 2 in vitro 693 and in vivo. Infect Immun. 1977;15(1):138-44.

694 17. Fontana MF, Vance RE. Two signal models in innate immunity. Immunol Rev. 695 2011;243(1):26-39. Epub 2011/09/03.

696 18. McEwan DL, Kirienko NV, Ausubel FM. Host translational inhibition by 697 Pseudomonas aeruginosa Exotoxin A Triggers an immune response in Caenorhabditis 698 elegans. Cell Host Microbe. 2012;11(4):364-74. Epub 2012/04/24.

699 19. Chakrabarti S, Liehl P, Buchon N, Lemaitre B. Infection-induced host 700 translational blockage inhibits immune responses and epithelial renewal in the Drosophila 701 gut. Cell Host Microbe. 2012;12(1):60-70. Epub 2012/07/24.

702 20. Asrat S, Dugan AS, Isberg RR. The frustrated host response to Legionella 703 pneumophila is bypassed by MyD88-dependent translation of pro-inflammatory 704 cytokines. PLoS Pathog. 2014;10(7):e1004229. Epub 2014/07/25.

705 21. Ingolia NT, Brar GA, Rouskin S, McGeachy AM, Weissman JS. The ribosome 706 profiling strategy for monitoring translation in vivo by deep sequencing of ribosome- 707 protected mRNA fragments. Nature protocols. 2012;7(8):1534-50. Epub 2012/07/28.

708 22. Ingolia NT, Ghaemmaghami S, Newman JR, Weissman JS. Genome-wide 709 analysis in vivo of translation with nucleotide resolution using ribosome profiling. 710 Science. 2009;324(5924):218-23. Epub 2009/02/14.

711 23. Ingolia NT, Lareau LF, Weissman JS. Ribosome profiling of mouse embryonic 712 stem cells reveals the complexity and dynamics of mammalian proteomes. Cell. 713 2011;147(4):789-802. Epub 2011/11/08.

714 24. Molofsky AB, Byrne BG, Whitfield NN, Madigan CA, Fuse ET, Tateda K, et al. 715 Cytosolic recognition of flagellin by mouse macrophages restricts Legionella 716 pneumophila infection. The Journal of experimental medicine. 2006;203(4):1093-104. 717 Epub 2006/04/12.

30 718 25. Ren T, Zamboni DS, Roy CR, Dietrich WF, Vance RE. Flagellin-deficient 719 Legionella mutants evade caspase-1- and Naip5-mediated macrophage immunity. PLoS 720 Pathog. 2006;2(3):e18. Epub 2006/03/23.

721 26. Ivanov SS, Roy CR. Pathogen signatures activate a ubiquitination pathway that 722 modulates the function of the metabolic checkpoint kinase mTOR. Nat Immunol. 723 2013;14(12):1219-28. Epub 2013/10/15.

724 27. Cairo G, Tacchini L, Schiaffonati L, Rappocciolo E, Ventura E, Pietrangelo A. 725 Translational regulation of ferritin synthesis in rat liver. Effects of chronic dietary iron 726 overload. The Biochemical journal. 1989;264(3):925-8. Epub 1989/12/15.

727 28. Pavitt GD, Ron D. New insights into translational regulation in the endoplasmic 728 reticulum unfolded protein response. Cold Spring Harbor perspectives in biology. 729 2012;4(6). Epub 2012/04/27.

730 29. Huang MT. Harringtonine, an inhibitor of initiation of protein biosynthesis. 731 Molecular pharmacology. 1975;11(5):511-9. Epub 1975/09/01.

732 30. Tscherne JS, Pestka S. Inhibition of protein synthesis in intact HeLa cells. 733 Antimicrobial agents and chemotherapy. 1975;8(4):479-87. Epub 1975/10/01.

734 31. Fresno M, Jimenez A, Vazquez D. Inhibition of translation in eukaryotic systems 735 by harringtonine. European journal of biochemistry / FEBS. 1977;72(2):323-30. Epub 736 1977/01/01.

737 32. Fontana MF, Shin S, Vance RE. Activation of host mitogen-activated protein 738 kinases by secreted Legionella pneumophila effectors that inhibit host protein translation. 739 Infect Immun. 2012;80(10):3570-5. Epub 2012/08/02.

740 33. von Moltke J, Ayres JS, Kofoed EM, Chavarria-Smith J, Vance RE. Recognition 741 of bacteria by inflammasomes. Annu Rev Immunol. 2013;31:73-106.

742 34. Hershey JW, Sonenberg N, Mathews MB. Principles of translational control: an 743 overview. Cold Spring Harbor perspectives in biology. 2012;4(12). Epub 2012/12/05.

744 35. Janssens S, Pulendran B, Lambrecht BN. Emerging functions of the unfolded 745 protein response in immunity. Nat Immunol. 2014;15(10):910-9. Epub 2014/09/19.

746 36. Tattoli I, Sorbara MT, Vuckovic D, Ling A, Soares F, Carneiro LA, et al. Amino 747 acid starvation induced by invasive bacterial pathogens triggers an innate host defense 748 program. Cell Host Microbe. 2012;11(6):563-75. Epub 2012/06/19.

749 37. Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus: NCBI gene 750 expression and hybridization array data repository. Nucleic acids research. 751 2002;30(1):207-10. Epub 2001/12/26.

31 752 38. Hou B, Reizis B, DeFranco AL. Toll-like receptors activate innate and adaptive 753 immunity by using dendritic cell-intrinsic and -extrinsic mechanisms. Immunity. 754 2008;29(2):272-82. Epub 2008/07/29.

755 39. Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, et al. Differential 756 gene and transcript expression analysis of RNA-seq experiments with TopHat and 757 Cufflinks. Nature protocols. 2012;7(3):562-78. Epub 2012/03/03.

758 40. Iwasaki S, Floor SN, Ingolia NT. Rocaglates convert DEAD-box protein eIF4A 759 into a sequence-selective translational repressor. Nature. 2016;534(7608):558-61. Epub 760 2016/06/17. 761 762

32 763 Figure and Table Legends

764

765 Fig. 1. Mapping of ribosome profiling reads to the genomic sequence of specific L.

766 pneumophila-induced genes of interest. (A-F) Ribosome footprint reads were mapped

767 to the genome and the number of footprints on the mRNAs for Gapdh (A), Csf2 (B), Gem

768 (C), Il23a (D), Il1a (E), and Il1b (F) was visualized. Numbers in parentheses show the

769 total read count of ribosome footprints found on the indicated transcript. Bracketed

770 numbers represent read count data range. Gray, uninfected BMMs. Red, ΔflaA-infected

771 BMMs. Green, ΔdotAΔflaA-infected BMMs. Blue, Δ7ΔflaA-infected BMMs

772

773 Fig. 2. mRNA superinduction controls the T4SS-dependent induction of host gene

774 expression in response to L. pneumophila. (A) The ratio of ribosome footprint and

775 RNAseq read counts for well-expressed transcripts (read count ≥100) in ΔflaA-infected

776 versus ΔdotAΔflaA-infected B6 BMMs was calculated for each annotated transcript (open

777 circles) in the dataset and plotted. (B-C) B6 BMMs were infected with ΔflaA or ΔdotA

778 ΔflaA L. pneumophila and at 6hrs post-infection proteins were measured in cell lysates

779 (B) or supernatants (C) by bead array. The T4SS-induction (ΔflaA/ΔdotAΔflaA) of

780 protein in supernatants (B) or lysates (C) and the T4SS-induction of mRNA (ΔflaA/Δ

781 dotAΔflaA) was plotted. Proteins were normalized to total protein levels measured by

782 BCA and RNAseq read counts was normalized to transcript length and the sum or their

783 respective mitochondrial protein coding genes. Data are averaged from four (A) or two

784 independent experiments (B-C). Orange circle, Il1a. Green circle, Il1b. Blue circle,

785 subset of inducible genes. Grey dotted line, y = x. Blue dotted line, linear regression

786 model. r2, coefficient of determination. See also Fig. 2- Source Data.

787

788 Fig. 3. Global induction of mRNAs and ribosome footprints in response to L.

789 pneumophila. (A-B) Ribosome footprint and RNAseq read counts were sorted for well-

790 expressed transcripts (read count ≥100) and normalized to the sum of their respective

791 mitochondrial protein coding genes. The ratio of ribosome footprint and RNAseq read

33 792 counts in (A) ΔflaA-infected and Δ7ΔflaA-infected B6 BMMs or (B) B6 or Myd88–/–

793 BMMs infected with ΔflaA L. pneumophila was calculated for each annotated transcript

794 (open circles) in the dataset and plotted. Data are averaged from two independent

795 experiments. Orange circle, Il1a. Green circle, Il1b. Blue circle, subset of inducible genes.

796 Grey dotted line, y = x. Blue dotted line, linear regression model. r2, coefficient of

797 determination. See also Fig. 3- Source Data.

798

799 Fig. 4. Ribosome occupancy does not explain the inducible innate immune response

800 to L. pneumophila. (A-D) Ribosome footprint and RNAseq read counts were sorted for

801 well-expressed transcripts (read counts ≥ 100) and normalized to CDS length and the sum

802 of their respective mitochondrial protein coding genes. The normalized read counts for

803 ribosome footprints and RNAseq for all well expressed annotated transcripts were plotted

804 for (A) uninfected, (B) ΔflaA, (C) ΔdotAΔflaA, or (D) Δ7ΔflaA L. pneumophila-infected

805 B6 BMMs. Red dots represent transcripts with low translation efficiency. Purple dots

806 represent a number of transcripts common to all conditions that appear to have

807 significantly higher ribosome occupancy. Data are averaged from three (A), four (B-C),

808 or two independent experiments (D). Orange circle, Il1a. Green circle, Il1b. Blue circles,

809 subset of inducible transcripts. Blue dotted line, linear regression model. Grey lines, 99%

810 prediction interval. r2, coefficient of determination. See also Table 1 and Fig. 4- Source

811 Data.

812

813 Table 1. Transcripts with ribosome occupancy eight times greater than the

814 condition average. Bolded, transcripts found in all conditions. Orange, transcripts found

815 in three conditions. Purple, transcripts found in two conditions. Data are averaged from

816 two independent experiments.

817

818 Fig. 5. Metagene profiles of L. pneumophila infected BMMs. (A-J) Metagene profiles

819 of uninfected (A-B), ΔflaA (C-D), ΔdotAΔflaA (E-F), Δ7ΔflaA (G-H) L.

820 pneumophila-infected B6 BMMs and a merge (I-J). Metagene profiles are depicted

34 821 relative to the translation start (A, C, E, G, J) and stop site (B, D, F, H, J). Metagene

822 analyses show peaks at every three nucleotides, corresponding to the codon-to-codon

823 shifts of the ribosome. Data are representative of two independent experiments (A-J).

824 Black line, uninfected. Red line, ΔflaA-infected. Green line, ΔdotAΔflaA-infected. Blue

825 line, Δ7ΔflaA-infected.

826

827 Fig. 6. L. pneumophila induced block of host protein synthesis occurs at the level of

828 translation initiation and elongation (A-H) Metagene profiles of B6 BMMs uninfected

829 (A-B) or infected with ΔflaA (C-D), ΔdotAΔflaA (E-F), or Δ7ΔflaA (G-H) L.

830 pneumophila in the presence (solid line) or absence (dashed line) of the drug

831 harringtonine to block translation initiation. Metagene profiles are depicted relative to the

832 translation start (A, C, E, G) and stop site (B, D, F, H). Data are representative of two

833 independent experiments (A-H). Solid line, no drug treatment. Dashed line, harringtonine

834 treatment. Black line, uninfected. Red line, ΔflaA-infected. Green line, ΔdotAΔflaA-

835 infected. Blue line, Δ7ΔflaA-infected.

836

837 Fig. 7. Immune-related genes do not have increased translation rates in response to

838 infection with L. pneumophila. (A-D) Read counts from paired samples treated with

839 harringtonine or left untreated were plotted for uninfected (A), ΔflaA (B), ΔdotAΔflaA

840 (C), or Δ7ΔflaA-infected (D) BMMs showing where cytokine-related transcripts (pink

841 circles; Csf1, Csf2, Cxcl1, Cxcl2, Dusp1, Dusp2, Ifnb1, Il10, Il12b, Il1a, Il1b, Il23a, Il6,

842 Lyz1, and Tnf) and housekeeping transcripts (blue circles; Gapdh, Rpl31, Rps17, and

843 Tuba1a) fall among all transcripts (black circles). Grey line, y=x. Data shown are

844 representative of two independent experiments. See Fig. 7- Source Data for individual

845 housekeeping and cytokine-related transcripts. 846

35 847 Supporting Information Captions

848

849 Fig. 1 – Supplement 1. Quanification of L. pneumophila infectivity. (A-B) ΔflaA or

850 ΔdotAΔflaA L. pneumophila infected BMMs were stained to mark extracellular (blue)

851 and all bacteria (red). (A) Representative image of ΔflaA L. pneumophila infected BMMs

852 showing extracellular (blue and red stain, blue arrow) and intracellular (red stain only,

853 yellow arrow) bacteria. (B) Individual BMMs (n=1,375) were analyzed for the presence

854 of at least one intracellular ΔflaA or ΔdotAΔflaA L. pneumophila bacterium. Image is the

855 same as in A with yellow circles marking infected cells and blue circles marking

856 uninfected cells. The average combined infectivity in these conditions is ~90%. See

857 supplemental methods for more details on counting.

858

859 Fig. 1 – Supplement 2. Ribosome profiling libraries show a strong bias in size

860 distribution. The fraction of total reads with a size of 26-34 nucleotides was plotted for

861 each ribosome profiling library used in this study. These graphs clearly show that the

862 ribosome profiling libraries used in this study have a strong bias for 27-28 nucleotide

863 fragments, consistent with the size of the footprint of the ribosome. Columns indicate

864 infection condition. Rows indicate BMM genotype and/or drug treatment.

865

866 Fig. 6 – Supplement 1. L. pneumophila induced block in host protein synthesis can

867 occur at the level of translation elongation and initiation. (A-B) Metagene profile plot

868 around the translation start (A) or stop (B) site of all harringtonine treated conditions

869 normalized to mitochondrial read counts of each condition. (C-F) Global weighted

36 870 averages across transcripts were calculated for BMMs left uninfected (C) or infected with

871 ΔflaA (D), ΔdotAΔflaA (E), or Δ7ΔflaA (F) L. pneumophila. Weighted averages were

872 generated by scaling each transcript’s ribosome occupancy profile according to the

873 average density from codon 250 to codon 349 and then averaging across the entire

874 condition. Transcripts with very low density in the 250-349 codon region (or shorter than

875 349 codons) are excluded from averaging. If the weighted average is less than 1, this

876 shows that this region has reduced ribosome footprints, while if the weighted average is

877 greater than 1 this shows that there are more ribosome footprints in this region. Following

878 a brief pulse of cells with harringtonine (red line) there is a change in the distribution of

879 ribosomes from the 5ʹ end of the mRNA to the 3ʹ end of the mRNA (i.e., as the ribosomes

880 continue to move 5ʹ to 3ʹ; C-F) as compared to untreated cells (blue line). This can be

881 seen by an increase in the weighted average at the 3ʹ end of mRNAs (C, E, F) but the lack

882 of this change shows a block in translation elongation (D). We also expect an

883 accumulation of ribosomes at the ATG following harringtonine treatment, which is also

884 seen as a peak in the weighted average at the start site (C-F). Data are representative of

885 two independent experiments (A-F).

886

887

888 Fig. 7 – Supplement 1. Individual mRNAs do not show evidence of preferential

889 translation. Cumulative read counts across the length of individual transcripts were

890 calculated and normalized to the sum of ribosome footprints of mitochondrial transcripts

891 in each respective library. Red = BMMs treated with harringtonine. Black = BMMs

37 892 untreated. Numbers in parentheses = total read counts. Rows = individual transcript.

893 Columns = infection condition.

894

38 A uninfected A ΔflaA A ΔdotAΔflaA A Δ7ΔflaA Riobosome Riobosome Riobosome Riobosome Gene Gene A Gene A Gene A Occupancy A Occupancy Occupancy Occupancy 1 Acta1 3137.40 Acta1 2489.70 Acta1 1633.44 Acta1 7105.62 2 Hist1h4f 1177.90 S100a11 1115.36 Hist1h4f 943.87 H2-Q7 3130.45 3 S100a11 870.51 Hist1h4f 1069.93 Rpl31 893.93 Hist1h4f 2195.70 4 Hist1h2aa 844.87 Rpl31 873.61 Hist1h2aa 822.96 Hist3h2ba 1715.98 5 Hist3h2bb-ps 699.92 Hist1h2ba 707.62 Hist3h2ba 625.90 H2afj 1524.09 Hist3h2bb- 6 Hist1h2ba 692.19 Hist3h2ba 670.32 S100a11 565.91 ps 1470.88 7 H2afj 686.47 Lyz1 533.22 Fus 532.18 Lyz1 1260.16 8 Cxcl3 675.23 Hist3h2bb-ps 524.22 Hist1h2ba 519.60 Hist1h2ba 1174.16 9 H2-T24 557.88 Fus 405.23 Lyz1 498.70 Cd52 1170.22 10 Lyz1 551.76 H2-T24 374.06 H2-Q7 371.10 Fus 1102.13 11 Hist3h2ba 509.95 Gm5803 345.78 Gm5803 368.22 H2-Q4 1022.17 12 Fus 480.25 H2-Q7 337.36 H2afj 356.99 Rpl38 1004.74 13 Saa1 475.25 Hist1h4i 302.55 Hist1h4i 348.34 Hist2h2ab 796.60 14 Gm5803 436.48 Cxcl3 281.86 Hist1h4k 318.08 H2-Q6 752.77 15 Hist1h4i 333.72 Saa1 265.08 Rrbp1 306.19 S100a11 717.99 16 Atp5e 315.00 Hist1h4n 244.74 Hist1h4j 301.73 Gm5803 692.32 17 Rrbp1 308.68 Rrbp1 225.85 Hist1h4a 298.67 Tmsb10 679.27 18 Mt1 308.36 Hist1h4j 218.84 Hist1h4h 295.99 H2-Q10 674.79 19 Hist1h4j 304.87 Hist1h4k 217.89 Hist1h4b 288.98 Rpl36 672.43 20 Hist1h4k 303.19 Atp5e 217.08 Hist1h4n 272.04 Mt1 672.29 21 Hist1h4h 293.11 H2-Q4 215.94 BC094916 259.08 Hist2h2bb 650.90 22 Hist1h4a 292.51 Hist1h4h 215.51 Hist1h4c 255.20 H2-Q7 629.56 23 Hist1h4b 280.60 H2afj 206.14 Cxcl3 252.49 H2-Q7 618.80 24 Gm11127 272.59 Hist1h4a 205.64 Atp5e 241.23 Atp5e 606.27 25 Hist1h4n 265.41 Hist1h4b 197.15 Saa1 220.10 H2-T24 601.41 26 Fkbp1a 264.22 Hist2h2bb 191.90 Myl12b 210.72 Rpl37 584.88 27 Hist1h4c Hist1h4c Hist1h4c 187.03 Gm7030 206.93 H2-T10 545.54 28 Gm7030 253.47 Mt1 185.28 Hist1h4i 529.15 29 Myl12b 247.71 Cd52 184.14 Gm11127 512.74 30 Rps17 234.41 Gm11127 183.08 Uqcrq 511.93 31 Cd52 231.77 Hist1h2bj 182.74 Emp1 494.39 Sh3bgrl 181.81 32 Hist1h2bf 484.53 33 Gm7030 481.28 34 Npc2 479.93 35 Hist1h2bj 478.33 36 Usmg5 477.21 37 Hmga2 468.10 895 Table 1

39 A B C uninfected uninfected [0-29] uninfected [0-4] (36,217) (3) (0)

ΔflaA ΔflaA ΔflaA (25,289) (1,585) (354)

ΔdotAΔflaA ΔdotAΔflaA ΔdotAΔflaA (37,002) (38) (6)

Δ7ΔflaA Δ7ΔflaA Δ7ΔflaA (18,545) (13) (27) [0-343]

Gapdh Csf2 Gem D E F uninfected (0) [0-4] uninfected [0-83] uninfected [0-259] (24) (56)

ΔflaA (95) ΔflaA ΔflaA (9,722) (21,019)

ΔdotAΔflaA (2) ΔdotAΔflaA ΔdotAΔflaA (594) (3,366)

Δ7ΔflaA (3) Δ7ΔflaA Δ7ΔflaA (4,474) (11,958)

Il23a Il1a Il1b start stop 5’ 3’ intron 5’ UTR CDS 3’ UTR exon exon

Fig 1 A T4SS-dependent induction 10

8 Il1a Il23a Il1b Csf2 Gem 6 inducible genes

4 Il1a Dusp2 Cxcl1 Il1b Tnf Il6 y = 0.8474x – 0.1691 2 2

(RNAseq Reads) Cxcl2

2 r = 0.7912 Dusp1 y = x log 0

–2 –2 0 2 4 6 8 10 log2(Ribosome Footprint Reads)

100 B BMM Lysates

Csf2 Il1a 10 Il1b

Il10 Cxcl1 Il1a Serpine1 Ccl2 Il1b 1 Vegfa Ccl4 Tnf Cxcl10 Ccl9 Il18 Vcam1 Ccl6 Ccl3 Il6 ( Δ flaA / dotA ) 0.1 Il12b y = 10^(1.002 log(x) – 0.8349) r 2 = 0.9934 Protein Lysates T4SS-Induction Protein Lysates

0.01 0.1 1 10 100 1000 mRNA T4SS-Induction (ΔflaA/ΔdotAΔflaA) 100 C BMM Supernatants

Il1a 10 Il1b Csf2 Il10 Serpine1 Tnf Ccl22 Il18 Il23a Kitl Il1b Il1a 1 Vegfa Ccl4 Cxcl1 Ccl6 Ccl2 Vcam1 Il12b Timp1 Il6 Ccl9 Cxcl10 Ccl3 ( Δ flaA / dotA ) 0.1 y = 10^(0.483 log(x) – 0.3259) r 2 = 0.5472 Protein Supernatants T4SS-Induction 0.01 0.1 1 10 100 1000 mRNA T4SS-Induction Fig 2 (ΔflaA/ΔdotAΔflaA) 7 effector-dependent induction A 8

6 Csf2 Il1a Il1b Il23a 4 inducible genes Gem Dusp2 2 Dusp1 Tnf Cxcl1 0 Cxcl2 Il1a Il1b (RNAseq Reads)

2 –2 Il6 y = 0.8492x – 0.0969 2 log r = 0.733 –4 y = x

–6 –6 –4 –2 0 2 4 6 8 log2(Ribosome Footprint Reads)

MyD88-dependent induction B 10

8 Il1a Cxcl1 Il1b 6 inducible genes Il1a Csf2 4 Cxcl2 Tnf Il6 Il23a 2 Dusp2 Il1b (RNAseq Reads) 2 0 Gem

log Dusp1 –2 y = 0.9126x – 0.2572 r 2 = 0.7995 y = x –4 –4 –2 0 2 4 6 8 10 log2(Ribosome Footprint Reads)

Fig 3

A 4 B 6 uninfected 5 3 4

2 3

2 1 1 Norm. Ribosome Reads 0 Norm. Ribosome Reads 0 2 4 6 8 2 6 8 4 12 22 24 28 30 32 34 36 10 14 16 18 20 26 10 12 –8 –6 –4 –2 −8 −2 −6 −4 –10 –28 –22 –20 –18 –16 –14 –12 −12 −10 −24 −26 stop start Position (Codons) Position (Codons)

C 4 D 6 ΔflaA 5 3 4

2 3

2 1 1 Norm. Ribosome Reads 0 Norm. Ribosome Reads 0 2 6 8 2 4 6 8 4 12 22 24 28 30 32 34 36 10 12 10 14 16 18 20 26 –8 –6 –4 –2 −8 −2 −6 −4 –10 –28 –22 –20 –18 –16 –14 –12 −12 −10 −24 −26 stop start Position (Codons) Position (Codons)

E 4 F 6 ΔdotAΔflaA 5 3 4

2 3

2 1 1 Norm. Ribosome Reads 0 Norm. Ribosome Reads 0 4 2 6 8 2 4 6 8 12 22 24 28 30 32 34 36 –8 –6 –4 –2 10 12 10 14 16 18 20 26 −8 −2 −6 −4 –28 –22 –20 –18 –16 –14 –12 –10 −26 −12 −10 −24 stop start Position (Codons) Position (Codons)

G 4 H 6 Δ7ΔflaA 5 3 4

2 3

2 1 1 Norm. Ribosome Reads 0 Norm. Ribosome Reads 0 4 2 4 6 8 2 6 8 12 22 24 28 30 32 34 36 –8 –6 –4 –2 10 14 16 18 20 26 10 12 −8 −2 −6 −4 –28 –22 –20 –18 –16 –14 –12 –10 −26 −12 −10 −24 stop start Position (Codons) Position (Codons)

I 4 J 6 uninfected 5 3 ΔflaA ΔdotAΔflaA 4 Δ7ΔflaA 2 3

2 1 1 Norm. Ribosome Reads 0 Norm. Ribosome Reads 0 2 4 6 8 2 6 8 4 12 22 24 28 30 32 34 36 10 14 16 18 20 26 10 12 –8 –6 –4 –2 −8 −2 −6 −4 –10 –28 –22 –20 –18 –16 –14 –12 −12 −10 −24 −26 stop start Fig 5 Position (Codons) Position (Codons) A Fig 6 G E C Norm. Ribosome Reads Norm. Ribosome Reads Norm. Ribosome Reads

Norm. Ribosome Reads 0 1 2 3 4 1 3 0 2 4 0 1 2 3 4 0 1 2 3 4 −12 −12 −12 −12 −10 −10 −10 −10 −8 −8 −8 −8 −6 −6 −6 −6 −4 −4 −4 −4 −2

−2 −2 −2 Position (Codons) Position (Codons) Position (Codons) Position (Codons) start start start start 2 2 2 2 4 4 4 4 6 6 6 6 8 8 8 8 uninfected +har. uninfected Δ7flaA +har. Δ7flaA ΔdotAflaA +har. ΔdotAflaA 10 10 10 10 ΔflaA +har. ΔflaA 12 12 12 12 14 14 14 14 16 16 16 16 18 18 18 18 20 20 20 20 22 22 22 22 24 24 24 24 26 26 26 26 28 28 28 28 30 30 30 30 32 32 32 32 34 34 34 34 36 36 36 36 H F D B Norm. Ribosome Reads Norm. Ribosome Reads Norm. Ribosome Reads Norm. Ribosome Reads 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 5 6 0 1 2 3 4 5

–28 –28 –28 –28 −26 −26 −26 −26 −24 −24 −24 −24 –22 –22 –22 –22 –20 –20 –20 –20 Position (Codons) Position (Codons) –18 Position (Codons) –18 Position (Codons) –18 –18 –16 –16 –16 –16 –14 –14 –14 –14 –12 –12 –12 –12 –10 –10 –10 –10 –8 –8 –8 –8 –6 –6 –6 –6 –4 –4 –4 –4 –2 –2 –2 –2 stop stop stop stop 2 2 2 2 4 4 4 4 6 6 6 6 8 8 8 8 10 10 10 10 12 12 12 12 A uninfected B ΔflaA 1M Genes 1M Genes Cytokines Cytokines Housekeeping Housekeeping 100k 100k 10k 10k 1k 1k 100 100 Harringtonine Reads Runoff Harringtonine Reads Runoff ...10 ...10

...10 100 1k 10k 100k 1M ...10 100 1k 10k 100k 1M Untreated Reads Untreated Reads C ΔdotAΔflaA D Δ7ΔflaA 1M Genes 1M Genes Cytokines Cytokines Housekeeping Housekeeping 100k 100k 10k 10k 1k 1k 100 100 Harringtonine Reads Harringtonine Reads Runoff Runoff ...10 ...10

...10 100 1k 10k 100k 1M ...10 100 1k 10k 100k 1M Untreated Reads Untreated Reads

Fig 7 l l A B l l Merge + DIC Merge l l l l l l l l l l l l l l l l l l l l l l l l l l l l l uninfected infected 120 All L.p. ExtracellularL.p. 110 100 90 80 70 60 50 40 30 20

% of macrophages 10 0 ΔflaA ΔdotAΔflaA

Figure 1– Supplement 1. Quantification of L. pneumophila infectivity. (A-B) ΔflaA or ΔdotAΔflaA L. pneumophila infected BMMs were stained to mark extracellular (blue) and all bacteria (red). (A) Representative image of ΔflaA L. pneumophila infected BMMs showing extracellular (blue and red stain, blue arrow) and intracellular (red stain only, yellow arrow) bacteria. (B) Individual BMMs (n=1,375) were analyzed for the presence of at least one intracellular ΔflaA or ΔdotAΔflaA L. pneumophila bacterium. Image is the same as in A with yellow circles marking infected cells and blue circles marking uninfected cells. The average combined infectivity in these conditions is ~90%. Data are combined from two independent experiments. See methods for more details on counting. uninfected ΔflaA ΔdotAΔflaA Δ7ΔflaA

RVKB007A RVKB007B RVKB007C 0.7 0.7 0.7 0.6 0.6 0.6 0.5 0.5 0.5 0.4 0.4 0.4 0.3 0.3 0.3 0.2 0.2 0.2 action of Reads action of Reads action of Reads B6 BMM 0.1 0.1 0.1 F r F r 0.0 F r 0.0 0.0 26 28 30 32 34 26 28 30 32 34 26 28 30 32 34 Length [nt] Length [nt] Length [nt] RVKB016A RVKB016C RVKB017A RVKB017C 0.5 0.5 0.5 0.5

0.4 0.4 0.4 0.4

0.3 0.3 0.3 0.3

0.2 0.2 0.2 0.2

0.1 0.1 0.1 0.1 action of Reads action of Reads action of Reads action of Reads B6 BMM F r 0.0 F r 0.0 F r 0.0 F r 0.0 26 28 30 32 34 26 28 30 32 34 26 28 30 32 34 26 28 30 32 34 Length [nt] Length [nt] Length [nt] Length [nt] RVKB016B RVKB016D RVKB017B RVKB017D 0.5 0.5 0.5 0.5

0.4 0.4 0.4 0.4

0.3 0.3 0.3 0.3

0.2 0.2 0.2 0.2

0.1 0.1 0.1 0.1 action of Reads action of Reads action of Reads action of Reads B6 BMM F r F r 0.0 0.0 F r 0.0 F r 0.0

+harringtonine 26 28 30 32 34 26 28 30 32 34 26 28 30 32 34 26 28 30 32 34 Length [nt] Length [nt] Length [nt] Length [nt] RVKB008A RVKB008B 0.5 0.5

0.4 0.4 0.3 0.3 Figure 1– Supplement 2. 0.2 0.2

0.1 0.1 Ribosome profiling action of Reads action of Reads B6 BMM F r 0.0 F r 0.0 libraries show a strong 26 28 30 32 34 26 28 30 32 34 Length [nt] Length [nt] bias in size distribution. RVKB008C RVKB008D The fraction of total reads 0.5 0.5 with a size of 26-34 0.4 0.4

0.3 0.3 nucleotides was plotted for –/– 0.2 0.2 each ribosome profiling 0.1 action of Reads 0.1 action of Reads

F r F r library used in this study. 0.0 0.0

Myd88 BMM 26 28 30 32 34 26 28 30 32 34 Length [nt] Length [nt] These graphs clearly show RVKB020A RVKB020B RVKB020C that the ribosome profiling 0.5 0.5 0.5 libraries used in this study 0.4 0.4 0.4 have a strong bias for 0.3 0.3 0.3 0.2 0.2 0.2 27-28 nucleotide 0.1 0.1 0.1 action of Reads action of Reads action of Reads B6 BMM fragments, consistent with F r F r 0.0 0.0 F r 0.0 26 28 30 32 34 26 28 30 32 34 26 28 30 32 34 the size of the footprint of Length [nt] Length [nt] Length [nt] the ribosome. Columns RVKB019A RVKB019B RVKB019C 0.5 0.5 0.5 indicate infection condition. 0.4 0.4 0.4 Rows indicate BMM 0.3 0.3 0.3

–/– genotype and/or drug 0.2 0.2 0.2

0.1 0.1 action of Reads action of Reads action of Reads 0.1 treatment. F r F r F r 0.0 0.0 0.0

Myd88 BMM 26 28 30 32 34 26 28 30 32 34 26 28 30 32 34 Length [nt] Length [nt] Length [nt] A 4 B 4 uninfected +har. uninfected +har. 3 ΔflaA +har. 3 ΔflaA +har. ΔdotAΔflaA +har. ΔdotAΔflaA +har. Δ7ΔflaA +har. Δ7ΔflaA +har. 2 2

1 1 Norm. Ribosome Reads 0 Norm. Ribosome Reads 0 2 4 6 8 2 6 8 4 12 12 22 24 28 30 32 34 36 10 14 16 18 20 26 10 –8 –6 –4 –2 −8 −2 −6 −4 –10 –28 –22 –20 –18 –16 –14 –12 −12 −10 −24 −26 stop start Position (Codons) Position (Codons) C uninfected D ΔflaA 4 4 Untreated Untreated +Harringtonine +Harringtonine 3 3 age age e r e r v v

2 2

eighted A 1 eighted A 1 W W

0 0 0 100 200 300 400 500 600 700 800 9001000 0 100 200 300 400 500 600 700 800 9001000 Position [codons] Position [codons] E ΔdotAΔflaA F Δ7ΔflaA 4 4 Untreated Untreated +Harringtonine +Harringtonine 3 3 age age e r e r v v

2 2

eighted A 1 eighted A 1 W W

0 0 0 100 200 300 400 500 600 700 800 9001000 0 100 200 300 400 500 600 700 800 9001000 Position [codons] Position [codons] Figure 6– Supplement 1. L. pneumophila induced block in host protein synthesis can occur at the level of translation elongation and initiation. (A-B) Metagene profile plot around the translation start (A) or stop (B) site of all harringtonine treated conditions normalized to mitochondrial read counts of each condition. (C-F) Global weighted averages across transcripts were calculated for BMMs left uninfected (C) or infected with ΔflaA (D), ΔdotAΔflaA (E), or Δ7ΔflaA (F) L. pneumophila. Weighted averages were generated by scaling each transcript’s ribosome occupancy profile according to the average density from codon 250 to codon 349 and then averaging across the entire condition. Transcripts with very low density in the 250-349 codon region (or shorter than 349 codons) are excluded from averaging. If the weighted average is less than 1 this shows that this region has reduced ribosome footprints, while if the weighted average is greater than 1 this shows that there are more ribosome footprints in this region. Following a brief pulse of cells with harringtonine (red line) there is a change in the distribution of ribosomes from the 5’ end of the mRNA to the 3’ end of the mRNA (i.e., as the ribosomes continue to move 5’ to 3’; C-F) as compared to untreated cells (blue line). This can be seen by an increase in the weighted average at the 3’ end of mRNAs (C, E, F) but the lack of this change shows a block in translation elongation (D). We also expect an accumulation of ribosomes at the ATG following harringtonine treatment, which is also seen as a peak in the weighted average at the start site (C-F).Data are representative of two independent experiments (A-F). uninfected ΔflaA ΔdotAΔflaA Δ7ΔflaA uninfected ΔflaA ΔdotAΔflaA Δ7ΔflaA ed) ed) ed) ed) ed) ed) ed) Untr (29 reads) ed) Untr (10433 reads) Untr (629 reads) Untr (4574 reads) Untr (4 reads) Untr (1741 reads) Untr (84 reads) Untr (301 reads) 1.0 1.0 1.0 1.0 6 6 6 Harr (55 reads) 6 Harr (42257 reads) Harr (662 reads) Harr (2070 reads) Harr (9 reads) Harr (6719 reads) Harr (100 reads) Harr (236 reads) mali z mali z mali z mali z mali z mali z mali z mali z 5 5 5 5 0.8 0.8 0.8 0.8 4 4 4 4 0.6 0.6 0.6 0.6 3 3 3 3 0.4 0.4 0.4 0.4 2 2 2 2 Il1a 0.2 0.2 0.2 0.2 1 1 1 1 ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r Cxcl1 0 0 0 0 0.0 0.0 0.0 0.0 Cu m Cu m Cu m Cu m Cu m Cu m Cu m Cu m

0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80

Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] ed) ed) ed) ed) ed) ed) ed) ed) 8 Untr (68 reads) 8 Untr (22449 reads) 8 Untr (3758 reads) 8 Untr (12442 reads) Untr (6 reads) Untr (281 reads) Untr (114 reads) Untr (158 reads) Harr (63 reads) Harr (53353 reads) Harr (2087 reads) Harr (7044 reads) Harr (1 reads) Harr (1054 reads) Harr (43 reads) Harr (63 reads) 0.15 0.15 0.15 0.15 mali z mali z mali z mali z mali z mali z mali z mali z 6 6 6 6 0.10 0.10 0.10 0.10 4 4 4 4 0.05 0.05 0.05 0.05 Il1b 2 2 2 2 ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r Dusp1 0 0 0 0 0.00 0.00 0.00 0.00 Cu m Cu m Cu m Cu m Cu m Cu m Cu m Cu m

0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 200 300 0 50 100 200 300 0 50 100 200 300 0 50 100 200 300

Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] ed) ed) ed) ed) ed) ed) ed) Untr (1 reads) Untr (293 reads) Untr (79 reads) Untr (1214 reads) ed) Untr (23 reads) Untr (6438 reads) Untr (964 reads) Untr (3849 reads) 0.12 0.12 0.12 0.12 Harr (0 reads) Harr (667 reads) Harr (49 reads) Harr (484 reads) Harr (28 reads) Harr (17723 reads) Harr (1038 reads) Harr (2405 reads)

mali z mali z mali z mali z mali z mali z mali z

z mali 2.0 2.0 2.0 2.0 0.08 0.08 0.08 0.08

Il6 1.0 1.0 1.0 1.0 0.04 0.04 0.04 0.04

ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r

r (mito−no reads ul

Cxcl3 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00

Cu m Cu m Cu m Cu m Cu m Cu m Cu m

m Cu

0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] ed) ed) ed) ed) ed) Untr (0 reads) ed) Untr (47 reads) ed) Untr (227 reads) ed) Untr (246 reads) Untr (2040 reads) Untr (1338 reads) Untr (1846 reads) Untr (1628 reads) Harr (0 reads) Harr (117 reads) Harr (75 reads) Harr (66 reads) Harr (170 reads) Harr (1552 reads) Harr (528 reads) Harr (272 reads) mali z mali z mali z mali z mali z mali z mali z mali z 0.20 0.20 0.20 0.20 0.020 0.020 0.020 0.020 0.10 0.10 0.10 0.10 0.010 0.010 0.010 0.010 ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r Il12b 0.00 Cu m 0.00 0.00 0.00 Cu m Cu m Cu m Cu m Cu m Cu m Cu m 0.000 0.000 0.000 0.000 Gm5803 0 50 100 200 300 0 50 100 200 300 0 50 100 200 300 0 50 100 200 300 0 50 100 200 300 0 50 100 200 300 0 50 100 200 300 0 50 100 200 300

Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] ed) ed) ed) ed) ed) ed) Untr (80 reads) Untr (43 reads) ed) Untr (49 reads) ed) Untr (301 reads) Untr (52620 reads) Untr (29537 reads) Untr (43274 reads) Untr (26698 reads) Harr (13 reads) Harr (60 reads) Harr (33 reads) Harr (24 reads) Harr (3696 reads) Harr (22725 reads) Harr (10669 reads) Harr (4552 reads) 0.030 0.030 0.030 0.030 4 4 4 4 mali z mali z mali z mali z mali z mali z mali z mali z 3 3 3 3 0.020 0.020 0.020 0.020 2 2 2 2 0.010 0.010 0.010 0.010 1 Il18 1 1 1 Lyz1 ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r 0 0 0 0 Cu m Cu m Cu m Cu m Cu m Cu m Cu m Cu m 0.000 0.000 0.000 0.000 0 50 100 150 0 50 100 150 0 50 100 150 0 50 100 150 0 50 100 150 0 50 100 150 0 50 100 150 0 50 100 150

Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] 150 ed) ed) ed) ed) ed) ed) ed) Untr (3 reads) Untr (1791 reads) Untr (42 reads) Untr (15 reads) ed) Untr (3699 reads) Untr (1858 reads) Untr (2569 reads) Untr (2448 reads) 1.0 1.0 1.0 1.0 Harr (8 reads) Harr (6695 reads) Harr (57 reads) Harr (28 reads) Harr (409 reads) Harr (2222 reads) Harr (420 reads) Harr (1062 reads) 0.30 0.30 0.30 0.30 mali z mali z mali z mali z mali z mali z mali z mali z 0.8 0.8 0.8 0.8 0.6 0.6 0.6 0.6 0.20 0.20 0.20 0.20 0.4 0.4 0.4 0.4 0.10 0.10 0.10 0.10 0.2 0.2 0.2 0.2 Csf2 ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 Cu m Cu m Cu m Cu m Cu m Cu m Cu m Cu m S100a11 0 20 40 60 80 120 0 20 40 60 80 120 0 20 40 60 80 120 0 20 40 60 80 120 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] ed) ed) ed) ed) ed) ed) ed) ed) 0.6 0.6 0.6 Untr (0 reads) Untr (386 reads) Untr (7 reads) Untr (28 reads) Untr (3644 reads) Untr (3146 reads) 0.6 Untr (4560 reads) Untr (5477 reads) Harr (2 reads) Harr (1781 reads) Harr (20 reads) Harr (6 reads) Harr (299 reads) Harr (3906 reads) Harr (1519 reads) Harr (724 reads) mali z mali z mali z mali z mali z mali z mali z mali z 0.20 0.20 0.20 0.20 0.4 0.4 0.4 0.4 0.10 0.10 0.10 0.10 0.2 0.2 0.2 0.2 Gem ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r Rpl31 0.0 0.0 0.0 0.0 Cu m 0.00 0.00 0.00 0.00 Cu m Cu m Cu m Cu m Cu m Cu m Cu m

0 50 100 150 200 250 300 0 50 100 150 200 250 300 0 50 100 150 200 250 300 0 50 100 150 200 250 300 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] ed) ed) ed) ed) ed) ed) ed) Untr (0 reads) Untr (100 reads) Untr (2 reads) Untr (4 reads) Untr (3891 reads) Untr (3199 reads) Untr (4574 reads) ed) Untr (6019 reads) 0.6 0.6 0.6 Harr (519 reads) 0.6 0.06 0.06 0.06 Harr (3 reads) 0.06 Harr (415 reads) Harr (4 reads) Harr (1 reads) Harr (4086 reads) Harr (2019 reads) Harr (1029 reads) mali z mali z mali z mali z mali z mali z mali z mali z 0.4 0.4 0.4 0.4 0.04 0.04 0.04 0.04 0.2 0.2 0.2 0.2 0.02 0.02 0.02 0.02 Il23a ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r Rps17 0.0 0.0 0.0 0.0 Cu m 0.00 0.00 0.00 0.00 Cu m Cu m Cu m Cu m Cu m Cu m Cu m

0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 0 20 40 60 80 100 140 0 20 40 60 80 100 140 0 20 40 60 80 100 140 0 20 40 60 80 100 140

Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons] Position [codons]

ed) Untr (18059 reads) ed) Untr (10983 reads) ed) Untr (13698 reads) ed) Untr (9813 reads) Harr (1978 reads) Harr (10350 reads) Harr (5215 reads) Harr (2191 reads) 1.5 1.5 1.5 1.5 mali z mali z mali z mali z 1.0 1.0 1.0 1.0 0.5 0.5 0.5 0.5 ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r ul reads (mito−no r 0 0 0 0 Tuba1a Cu m Cu m Cu m Cu m

0 100 200 300 400 0 100 200 300 400 0 100 200 300 400 0 100 200 300 400

Position [codons] Position [codons] Position [codons] Position [codons]

Figure 7– Supplement 1. Individual mRNAs do not show evidence of preferential translation. Cumulative read counts across the length of individual transcripts were calculated and normalized to the sum of ribosome footprints of mitochondrial transcripts in each respective library. Red = BMMs treated with harringtonine. Black = BMMs untreated. Numbers in parentheses = total read counts. Rows = individual transcript. Columns = infection condition. Data are representative of two independent experiments.