!1

Intracellular bacterial pathogens trigger the formation of U bodies through metabolic stress induction Jessica Tsalikis1,#, Ivan Tattoli1,2, #, Arthur Ling1, Matthew T. Sorbara2, David O. Croitoru1, Dana J. Philpott2 and Stephen E. Girardin1,*

1Department of Laboratory Medicine and Pathobiology, 2 Department of Immunology, University of Toronto,

M6G 2T6, Toronto, Canada.

Running title: Intracellular bacteria and metabolic stress trigger U bodies

*Address correspondence to: Stephen E. Girardin: Medical Sciences Building, Room 6336, University of Toronto, M5S 1A8, Toronto, Ontario, Canada. E-mail: stephen.girardin@utoronto.ca, Tel: (1) 416 978 7507, Fax: (1) 416 978 5959. # Contributed equally

Keywords: Shigella, Salmonella, Listeria, U snRNA, Metabolic stress, NF-κB

CAPSULE Background: The impact of metabolic stress on host response to bacterial infection remains poorly characterized. Results: Intracellular bacteria (Shigella, Salmonella, Listeria) induce cytoplasmic U bodies through metabolic stress. Conclusion: Bacterial infection and metabolic stress affect the splicing machinery. Significance: Regulation of U snRNA maturation is a novel checkpoint in innate immunity. ABSTRACT

Invasive bacterial pathogens induce an

amino acid starvation (AAS) response in infected host cells, which controls host defense in part by promoting autophagy. However, whether AAS has additional significant effects on the host response to intracellular bacteria remains poorly characterized. Here we showed that Shigella, Salmonella and Listeria interfere with spliceosomal U snRNA maturation in the cytosol. Bacterial infection resulted in the re-routing of U snRNAs and their cytoplasmic escort, the Survival Motor Neuron (SMN) complex, to processing bodies (P bodies), thus forming U snRNA bodies (U bodies). This process likely contributes to the decline in the cytosolic levels of U snRNAs and of the SMN complex proteins SMN and DDX20 that we observed in infected cells. U body formation was triggered by membrane damage in infected cells, and was associated with the induction of metabolic stresses, such as AAS or endoplasmic reticulum

(ER) stress. Mechanistically, targeting of U snRNAs to U bodies was regulated by translation initiation inhibition and the ATF4/ATF3 pathway, and U bodies rapidly disappeared upon removal of the stress, suggesting that their accumulation represented an adaptive response to metabolic stress. Importantly, this process likely contributed to shape the host response to invasive bacteria, since down-regulation of DDX20 expression using short hairpin RNA (shRNA) amplified ATF3- and NF-κB-dependent signaling. Together, these results identify a critical role for metabolic stress and invasive bacterial pathogens in U body formation and suggest that this process contributes to host defense.

RNA splicing is a fundamental process regulating gene expression, which involves the modification of nascent pre-messenger RNA via removal of introns and joining of exonic sequences. In eukaryotes, the machinery driving the splicing process depends on the spliceosome – a set of core catalytic RNA-protein complexes and more than 150 other protein splicing factors (1). Uridine-rich small nuclear ribonucleic acid (U snRNA) molecules function as building blocks of the spliceosome and associate with specific sets of proteins to form complexes known as uridine-rich small nuclear ribonucleoproteins (U snRNPs) (2-4). The SMN complex, which consists of 9 different proteins including Survival of Motor Neuron (SMN) and Gemin proteins 2-8, is directly involved in the cytoplasmic maturation of U snRNAs (5). The major spliceosome, which is made up of U snRNAs U1, U2,

http://www.jbc.org/cgi/doi/10.1074/jbc.M115.659466The latest version is at JBC Papers in Press. Published on July 1, 2015 as Manuscript M115.659466

Copyright 2015 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

!2

U4, U5 and U6, is responsible for splicing the majority of mRNA transcripts within the cell, whereas the minor spliceosome serves to splice a less frequent class of pre-mRNA introns and is comprised of U11, U12, U4atac and U6atac.

With the exception of U6 and U6atac, seven of the nine spliceosomal U snRNAs are exported into the cytoplasm via the CBC/PHAX/RanGTP transport complex guided by the m7G cap of the nascent U snRNAs (6). After association with free Gemin 5 in the cytoplasm, the U snRNAs are then directed to the SMN complex, which facilitates the assembly of a stable “ring”-shaped heptameric core consisting of Sm proteins B, D1, D2, D3, E, F and G. The SMN complex recognizes and binds to not only the Sm proteins, but also to the U snRNAs directly via stringent recognition of specific uridine-rich sequence and structural motifs (7). The resulting Sm+ U snRNPs undergo a maturation step in which the 5’ 7-trimethylguanosine cap is hypermethylated by the TGS1 enzyme to form a 2, 2, 7-trimethylguanosine (TMG) cap. The mature U snRNPs are then imported back into the nucleus and undergo further modifications in the Cajal bodies prior to assembly into the spliceosome complexes (8). It has been well studied that a primary means of energy conservation during times of cellular stress occurs by halting protein translation. One of the important stress pathways that drive protein translation arrest is the integrated stress response (ISR) pathway, which causes phosphorylation of the translation initiation factor eIF2α (9). Translation initiation involves the interaction of eIF2 with the initiator Met-tRNA and GTP to form a ternary complex, followed by hydrolysis of eIF2-GTP complex via interaction with the guanine nucleotide exchange factor eIF2B (9). Phosphorylation of the eIF2α subunit causes the protein to act as competitive inhibitor of guanine nucleotide exchange factor eIF2B, thus preventing the recycling of eIF2-GTP. This results in a translational blockade on the bulk of cellular transcripts, while paradoxically increasing expression of stress response genes such as transcription factors ATF4 and ATF3 (10). Consequently, expression of downstream adaptive response genes is induced, including those involved in amino acid (AA) synthesis, cellular redox status and regulation of autophagy and apoptosis (11,12). Four protein kinases are known to induce eIF2α phosphorylation in response to different environmental stresses: general controlled nonderepressible 2 (GCN2), heme-regulated inhibitor (HRI), protein kinase R (PKR), and PKR-like

endoplasmic reticulum kinase (PERK) (13). GCN2 and PERK are triggered by AA depletion and ER stress within the cell, respectively, while HRI responds to oxidative stress. Furthermore, the PKR pathway is activated by double stranded RNA from viruses and elicits a number of downstream antiviral responses (14). "

In addition to the ISR, signaling pathways dependent on the kinase mammalian target of rapamycin (mTOR) also play major roles in the control of gene expression in response to multiple metabolic cues, including AA and growth factor levels, oxygen tension or cellular ATP levels (15-17). Similar to the ISR, mTOR inhibition during metabolic starvation also results in translation initiation inhibition; while ISR induction targets eIF2α, mTOR inhibition blocks the initiation step of translation through the dephosphorylation of 4E-BP proteins (18).

We recently demonstrated that infection with intracellular bacterial pathogens resulted in the induction of an AA starvation response in host cells, and that this process was part of a defense program that promoted autophagy induction in infected host cells (19-22). In addition to autophagy, AA starvation in infected host cells resulted in a global inhibition of mTOR and induction of the GCN2/eIF2α/ATF3 signaling axis of the ISR. Interestingly, the kinetics of AA starvation during infection differed between bacteria; indeed, while infection with Shigella resulted in a sustained AA starvation (21), this effect was only transient (peaking at 1-2h post-infection) in Salmonella- and Listeria-infected cells (21,22). Mechanistically, the induction of an AA starvation response in infected cells was shown to be dependent on host membrane damage (19).

While the induction of AA starvation responses was shown to trigger protective autophagic responses in infected cells, it remains unclear if other cellular processes affected by AA starvation could play a role in host defense against bacterial pathogens. We speculated that this would likely be the case, given the critical importance played by mTOR signaling and the ISR pathway in the control of numerous cellular processes. Here we demonstrate that bacterial infection dramatically affected spliceosomal U snRNA maturation, and that this regulation was dependent on the induction of metabolic stress pathways. Importantly, our results further indicate that this regulation directly impacted on the induction of NF-κB- and ATF3-dependent signaling in infected cells. We propose that the control of U snRNA maturation, similar to translation

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

!3

initiation, represents an important checkpoint linking metabolic stress with host defense. EXPERIMENTAL PROCEDURES Antibodies and reagents Rabbit anti-ATF3 (sc-188), rabbit anti-ATF4 (sc-200), mouse anti-DDX20 (sc-57007), rabbit anti-DDX20 (H-145), mouse anti-TMG (K121), Santa Cruz Biotechnology; mouse anti-tubulin clone DM1A (T9026), Sigma-Aldrich; rabbit anti-mTOR (7C10), rabbit anti-Ge-1, Cell Signalling; rabbit anti-DDX6 (A300-460A), Bethyl Laboratories; mouse anti-puromycin (clone 4G11), Millipore; mouse anti-ribosomal rRNA 5.8S, Novus Biologicals; anti-SMN was from the Dr. Glenn Morris (Oswestry, UK); mouse anti-Sm (Smith antigen Y12, ab3138) was from Abcam; FITC-conjugated Goat anti-rabbit and Cy3-conjugated Goat anti-mouse, Jackson Immuno-Research Laboratories. 40, 6-diamidio-2-phenylindole (DAPI), Vector Laboratories; rapamycin, LKT Laboratories; thapsigargin, cycloheximide, puromycin, carbonyl cyanide m-chlorophenyl hydrazone, Sigma-Aldrich; pateamine A was a gift from Dr. Imed Gallouzi (McGill University); Krebs Ringer Bicarbonate (KRB) buffer (118.5 mM NaCl, 4.74 mM KCl, 1.18 mM KH2PO4, 23.4 mM NaHCO3, 5 mM glucose, 2.5 mM CaCl2, and 1.18 mM MgSO4, pH 7.6). Cell culture and bacterial strains The human epithelial HeLa cell line (American Type Culture Collection) was cultured as previously described (21). The bacterial strains Shigella flexneri (invasive M90T strain), Salmonella typhimurium (SL1344) and Listeria monocytogenes (10403S) were grown in Tryptic Soy Broth (Becton Dickinson), Luria-Bertani broth (Invitrogen), and Brain Heart Infusion broth (Becton Dickinson), respectively. Bacterial infections Overnight bacterial cultures of Shigella, Salmonella, and Listeria were used for infection as previously described (21). Immunofluorescence microscopy analysis Samples were fixed onto coverslips with 4% formaldehyde for 10 min at room temperature, rinsed three times in PBS for 5 min, and permeabilized via 0.1% Triton X-100 for 10 min, and incubated with antibodies as previously described (21).

Western blotting, RNA isolation and quantitative RT-PCR Western blotting, RNA isolation and quantitative RT-PCR were performed as previously described (21). shRNA lentiviral transduction shRNA sequences for transient lentiviral knockdown were cloned into the pLKO.1 vector (Addgene) and transfected along with the lentiviral packaging/envelope vectors psPAX2 and pMD2.G into HEK 293T cells using Lipofectamine2000 (Life Technologies). Supernatants were collected 48 hours post-transfection and HeLa cells were transduced with 1-2 ml of lentiviral particles. The cells were selected with puromycin 24 hours post-transduction and harvested after 3-4 days of selection. The following sequences were used: ATF3 5’-TCACAGGAAGAAAGCAGAAAGTTCA-3’, ATF4 5’-CCTCAGTGCATAAAGGAGGAA-3’, DDX20 5’-GAATTCCAGTGATCCAAGTCT-3’ and 5’-GCACAGCAGAGCACAACATTT-3’. U body quantification Cells with 2 or more U bodies for each condition were manually quantified upon immunofluorescence staining and represented as a percentage over the total number of cells counted. For each analysis, at least 100 cells from randomly selected fields were counted for each time point and condition, in at least three independent experiments. Results are expressed as means ± SEM of data obtained in these independent experiments. Surface sensing of translation (SUnSEt) assay The SUnSEt assay was conduced as previously described (Schmidt et al., 2009). Cells were stimulated with either thapsigargin, KRB, cycloheximide or infected with Shigella. After 4 hours, puromycin (10 µg/ml) was added directly to the cells for 10 minutes. Cells were then harvested and analyzed via western blot or immunofluorescence as described above. Puromycin antibody was used at 1:10,00 dilution for WB and 1:12,500 for IF. Super-resolution microscopy Images were acquired on a Zeiss Elyra SP1 super-resolution microscope using Structured Illumination Microscopy (SIM) imaging mode. The microscope was equipped with a 63x/1.4 Plan-Apo objective, with a zoom factor of 1.7, producing a final effective voxel size of 40nm lateral and 90nm axial. Rendering was done with Imaris Bitplane software, using a

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

!4

surface detail level of 10nm and automatic thresholding. Statistical analysis Significant differences between mean values were evaluated using a one-sample or unpaired Student’s t-tests using Prism 5.0. All the experiments presented are representative or pooled from at least three independent experiments. RESULTS Bacterial infection affects U snRNA levels and splicing and induces U bodies Spliceosomal U snRNAs are transiently exported to the cytosol after synthesis; at which point the U snRNAs are escorted by proteins of the SMN complex and receive a 2,2,7-trimethylguanosine (TMG) cap that is unique to this class of RNAs (2). Using an antibody recognizing the TMG cap of U snRNAs, we serendipitously observed that human epithelial HeLa cells infected with the invasive bacterial pathogen Shigella flexneri displayed reduced levels of nuclear TMG staining (Fig. 1A), indicative of decreased levels of mature U snRNAs. The decline in U snRNA levels was particularly evident in heavily infected cells (Fig. 1A, arrows). Using quantitative PCR (qPCR), we further observed that while the total level of all U snRNAs was down-regulated, albeit modestly, in Shigella-infected cells, there was a dramatic reduction of cytosolic U snRNAs that co-immunoprecipitated with SmB/B’ of the Sm complex in infected cells (Fig. 1B), suggesting that Shigella infection likely inhibited the cytosolic stage of U snRNA maturation. In agreement, the cytosolic levels of both DDX20, a component of the SMN complex also known as Gemin 3, as well as against the SMN protein were decreased upon Shigella infection, while the nuclear levels remained unchanged (Fig. 1C).

The above results prompted us to analyze further the impact of Shigella infection on the U SnRNA-interacting proteins of the SMN and Sm complexes. Using antibodies against DDX20 and SMN in immunofluorescence experiments, we observed in uninfected conditions that DDX20 and SMN stainings were diffuse in the cytosol and also found in discrete nuclear foci known as gems (Fig. 2A, top). Strikingly, infection with Shigella resulted in the accumulation of bright cytosolic DDX20+ and SMN+ foci in infected cells (Fig. 2A, bottom). In heavily infected cells, SMN+ cytosolic foci were found in greater number than DDX20+ foci; DDX20+ foci were always also SMN+. DDX20+ foci also

contained the U snRNA-associated protein SmB/B’ (Fig. 2B) and were partially positive for TMG cap staining (Fig. 2C), suggesting that these cytosolic foci contain maturing U snRNAs associated with the SMN complex and the heptameric Sm ring. However, it is worth noting that TMG staining was faint in these foci, and often even undetectable (data not shown), suggesting that DDX20/SMN+ foci did not always contain U snRNAs. Moreover, we observed that the number of DDX20+ foci increased over time in Shigella-infected cells (Fig. 2D). DDX20+ foci were also identified in the cytosol of the human intestinal epithelial cell line SW480 following infection with Shigella (data not shown).

A recent report showed that U snRNA could be degraded in cytosolic processing bodies (or P bodies) (23). Since infection with Shigella resulted in decreased levels of U snRNAs and a strong decline in Sm complex-associated cytosolic U snRNAs (see above), we aimed to identify if cytosolic DDX20+ foci were associated with P bodies. Using an antibody against DDX6, a marker of P bodies, we observed that while DDX20 was absent from DDX6+ P bodies in uninfected condition (Fig. 3A), a remarkable colocalization of DDX20+ foci with P bodies was observed in cells infected with Shigella for 1h (Fig. 3B), thus strongly suggesting that SMN complex-associated U snRNAs are re-routed to P bodies for degradation during infection. Super-resolution imaging identified various degrees of fusion between DDX20+ foci and P bodies (Fig. 3C). A similar cytosolic structure containing the SMN complex and associated with P bodies has been previously reported in Drosophila ovarian nurse cells and was named U body (24). As we believe the DDX20+ foci observed here in mammalian cells are equivalent to U bodies found in Drosophila ovaries, we used the latter term herein.

While the colocalization between U and P bodies was obvious in the early times of infection (1-2h post-infection (p.i.)), we also noticed that the number of DDX6+ P bodies sharply declined at 3-4h p.i., while U bodies continued to accumulate (Figs. 4A-B). This observation was confirmed using an antibody against Ge-1, another marker of P bodies (Fig. 4C). Quantification of DDX6+ P bodies in Shigella-infected cells revealed that P body decline started as early as 1h p.i. (Fig. 4D). Although the molecular mechanism underlying this P body decline remains unclear, we noticed that mitochondrial stress induced by the H+ ionophore carbonyl cyanide m-chlorophenyl hydrazone (CCCP) also resulted in P body disappearance (Fig. 4E); because we previously

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

!5

noticed that Shigella infection causes mitochondrial stress (25), it is possible that Shigella alters P body number as a result of mitochondrial stress.

SGs represent another type of cytosolic RNA granules that are induced by stress. Our previous work demonstrated that SGs are induced by Shigella infection (21) and accumulate with similar kinetics as U bodies, indicating a possible link between these two types of RNA granules. However, simultaneous visualization of SGs and U bodies using antibodies against Tia-1 and DDX20, respectively, revealed that these structures were mainly mutually exclusive in infected cells (Fig. 4F), as SGs and U bodies were rarely observed in the same cells. Together, these results identify U bodies as a new type of cytosolic structures induced by infection and define their relationship with the well-characterized P bodies and SGs.

We next aimed to determine whether U body formation was specific to Shigella infection. Interestingly, we observed that U bodies formed in cells infected with Salmonella (Fig. 5A) and Listeria (Fig. 5B) although, in contrast to Shigella infection, the number of Salmonella- and Listeria-induced U bodies peaked at 2-3h p.i., and declined sharply by 4h p.i. (Figs. 5 C, D).

Because we previously showed that Shigella induces a sustained inhibition of mTOR signaling (21), while this effect was only transient in both Salmonella- and Listeria-infected cells (21,22),! we hypothesized that U body formation in infected cells could be driven by mTOR inhibition. In support for this, U body accumulation in Salmonella-infected cells correlated with a decline in the localization of mTOR to endomembranes (Fig. 6A), indicative of mTOR signaling inhibition caused by amino acid (AA) starvation (26).!

Intracellular bacteria trigger host membrane damage, resulting in the induction of an AA starvation pathway dependent on inhibition of mTOR (21). In order to directly demonstrate that this pathway was responsible for the formation of U bodies in infected cells, we used a Salmonella mutant strain, ΔSPI-1/Inv, which invades epithelial cells efficiently but does not induce Salmonella Pathogenicity Island (SPI)-1-dependent membrane damage (19). While wild type (WT) Salmonella triggered U body formation in cells accumulating the membrane damage marker NDP52, infection with ΔSPI-1/Inv Salmonella resulted in dramatically reduced number of U bodies (Figs. 6B,C), thus suggesting that Salmonella induces U bodies through a pathway dependent on membrane damage. In

agreement, aseptic membrane damage caused by digitonin, transfection reagents or the lysosomal damaging drug glycyl-L-phenylalanine 2-naphthylamide (GPN) all triggered accumulation of U bodies (data not shown). Finally, in order to directly demonstrate that the reduced levels of U bodies in cells infected with ΔSPI-1/Inv Salmonella was not due to active suppression of U bodies by the bacteria but caused by the reduced level of membrane damage, cells were infected with ΔSPI-1/Inv Salmonella in the presence of GPN. Addition of GPN was sufficient to trigger U body formation (Fig. 6D), thus demonstrating that U body formation in Salmonella-infected cells is caused by membrane damage.

Metabolic stress induces reversible accumulation of U bodies The above results suggest a link between bacteria-induced AA starvation response, mTOR inhibition and perhaps other stimuli that induce metabolic stress response pathways in the induction of U bodies. Consistently, incubation of cells in an AA starvation medium or treatment with the endoplasmic reticulum (ER) stress inducer thapsigargin both potently triggered the formation of U bodies, which co-localized with P bodies, as well as a decrease in TMG staining (Figs 7A,B). Importantly, addition of the mTOR inhibitor rapamycin also triggered U bodies (Fig. 7C). In contrast, viral infection, stimulation with TNF or the bacterial peptidoglycan molecules MDP and C12-iE-DAP, failed to induce U bodies (Fig. 7C and data not shown). Moreover, other cellular stresses, such as serum starvation, mitochondrial depolarization, hypoxia, PI3K inhibition and oxidative stress did not induce U bodies (data not shown), suggesting that this process was specifically associated with certain metabolic stresses. Interestingly, we also observed that the accumulation of U bodies was reversible, as replacement of the AA starvation buffer with normal medium resulted in rapid disappearance of U bodies (Fig. 7D). Together, these results demonstrate that the targeting of U snRNPs to P bodies is a reversible cellular response to metabolic stress inducers, including AA starvation, ER stress and mTOR inhibition.

U body formation is dependent on translation initiation inhibition and ATF4/ATF3 expression Down-regulation of mTOR activity, AA starvation and ER stress, stimuli that all trigger U body formation, have the common consequence of inhibiting translation initiation (Fig. 8A), but it is

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

!6

unclear if infection with intracellular pathogens elicits a similar blockade. Using a SUnSET assay to monitor translation activity (27), we observed that Shigella infection and metabolic stress conditions indeed caused translation inhibition (Figs. 8B-C). Interestingly, treatment with pateamine A, a specific inhibitor of translation initiation, was sufficient to trigger U bodies in the absence of infection or metabolic stress (Fig. 8D). In contrast, the translation elongation inhibitors cycloheximide and puromycin failed to induce U bodies (Fig. 8D), demonstrating that down-regulation of the initiation phase of translation, and not translation inhibition in general, is critical for this process. Notably, in contrast to cells infected with Shigella (see above Fig. 4F), most cells treated with pateamine A accumulated both SGs and U bodies, and moreover, a significant proportion of the U bodies induced by pateamine A treatment appeared to localize adjacent to cytoplasmic SGs (Fig. 8E), suggesting a potential functional link between U bodies and SGs during translation initiation inhibition.

A well-characterized consequence of translation initiation inhibition is the activation of the ATF4/ATF3 signaling axis (28). While knockdown of ATF3 expression (Fig. 9A) did not affect the number of DDX20-positive nuclear gems, it significantly decreased the number of cytosolic U bodies in response to metabolic stress (Figs 9B-D); similar results were obtained following ATF4 knockdown (Figs 9E,F). Thus, the ATF4/ATF3 signaling axis controls U body formation in response to metabolic stress and translation initiation inhibition.

Knockdown of the U body component DDX20 amplifies ATF4/ATF3- and NF-κB-dependent signaling The above results suggest that the modulation of U snRNA cytosolic maturation represent an adaptive cellular response to metabolic stress and bacterial infection. In order to further delineate the functional importance of targeting cytosolic U snRNA maturation and levels during metabolic stress, we directly targeted the U snRNA maturation step by knocking-down the expression of the essential SMN complex component DDX20 (Fig. 10A) using short hairpin RNA (shRNA). Interestingly, cells expressing shRNA against DDX20 (shDDX20) displayed increased expression of ATF4 in response to Shigella infection or stimulation with the ER stress inducer thapsigargin as compared to control cells expressing shRNA against a scramble sequence (shSCR) (Fig. 10B). Similarly, ATF3 expression was increased in

shDDX20 cells (Fig. 10C) in response to Shigella, thapsigargin and AA starvation, thus showing that targeted down-regulation of SMN complex function results in increased cellular responses to metabolic stress. This further suggests the existence of a positive regulatory feedback loop amplifying the activation of the ATF4/ATF3 axis through reduced U snRNA maturation during metabolic stress.

In addition to the ATF4/ATF3 signaling axis, infection with Shigella results in a potent activation of a pro-inflammatory pathway dependent on NF-κB. Using an NF-κB-dependent luciferase reporter construct, we observed that NF-κB activation by Shigella (Fig. 10D) and MyD88 (Fig. 10E) over-expression was increased in shDDX20 cells compared to shSCR cells, in agreement with a previous study that identified a role for DDX20 in NF-κB signaling (29). This potentiation effect was specific to NF-κB since induction by the antiviral protein MAVS of a luciferase reporter whose expression is regulated by an interferon-sensitive response element (ISRE) was similar in shDDX20 and shSCR cells (Fig 10F). Thus, targeting the U snRNA maturation step through knockdown of DDX20 amplifies cellular stress and pro-inflammatory responses to the bacterial pathogen Shigella, suggesting a role for this process in innate immunity. DISCUSSION

Here we report that upon metabolic stress,

various components of the splicing machinery are re-localized into cytosolic granules known as U bodies. Furthermore we show that ER stress, AA starvation and bacterial infection result in a decrease in cytosolic U snRNA levels. Because of their co-localization with cytosolic P bodies, it is possible that U snRNAs are being sent to P bodies for degradation as a means of fine-tuning their availability for splicing during times of stress. In support of this, Parker et al. recently demonstrated that defective U snRNAs are degraded within P bodies in yeast and that this interaction represents a major quality control mechanism for assembly-defective U snRNAs. It is possible that even minute fluctuations in the homeostatic ratios of U snRNAs to mRNAs present within the cell could have global effects on splicing, and U bodies may thus represent a means of restoring this balance during times of metabolic stress, when transcription and mRNA levels drop. The fact that U bodies co-localize with P bodies (24) and are regulated by nutrient starvation (30) in Drosophila

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

!7

cells suggests that the process characterized here is conserved in evolution.

We also showed that localization of U snRNAs and other core spliceosomal factors into cytosolic U bodies is a reversible phenomenon, as the percent of U body-positive cells observed in AA–starvation condition decreased to control levels upon the addition of DMEM complete media. The dependency of U bodies on a persistent metabolic stress raises the possibility that U snRNAs are sent to U bodies to be stored rather than degraded - a process akin to that by which mRNA transcripts sent to stress granules (SGs) for storage as a means of more rapid translation of those transcripts once the stress has been relieved. Hua et al. demonstrated that SMN localizes to stress granules in the cytosol and is implicated in their assembly during times of oxidative stress (31). The frequent positioning of U bodies adjacent to stress granules in the cytosol (see Fig. 8E) supports the notion that U bodies and SGs can exchange material, although further work would be needed to elucidate whether this were the case.

In addition to the maturation and assembly of U snRNAs, there is also evidence suggesting that DDX20 may play a role in regulating gene expression via microRNAs. Indeed, it is interesting to note that DDX20/Gemin3 has been identified early on as a core member of the RNA-induced silencing complex (RISC) (32,33), a complex that is essential for microRNA-mediated RNA silencing. The fact that RISC members such as Argonaute-2 have been observed in both P bodies and stress granules, and that U bodies often co-localize with these structures, suggest that perhaps DDX20 plays a splicing-independent role, utilizing its helicase function in directing microRNAs into the RISC complex. If this were the case, the formation of U bodies could represent a way to regulate the microRNA machinery in a metabolic stress-dependent manner by incorporating the RNA helicase DDX20 into the RISC. Whether metabolic stress regulates microRNA activity in a DDX20-dependent manner remains to be characterized.

Consistent with the notion that DDX20 expression can regulate microRNA activity, Takata et al. observed that DDX20 knockdown resulted in impaired function of the NF-κB-suppressive microRNA miR-140-3p (29). Using luciferase reporter plasmids, it was observed that DDX20 deficient cells display increased NF-κB activity, as well as increased expression of IL-8, in response to TNFα stimulation (29), and this enhancement of NF-

κB activity was attributed to impaired suppressive function of the microRNA-140.

It is well characterized that inhibition of translation initiation serves as a critical checkpoint in the regulation of metabolic stress, since it represents the converging point of stress-associated pathways regulated by the eIF2α kinases and by mTOR signaling. This regulatory system plays a fundamental role in cellular adaptation to nutrient stress and is conserved from yeast to mammals. Importantly, the translation initiation checkpoint serves two major roles: (i) to down-regulate the energy-demanding translation activity in conditions of limited access to nutrients, thereby re-affecting resources to other vital cellular processes; (ii) to trigger cellular stress responses pathways, such as the ATF4/ATF3 signaling axis, which favor cellular adaptation to metabolic stress in part through the upregulation of processes aiming at increasing access to nutrient, such as the expression of AA transporters. Our data suggest that the dynamic regulation of U snRNAs at the level of their cytosolic maturation may represent another important checkpoint during metabolic stress. While the main purpose of this regulation might be the fine-tuning of U snRNA levels to avoid unnecessary use of metabolites in nutrient-poor conditions, our results with shDDX20 cells suggest that, similar to the translation initiation checkpoint, the control of U snRNA maturation also likely contributes to the cellular adaptation to metabolic stress and to host defense, by controlling ATF3- and NF-κB-dependent pathways. In this context, it is interesting to notice that U body formation is dependent on translation initiation inhibition, suggesting that these two checkpoints are tightly interconnected.

We report here that the re-localization of the splicing machinery into U bodies is also induced upon infection with the intracellular bacteria Listeria, Salmonella, and Shigella. Notably, the kinetics of U body induction parallels the kinetics of starvation induced by the intracellular lifestyles of the pathogens (19,21,22). We observed that during Shigella infection, similar to the progressive increase in the phosphorylation levels of GCN2 and eIF2α, U body induction increased progressively throughout the infection. Listeria and Salmonella in contrast induce a transient state of amino acid starvation that peaks at 1-2 hours p.i., a trend we report to also be the case for U body formation (19, 21, 22). Our data suggests that the mTOR pathway, which is inhibited upon infection and is associated with cell stress, energy and growth, is also implicated in U body formation, as inhibition

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

!8

of mTORC1 via rapamycin is a potent inducer of U bodies and inhibition of the mTOR pathway in response to Shigella, Salmonella or Listeria infections also occur with kinetics that correlates with U body formation (21). Lastly, membrane damage induced by Salmonella type 3 secretion system virulence factors or aseptically via digitonin or GPN, appears to be required for both starvation response and U body induction.

Notably, we observed that Shigella infection not only results in an accumulation of cytosolic U bodies, but also leads to disassembly of P bodies by 4 hours p.i.. This finding is consistent with a report by Eulalio et al. in which the authors observed that both Salmonella and Shigella infections resulted in a significant reduction in the number of cytosolic P bodies, a result that was most notable at mid- to late-phases of infection (34). Interestingly, staining for Tia-1 positive SGs in Shigella-infected cells revealed that the formation of U bodies and SGs appear to be mutually exclusive events during infection. Notably, this phenomenon was not observed upon treatment with the translational inhibitor pateamine A, which induced both U bodies and SGs with the structures often localized adjacent to one another. It would be of interest to examine the localization of other SG markers such as eIF4F, eIF4B, or PABP, to confirm whether mutual exclusivity with U bodies upon infection is representative of the global population of SGs or merely a subset of Tia-1 positive SGs (35). There have been many reported examples of Shigella manipulation of host functions, such as protein synthesis, and given that Shigella induces translational arrest upon infection, it would be interesting to explore whether pathogens exploit the P body/SG/U body axis to subvert host defenses. By providing a novel link between U snRNA levels, the SMN complex and cellular adaptation to metabolic stress, our data might have implications for the understanding of SMA, a severe neurodegenerative disease that results from a genetic defect in SMN1, which encodes SMN (2). Importantly, variable residual expression of SMN is observed in SMA patients despite SMN1 defects because of the existence of SMN2, a duplicated gene encoding an identical SMN protein, and pathology severity inversely correlates with the levels of residual SMN expression from SMN2. SMN2 expression only partially rescues SMN1 defect because of a silent mutation affecting alternative splicing of exon 7, which generates a shorter and unstable protein (36). Based on our data, one might speculate that severe decrease in SMN levels would

alter the cellular capacity to physiologically and reversibly regulate U snRNA levels during metabolic stress, resulting in increased sensitivity to metabolic stress. It would be of interest to determine if metabolic stress is a critical driver of neuronal cell death in SMA. In support for this, several recent findings have highlighted a link between sensitivity of SMA cells and metabolic pathways, including those dependent on mTOR (37), PTEN (38) and autophagy (39,40). In light of our results, we speculate that increasing autophagy would reduce the sensitivity of SMA cells undergoing AA starvation by providing an additional source of AA. Similarly, increasing the activity of neurotrophic cells that provide nutrients to the motor neurons of the spinal cord could contribute to protect these cells from cell death in SMA patients. Another interesting angle would be to characterize the effect of metabolic stress on the usage versus skipping of SMN2 exon 7. Notably, a recent report showed that SMN2 splicing defect was exacerbated by starvation (41), reinforcing the notion that metabolic stress might be a key aspect of neuronal sensitivity in SMA. Although the mechanisms that control and coordinate the activity of the splicing machinery remain poorly defined, regulated expression of splicing-associated factors, such SF2/ASF (42), polypyrimidine tract-binding proteins (43) or SR proteins (44) can affect splicing globally. In contrast, the synthesis and cytosolic maturation of U snRNPs, which represent essential building blocks of the spliceosome, is generally believed to be constitutively active and not subject to extensive cellular regulation, although a previous study showed that the activity of the SMN complex was regulated by oxidative stress (45). Interestingly, widespread pre-mRNA splicing defects in numerous transcripts have been reported in SMN-deficient tissues and can be attributed to alterations in the levels of mature U snRNAs (46), suggesting that the dynamic control of U snRNA levels might play a role in splice site determination, and thereby in alternative splicing regulation. In this regard, our study reveals a possible new mechanism through which the cellular splicing program could be globally altered, which is potentially dependent on the regulation of U snRNA levels through redirection of the cytosolic pool of maturing U snRNAs towards degradative P bodies, resulting in the formation of U bodies. Future studies should reveal whether the control of U snRNA levels and U body formation result in a global reprogramming of alternative splicing in conditions of metabolic stress.

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

!9

ACKNOWLEDGMENTS This work was supported by grants from Canadian Institutes of Health Research (D.J.P and S.E.G). We thank Dan Stevens (Zeiss) for assistance with super-resolution microscopy. CONFLICT OF INTEREST The authors declare that they have no conflicts of interest with the contents of this article. AUTHOR CONTRIBUTIONS SEG conceived and coordinated the study and wrote the manuscript. IT assisted in the conception of the paper, as well as designed, performed and analyzed the experiments shown in Figures 1-6 and 10. JT designed, performed and analyzed the experiments shown in Figures 1, 3, 5, 7-9 and assisted in writing of the paper. AL performed experiments in Figures 1 and 9. DOC and MTS conducted experiments in Figures 10 and 6, respectively. DJP co-coordinated the study and provided critical comments for the manuscript. All authors reviewed the results and approved the final version of the manuscript. REFERENCES 1. Valadkhan, S., and Jaladat, Y. (2010) The spliceosomal proteome: at the heart of the largest

cellular ribonucleoprotein machine. Proteomics 10, 4128-4141 2. Neuenkirchen, N., Chari, A., and Fischer, U. (2008) Deciphering the assembly pathway of

Sm-class U snRNPs. FEBS letters 582, 1997-2003 3. Otter, S., Grimmler, M., Neuenkirchen, N., Chari, A., Sickmann, A., and Fischer, U. (2007)

A comprehensive interaction map of the human survival of motor neuron (SMN) complex. The Journal of biological chemistry 282, 5825-5833

4. Pellizzoni, L., Yong, J., and Dreyfuss, G. (2002) Essential role for the SMN complex in the specificity of snRNP assembly. Science 298, 1775-1779

5. Chang, T. H., Tung, L., Yeh, F. L., Chen, J. H., and Chang, S. L. (2013) Functions of the DExD/H-box proteins in nuclear pre-mRNA splicing. Biochimica et biophysica acta 1829, 764-774

6. Suzuki, T., Izumi, H., and Ohno, M. (2010) Cajal body surveillance of U snRNA export complex assembly. The Journal of cell biology 190, 603-612

7. Yong, J., Golembe, T. J., Battle, D. J., Pellizzoni, L., and Dreyfuss, G. (2004) snRNAs contain specific SMN-binding domains that are essential for snRNP assembly. Molecular and cellular biology 24, 2747-2756

8. Morris, G. E. (2008) The Cajal body. Biochimica et biophysica acta 1783, 2108-2115 9. Wek, R. C., Jiang, H. Y., and Anthony, T. G. (2006) Coping with stress: eIF2 kinases and

translational control. Biochemical Society transactions 34, 7-11 10. Ait Ghezala, H., Jolles, B., Salhi, S., Castrillo, K., Carpentier, W., Cagnard, N., Bruhat, A.,

Fafournoux, P., and Jean-Jean, O. (2012) Translation termination efficiency modulates ATF4 response by regulating ATF4 mRNA translation at 5' short ORFs. Nucleic acids research 40, 9557-9570

11. Jiang, H. Y., Wek, S. A., McGrath, B. C., Lu, D., Hai, T., Harding, H. P., Wang, X., Ron, D., Cavener, D. R., and Wek, R. C. (2004) Activating transcription factor 3 is integral to the eukaryotic initiation factor 2 kinase stress response. Molecular and cellular biology 24, 1365-1377

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

! 10

12. Teske, B. F., Wek, S. A., Bunpo, P., Cundiff, J. K., McClintick, J. N., Anthony, T. G., and Wek, R. C. (2011) The eIF2 kinase PERK and the integrated stress response facilitate activation of ATF6 during endoplasmic reticulum stress. Molecular biology of the cell 22, 4390-4405

13. Donnelly, N., Gorman, A. M., Gupta, S., and Samali, A. (2013) The eIF2alpha kinases: their structures and functions. Cellular and molecular life sciences : CMLS 70, 3493-3511

14. Balachandran, S., and Barber, G. N. (2007) PKR in innate immunity, cancer, and viral oncolysis. Methods Mol Biol 383, 277-301

15. Abdel-Nour, M., Tsalikis, J., Kleinman, D., and Girardin, S. E. (2014) The emerging role of mTOR signalling in antibacterial immunity. Immunology and cell biology 92, 346-353

16. Laplante, M., and Sabatini, D. M. (2012) mTOR signaling in growth control and disease. Cell 149, 274-293

17. Tsalikis, J., Croitoru, D. O., Philpott, D. J., and Girardin, S. E. (2013) Nutrient sensing and metabolic stress pathways in innate immunity. Cellular microbiology 15, 1632-1641

18. Kumar, V., Sabatini, D., Pandey, P., Gingras, A. C., Majumder, P. K., Kumar, M., Yuan, Z. M., Carmichael, G., Weichselbaum, R., Sonenberg, N., Kufe, D., and Kharbanda, S. (2000) Regulation of the rapamycin and FKBP-target 1/mammalian target of rapamycin and cap-dependent initiation of translation by the c-Abl protein-tyrosine kinase. The Journal of biological chemistry 275, 10779-10787

19. Tattoli, I., Philpott, D. J., and Girardin, S. E. (2012) The bacterial and cellular determinants controlling the recruitment of mTOR to the Salmonella-containing vacuole. Biol Open 1, 1215-1225

20. Tattoli, I., Sorbara, M. T., Philpott, D. J., and Girardin, S. E. (2012) Bacterial autophagy: The trigger, the target and the timing. Autophagy 8

21. Tattoli, I., Sorbara, M. T., Vuckovic, D., Ling, A., Soares, F., Carneiro, L. A., Yang, C., Emili, A., Philpott, D. J., and Girardin, S. E. (2012) Amino acid starvation induced by invasive bacterial pathogens triggers an innate host defense program. Cell host & microbe 11, 563-575

22. Tattoli, I., Sorbara, M. T., Yang, C., Tooze, S. A., Philpott, D. J., and Girardin, S. E. (2013) Listeria phospholipases subvert host autophagic defenses by stalling pre-autophagosomal structures. The EMBO journal 32, 3066-3078

23. Shukla, S., and Parker, R. (2014) Quality control of assembly-defective U1 snRNAs by decapping and 5'-to-3' exonucleolytic digestion. Proceedings of the National Academy of Sciences of the United States of America 111, E3277-3286

24. Liu, J. L., and Gall, J. G. (2007) U bodies are cytoplasmic structures that contain uridine-rich small nuclear ribonucleoproteins and associate with P bodies. Proceedings of the National Academy of Sciences of the United States of America 104, 11655-11659

25. Carneiro, L. A., Travassos, L. H., Soares, F., Tattoli, I., Magalhaes, J. G., Bozza, M. T., Plotkowski, M. C., Sansonetti, P. J., Molkentin, J. D., Philpott, D. J., and Girardin, S. E. (2009) Shigella induces mitochondrial dysfunction and cell death in nonmyleoid cells. Cell host & microbe 5, 123-136

26. Sancak, Y., Peterson, T. R., Shaul, Y. D., Lindquist, R. A., Thoreen, C. C., Bar-Peled, L., and Sabatini, D. M. (2008) The Rag GTPases bind raptor and mediate amino acid signaling to mTORC1. Science 320, 1496-1501

27. Schmidt, E. K., Clavarino, G., Ceppi, M., and Pierre, P. (2009) SUnSET, a nonradioactive method to monitor protein synthesis. Nature methods 6, 275-277

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

! 11

28. Rutkowski, D. T., and Kaufman, R. J. (2003) All roads lead to ATF4. Developmental cell 4, 442-444

29. Takata, A., Otsuka, M., Yoshikawa, T., Kishikawa, T., Kudo, Y., Goto, T., Yoshida, H., and Koike, K. (2012) A miRNA machinery component DDX20 controls NF-kappaB via microRNA-140 function. Biochemical and biophysical research communications 420, 564-569

30. Buckingham, M., and Liu, J. L. (2011) U bodies respond to nutrient stress in Drosophila. Experimental cell research 317, 2835-2844

31. Hua, Y., and Zhou, J. (2004) Survival motor neuron protein facilitates assembly of stress granules. FEBS letters 572, 69-74

32. Hutvagner, G., and Zamore, P. D. (2002) A microRNA in a multiple-turnover RNAi enzyme complex. Science 297, 2056-2060

33. Mourelatos, Z., Dostie, J., Paushkin, S., Sharma, A., Charroux, B., Abel, L., Rappsilber, J., Mann, M., and Dreyfuss, G. (2002) miRNPs: a novel class of ribonucleoproteins containing numerous microRNAs. Genes & development 16, 720-728

34. Eulalio, A., Frohlich, K. S., Mano, M., Giacca, M., and Vogel, J. (2011) A candidate approach implicates the secreted Salmonella effector protein SpvB in P-body disassembly. PloS one 6, e17296

35. Kedersha, N., Tisdale, S., Hickman, T., and Anderson, P. (2008) Real-time and quantitative imaging of mammalian stress granules and processing bodies. Methods in enzymology 448, 521-552

36. Lorson, C. L., Hahnen, E., Androphy, E. J., and Wirth, B. (1999) A single nucleotide in the SMN gene regulates splicing and is responsible for spinal muscular atrophy. Proceedings of the National Academy of Sciences of the United States of America 96, 6307-6311

37. Kye, M. J., Niederst, E. D., Wertz, M. H., Goncalves, I. D., Akten, B., Dover, K. Z., Peters, M., Riessland, M., Neveu, P., Wirth, B., Kosik, K. S., Sardi, S. P., Monani, U. R., Passini, M. A., and Sahin, M. (2014) SMN regulates axonal local translation via miR-183/mTOR pathway. Human molecular genetics

38. Ning, K., Drepper, C., Valori, C. F., Ahsan, M., Wyles, M., Higginbottom, A., Herrmann, T., Shaw, P., Azzouz, M., and Sendtner, M. (2010) PTEN depletion rescues axonal growth defect and improves survival in SMN-deficient motor neurons. Human molecular genetics 19, 3159-3168

39. Custer, S. K., and Androphy, E. J. (2014) Autophagy dysregulation in cell culture and animals models of spinal muscular atrophy. Molecular and cellular neurosciences 61, 133-140

40. Garcera, A., Bahi, N., Periyakaruppiah, A., Arumugam, S., and Soler, R. M. (2013) Survival motor neuron protein reduction deregulates autophagy in spinal cord motoneurons in vitro. Cell death & disease 4, e686

41. Sahashi, K., Hua, Y., Ling, K. K., Hung, G., Rigo, F., Horev, G., Katsuno, M., Sobue, G., Ko, C. P., Bennett, C. F., and Krainer, A. R. (2012) TSUNAMI: an antisense method to phenocopy splicing-associated diseases in animals. Genes & development 26, 1874-1884

42. Moore, M. J., Wang, Q., Kennedy, C. J., and Silver, P. A. (2010) An alternative splicing network links cell-cycle control to apoptosis. Cell 142, 625-636

43. Boutz, P. L., Stoilov, P., Li, Q., Lin, C. H., Chawla, G., Ostrow, K., Shiue, L., Ares, M., Jr., and Black, D. L. (2007) A post-transcriptional regulatory switch in polypyrimidine tract-

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

! 12

binding proteins reprograms alternative splicing in developing neurons. Genes & development 21, 1636-1652

44. Calarco, J. A., Superina, S., O'Hanlon, D., Gabut, M., Raj, B., Pan, Q., Skalska, U., Clarke, L., Gelinas, D., van der Kooy, D., Zhen, M., Ciruna, B., and Blencowe, B. J. (2009) Regulation of vertebrate nervous system alternative splicing and development by an SR-related protein. Cell 138, 898-910

45. Wan, L., Ottinger, E., Cho, S., and Dreyfuss, G. (2008) Inactivation of the SMN complex by oxidative stress. Molecular cell 31, 244-254

46. Zhang, Z., Lotti, F., Dittmar, K., Younis, I., Wan, L., Kasim, M., and Dreyfuss, G. (2008) SMN deficiency causes tissue-specific perturbations in the repertoire of snRNAs and widespread defects in splicing. Cell 133, 585-600

FIGURE LEGENDS

Figure 1. Shigella infection affects U snRNA levels. (A) HeLa cells infected with Shigella for 4 hours, analyzed by immunofluorescence with antibodies against tri-methyl guanosine (TMG). (B) Lysates from cells infected with Shigella for 4 hours were immunoprecipitated with Sm antibody, followed by quantitative RT-PCR for the cellular levels of spliceosomal U snRNA (top). Results are shown as fold over the untreated condition n=3. (C) Protein levels of DDX20 and SMN in cytoplasmic, nuclear, and total lysates during Shigella (S.f) infection analyzed by western blot, as compared to p84 and tubulin loading controls. Figure 2. Shigella infection induces the formation of cytosolic U bodies. (A, B) Immunofluorescence analysis of HeLa cells infected with Shigella for 4 hours, stained with antibodies against DDX20 and SMN (A) or polyclonal DDX20 and Sm (B). (C) Immunofluorescence analysis of HeLa cells infected with Shigella for 4 hours vs uninfected, stained with tri-methyl guanosine (TMG) and polyclonal DDX20 antibodies. A magnified region is shown, for which contract was increased to reveal the TMG+ foci. (D) Quantification of % cells positive for U bodies during Shigella infection over 4 hours. Figure 3. Shigella-induced U bodies colocalize with P bodies at early stages of infection. (A, B) Immunofluorescence analysis of HeLa cells either left uninfected (A) or infected with Shigella for 1 hour (B), stained with an antibodies against DDX20 and DDX6. (C) Super-resolution microscopy analysis of HeLa cells infected with Shigella for 1 hour, stained with antibodies for DDX20 and DDX6 (left). Scale bar=5 µm. 3D-rendering showing two projections for four distinct foci, showing colocalization between U body and P body markers (green=DDX6, red=DDX20) (right). Scale bars=200nm. Figure 4. Prolonged Shigella infection results in disappearance of P bodies. (A-C) Immunofluorescence of HeLa cells infected with Shigella for 4 hours, stained with DDX6 and DDX20 (A), SMN (B), or Ge-1 (C). (D) Quantification of percent cells positive for U bodies (DDX20) or P bodies (DDX6) over a 4 hour Shigella infection. (E) Immunofluorescence microscopy analysis of HeLa cell treated with 10 uM CCCP for 5 hours, stained with DDX6. (F) Immunofluorescence miscropy analysis of HeLa cells infected with Shigella for 4 hours, stained with Tia-1 and DDX20. Figure 5. Induction of cytosolic U bodies upon infection with Salmonella and Listeria.

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

! 13

(A, B) Immunofluorescence analysis of HeLa cells infected with mCherry Salmonella WT (A) or GFP Listeria (B) for 2 hours, stained with DDX20 antibody. Note, mCherry Salmonella WT has been false colored green for consistency. (C, D) Quantification of U body formation in Salmonella (C) and Listeria (D) infected cells over 4 hours. Figure 6. Bacteria-induced U body formation is dependent on membrane damage. (A, B) Immunofluorescence of HeLa cells infected with either WT or ΔSPI-1/Inv Salmonella for two hours stained for DDX20, mTOR or NDP52. (C) Quantification of % cells positive for U bodies in WT vs. ΔSPI-1/Inv Salmonella-infected HeLa cells over 4 hours. (D) Immunofluorescence of HeLa cells infected with ΔSPI-1/Inv Salmonella for 2 hours, followed by a 15 min treatment of glycyl-L-phenylalanine-2-napthylamide (GPN), stained with NDP52 and DDX20. Figure 7. Metabolic stresses induce the formation of cytosolic U bodies. (A) Immunofluorescence analysis of HeLa cells treated with KRB buffer for 4 hours stained with DDX6 and DDX20 (super resolution microscopy). (B) Immunofluorescence analysis of HeLa cells treated with amino acid free KRB buffer or 10 uM thapsigargin for 4 hours stained with TMG. (C) Quantification of the percent of HeLa cells positive for U bodies when treated with various stimuli. ***p<0.001 and **p<0.01 over CTR. (D) Quantification of the percent of HeLa cells positive for U bodies upon treatment with KRB for 6 hours or KRB for 4 hours with 2 hours of recovery in DMEM complete medium. Figure 8. U body formation is dependent on translation initiation inhibition. (A) Key upstream pathways involved in eliciting translation initiation inhibition in response to stress. (B) SUnSEt assay to measure active translation in HeLa cells via incorporation of puromycin. Unstimulated cells without puromycin added, along with non-stimulated, cycloheximide treated, thapsigargin treated and Shigella infected cells treated with puromycin for 10 min, analyzed using anti-puromycin antibody via western blot. (C) SUnSEt assay on HeLa cells infected with Shigella for 4 hours, stained with anti-puromycin antibody. (D) Percent of HeLa cells positive for U bodies upon treatment with various translational inhibitors. (E) Immunofluorescence analysis of pateamine A (2 hour) treated HeLa cells, stained for Tia-1 and DDX20. Figure 9. U body formation is controlled by the ATF4/ATF3 signaling axis. (A, B) Transient lentiviral shRNA knockdown of ATF3 in HeLa cells. (C, D) Immunofluorescence visualization (C) and quantification via percent of scramble versus ATF3 knockdown cells positive for U bodies (D) upon treatment with 5 uM thapsigargin for 4 hours. **, p<0.01. Analysis of lentiviral shRNA knockdown of ATF4 versus scramble HeLa cells under control and 5 uM thapsigargin treatment via western blot using antibodies against ATF4 and the housekeeping control, tubulin (E) and by immunofluorescence using antibodies against DDX6 and DDX20 (F). Figure 10. Knockdown of the U body component DDX20 affects the ATF4/ATF3 pathway and host response to infection. (A) Transient lentiviral shRNA knockdown of DDX20 in HeLa cells. (B) ATF4 levels analyzed via western blot in scramble vs. DDX20-knockdown HeLa cells infected with Shigella (S.f) or treated with 5 µM thapsigargin for 4 hours. (C) Expression of ATF3 scramble vs. DDX20-knockdown HeLa cells upon stimulation with thapsigargin, KRB buffer or Shigella infection, analyzed via quantitative PCR. (D-F) Fold induction of Igκ or ISRE luciferase activity upon Shigella infection (D), MyD88 (E) and MAVS overexpression (F) in scramble vs. DDX20 knockdown cells.

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

0

0,2

0,4

0,6

0,8

1

1,2

1,4

U1 U2 U4 U5 U6 U11 U12 U4atac U6atac

Exp

ress

ion

rela

tive

to

unin

fect

ed

00,20,40,60,81

1,2

U1 U2 U4 U5 U6 U11 U12 U4atac U6atac

Exp

ress

ion

rel

ativ

e to

u

nin

fect

ed

BU snRNAs (Total)

U snRNAs (IP cytosolic Sm)

CTRShigella 4h

Fig. 1

*

*** *** *** **

***

* *

*** **

DAPI

TMGA

C

SMN

p84

Tubulin

DDX20

Totallysate

Cytosol Nucleus

C S.f C S.f C S.f

14

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

Fig. 2

Shigella

0 1 2 3 40

20

40

60

80

100

Time (h)

U b

odie

s (%

cel

ls)

D

CTRTMG RNApoly DDX20

DAPI Merge

Shigella 4h

A

C

B

TMG RNApoly DDX20

DAPI Merge

Contrast-enhanced

CTR

Shigella

4h

CTR

Shigella

4h

SMN DDX20

DAPI Merge

poly DDX20 SmB/B’

DAPI Merge

SMN DDX20

DAPI Merge

poly DDX20 SmB/B’

DAPI Merge

15

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

Shigella 1hDDX6

DAPI

DDX20

Merge

A CTRDDX6

DAPI

DDX20

Merge

C Shigella 1h

DDX6 /DDX20

1 1

2 2

3 3

4 4

B

Fig. 3

16

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

Shigella 4h

0 2 40

20

40

60

80

100

hours

%

cel

ls w

ith D

DX

6 or

DD

X20

gra

nule

s DDX6DDX20

DDX6 DDX6

CTR CCCP 10 µM 5hE

Shigella 4hCD

DDX6

DAPI

SMN

Merge

DDX6

DAPI

Ge-1

Merge

Fig. 4 A Shigella 4h

DDX6

DAPI

DDX20

Merge

B

F DAPI Tia-1

DDX20 Merge

17

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

Salmonella

0 1 2 3 40

20

40

60

80

Time (h)

U b

odie

s (%

cel

ls)

A

Salmonella-mCherry WT 2h

C Listeria

0 1 2 3 40

10

20

30

40

Time (h)

U b

odie

s (%

cel

ls)D

BDDX20

Listeria-GFP WT 2h

DDX20

Merge

DAPI

mCherry GFP

DAPI

Merge

DDX20

A

Fig. 5

18

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

Salmonella

0 1 2 3 40

25

50

75 WTDelta SPI-1/Inv

Time (h)

B

NDP52 DDX20

DAPI Merge

NDP52 DDX20

DAPI Merge

Salmonella WT 2h

Salmonella ΔSPI-1/Inv 2h

C

U b

odie

s (%

cel

ls)

**

**

**

***

A Salmonella WT 2h

D Salmonella ΔSPI-1/Inv 2h/GPN 15’

NDP52 DDX20

DAPI Merge

DDX20 mTOR

DAPI Merge

Fig. 6

19

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

CTR MDP iE-DAP TNF Rapa. AAS Thapsi.0

5

10

15

20

25

30

35

40

**

***

***

U b

odie

s (%

cel

ls)

AA starvation (AAS) 4h B

DDX6 DDX20

DAPI Merge

A

DAPI TMG Merge

CTR

Thapsi4h

AAS4h

C D

U b

odie

s (%

cel

ls)

Fig. 7

20

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

CTR AAS PatA CHX Puromycin0

10

20

30

40

U b

odie

s (%

cel

ls)

D

α�Tubulin

α-Puro

Puro 10’

CTR

Shigella

Thapsi.

AA

S

CHX

No

Puro

.BA

eIF2a-P 4E-BPs

Translation initiation

mTOR

AA starvation ER stress

Rapamycin

Merge DAPI α�PuromycinC

Translation elongation

CHX,Puromycin

EPateamine A 2h

Tia-1/DDX20

Shigella 4h

Fig. 8

21

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

ATF3

Tubulin

shATF3

shScr.

A

shATF3 shSCR shATF3 shSCR0.0

2.5

5.0

7.5

10.0

12.5

CTR Thapsigargin 4h

**U

bod

ies (

% c

ells

)

shATF3 shSCR shATF3 shSCR

0

10

20

30

40

50

60

CTR Thapsigargin 4h

Nuc

lear

gem

s (%

cel

ls)

DAPI/DDX20

D

B Thapsigargin 4hCTR

shScr

shATF3

C

ATF4

CTR Thap. CTR Thap.

Scr KD ATF4 KD

Tubulin

E F

DAPI/DDX6/DDX20

Thapsigargin 4hCTR

shScr

shATF4

Fig. 9

22

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

A

DDX20

DDX20

Scr DDX20

100 kDa

70 kDa

130 kDa

*

p84

shRNA

Tubulin

ATF4

sh SCR

C S. f Th. C S.f Th.

sh DDX20

B

C

CTR Shigella0

5

10

15

20

Igk

Luci

fera

se (f

old) ShScramble

shDDX20

CTR MyD88 OE0

50

100

150

Igk

Luci

fera

se (f

old) ShScramble

shDDX20

CTR MAVS OE0

10

20

30

ISRE

Luc

ifera

se (f

old) ShScramble

shDDX20

D E F

Fig. 10

23

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from

Philpott and Stephen E. GirardinJessica Tsalikis, Ivan Tattoli, Arthur Ling, Matthew T. Sorbara, David O. Croitoru, Dana J.

stress inductionIntracellular bacterial pathogens trigger the formation of U bodies through metabolic

published online July 1, 2015J. Biol. Chem. 

  10.1074/jbc.M115.659466Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

by guest on June 1, 2018http://w

ww

.jbc.org/D

ownloaded from