RNASEK Is Required for Internalization of Diverse Acid-Dependent Viruses

RNASEK is required for internalization of diverse acid-dependent viruses

Brent A. Hacketta, Ari Yasunagaa, Debasis Pandaa, Michael A. Tartella, Kaycie C. Hopkinsa, Scott E. Hensleyb, and Sara Cherrya,1

aDepartment of Microbiology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104; and bVaccine Center, Wistar Institute, Philadelphia, PA 19104

Edited by Robert A. Lamb, Northwestern University, Evanston, IL, and approved May 15, 2015 (received for review December 16, 2014) Viruses must gain entry into cells to establish infection. In general, The molecular mechanisms involved in these uptake mecha- viruses enter either at the plasma membrane or from intracellular nisms are complex and rely on key molecules and organelles that endosomal compartments. Viruses that use endosomal pathways are essential for cellular viability, as these uptake mechanisms are dependent on the cellular factors that control this process; bring nutrients and other metabolites into the cytosol for cellular however, these genes have proven to be essential for endogenous growth and survival. Indeed, these processes and proteins in- cargo uptake, and thus are of limited value for therapeutic inter- volved are highly conserved from yeast to humans (14, 15). vention. The identification of genes that are selectively required Depending on the virus entry requirements, some viruses fuse for viral uptake would make appealing drug targets, as their in- within early endosomal vesicles, whereas others traffic to more hibition would block an early step in the life cycle of diverse viruses. acidic compartments or macropinosomes for entry. Because At this time, we lack pan-antiviral therapeutics, in part because many viruses are dependent on these endosomal trafficking of our lack of knowledge of such cellular factors. RNAi screening pathways for entry, much effort has been made in identifying the has begun to reveal previously unknown genes that play roles in specific cellular genes required for viral entry (16). Therapeutics viral infection. We identified dRNASEK in two genome-wide RNAi targeting entry are appealing because it is the first step in the screens performed in Drosophila cells against West Nile and Rift infection cycle, and many viruses use common pathways; thus, Valley Fever viruses. Here we found that ribonuclease kappa inhibition may be broadly antiviral, rather than active against (RNASEK) is essential for the infection of human cells by divergent only a specific virus. Furthermore, many viruses have high mu- MICROBIOLOGY and unrelated positive- and negative-strand-enveloped viruses from tation rates and rapidly evolve resistance to therapeutics tar- the Flaviviridae, Togaviridae, Bunyaviridae, and Orthomyxoviridae geting virally encoded genes. Conversely, therapeutics against families that all enter cells from endosomal compartments. In con- host encoded targets would likely be more difficult for the virus trast, RNASEK was dispensable for viruses, including parainfluenza to evade. virus 5 and Coxsackie B virus, that enter at the plasma membrane. Recent advances in functional genomic technologies have fa- RNASEK is dispensable for attachment but is required for uptake cilitated the use of unbiased genome-wide RNAi screens to of these acid-dependent viruses. Furthermore, this requirement ap- identify cellular genes required for viral infection (17). Such pears specific, as general endocytic uptake of transferrin is unaf- approaches allow for the discovery of otherwise unknown genes fected in RNASEK-depleted cells. Therefore, RNASEK is a potential that play essential roles in infection. We recently performed such ’ host cell Achilles heel for viral infection. screens in insect cells against two disparate insect-borne human pathogens: the flavivirus West Nile virus (WNV) and the arbovirus | entry | endocytosis | clathrin-mediated endocytosis | bunyavirus RVFV (18, 19). These are both arthropod-borne macropinocytosis human pathogens for which there are no vaccines or therapeutics. Furthermore, these viruses are quite divergent: WNV is a iral pathogens are quite diverse in their replication strate- flavivirus that is a globally important cause of encephalitis (20), Vgies; however, all viruses must enter cells to initiate their replication cycles. The first step involves binding of virus parti- Significance cles to the cell surface. Such interactions can involve attachment factors, which have low affinity but concentrate viruses on the Many viruses, including those of global concern, are dependent surface of cells, and receptors that intricately interact with viral on internalization for their entry. We found that ribonuclease envelope glycoproteins, which, in addition to binding, promote kappa (RNASEK) is required for infection of every virus we other aspects of infection such as internalization. Although a tested that enters cells through an acid-dependent pathway, plethora of receptors and pathways can be used, most viruses including dengue, West Nile, Sindbis, Rift Valley Fever, and in- take advantage of the cellular endocytic machinery and pene- fluenza viruses. Mechanistically, we found that RNASEK has no trate from within the cytosol (reviewed in refs. 1–3). Clathrin- effect on virus binding to cells but, rather, is required for their mediated endocytosis, macropinocytosis, and caveolin-mediated uptake. RNASEK was required for diverse viruses that are de- endocytosis are the best-studied forms of uptake used by viruses. pendent on clathrin-mediated endocytosis for entry, but we Clathrin-mediated endocytosis is the most common mechanism found that RNASEK was dispensable for general endocytic used by small viruses, as clathrin-coated vesicles have a diameter – uptake. Therefore, RNASEK appears to play a unique role in of 60 200 nm and can be enlarged to fit even larger particles (4, viral uptake and may be a therapeutically viable target to 5). This pathway is constitutive on most cells, and some viruses inhibit major human viral pathogens. use preexisting clathrin-coated pits for entry (e.g., dengue virus), whereas others induce the formation of these structures (e.g., Author contributions: B.A.H. and S.C. designed research; B.A.H., A.Y., D.P., M.A.T., and influenza virus) (6, 7). Macropinocytosis is an actin-dependent K.C.H. performed research; S.E.H. contributed new reagents/analytic tools; B.A.H. and S.C. endocytic process for the nonselective uptake of nutrients in analyzed data; and B.A.H. and S.C. wrote the paper. response to receptor engagement. It is the predominant pathway The authors declare no conflict of interest. for many larger viruses, including vaccinia virus, but is also used This article is a PNAS Direct Submission. by others, including influenza virus under some conditions (8– 1To whom correspondence should be addressed. Email: [email protected]. 10). It also remains unclear which pathways are used by some This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. viruses, including Rift Valley Fever virus (RVFV) (11–13). 1073/pnas.1424098112/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1424098112 PNAS Early Edition | 1of6 Downloaded by guest on September 27, 2021 and RVFV is a bunyavirus that causes significant morbidity and RNASEK Promotes Flavivirus Infection in Human Cells. RNASEK is a mortality in livestock and humans in Africa (21). In our screens, highly conserved protein with 51% homology between Drosophila there were only three genes that promoted infection by both viruses: and humans (25). Therefore, we tested whether these flaviviruses dRAB5, dSTX7, and dRNASEK (CG40127). The functions of require hRNASEK for infection in human cells. We obtained two RAB5 and STX7 have been described, but little is known about independent siRNAs against hRNASEK and found that they ef- ribonuclease kappa (RNASEK). Both RAB5 and STX7 are in- ficiently deplete hRNASEK in human osteosarcoma cells (U2OS), volved in endosomal transport and have roles in viral entry (22, 23). as measured by RT-qPCR (Fig. 1D). We challenged these cells Both WNV and RVFV are enveloped RNA viruses that require an with WNV (KUN) or DENV2 (NGC) and monitored infection by acidic compartment for entry (12, 13, 24). RNASEK is a single- RT-qPCR or automated microscopy. We found that depletion of copy, 137-aa protein conserved from insects to humans (25–27) hRNASEK significantly attenuates infection of these flaviviruses in with an unknown function. We set out to determine the role of human cells (Fig. 1 E and F and Fig. S2 A and B). Again, we ob- RNASEK in viral infection and found that RNASEK is required served no change in cell number or growth properties of hRNA- for internalization of a diverse panel of viruses of medical concern. SEK-depleted cells (Fig. S2C). We also tested a different human cell line, HEK293T cells, and found that depletion of hRNASEK Results is efficient in these cells and attenuates WNV (KUN) infection RNASEK Promotes Flaviviral Infection in Insect Cells. Because we had (Fig. S3A). Our initial studies monitored infection of WNV at a identified dRNASEK as promoting infection in our genome- late point postinfection, so we next tested whether depletion of wide RNAi screens, we set out to verify this outside of a hRNASEK affects infection at early points. We depleted screening format with independent RNAi reagents. First, we hRNASEK in U2OS cells and monitored infection by WNV verified that we efficiently depleted dRNASEK, as measured by (KUN) 8 hours post infection (hpi), and observed significantly decreased infection at this early time (Fig. S3B). RT-quantitative PCR (qPCR), compared with control knock- down (Fig. 1A). We challenged these cells with WNV [Kunjin RNASEK Is Required for Infection of Diverse pH-Dependent Viruses. strain (KUN)], as it is closely related to WNV strains circulating Our initial RNAi screens identified dRNASEK as required for globally (but can be used under biosafety level 2 containment), both WNV (NY2000) and RVFV (MP12) infection. Therefore, and monitored infection by RT-qPCR and automated micros- we tested the role of hRNASEK in RVFV infection of human copy. We found that depletion of dRNASEK significantly at- cells. Indeed, we found that depletion of hRNASEK significantly tenuated infection of WNV in Drosophila cells (Fig. 1B and Fig. reduces infection by RVFV (MP12), as measured by RT-qPCR S1 A and B). We extended this study to dengue virus because it is and immunoblot (Fig. 2A and Fig. S4A). Because humans are a pandemic threat (two-fifths of the world’s population is at risk, infected by three families of arthropod-borne viruses, the flavi- and it is the fastest-spreading tropical disease and arthropod- viruses, the bunyaviruses, and the alphaviruses, and we found borne viral pathogen worldwide). We challenged control or that both flaviviruses and bunyaviruses are sensitive to RNA- RNASEK-depleted cells with dengue virus serotype 2 (DENV2) SEK, we tested the role of a hRNASEK in alphavirus infection. [S2 or New Guinea C (NGC)] and monitored infection by RT- We used the prototypical alphavirus, Sindbis virus (SINV, HRsp qPCR or automated microscopy and observed that loss of strain), and found that depletion of hRNASEK attenuates in- dRNASEK significantly attenuates infection (Fig. 1C and Fig. S1 A fection, as measured by RT-qPCR and immunoblot (Fig. 2B and and B). We observed no change in cell number or growth prop- Fig. S4B). Aside from being arthropod-borne, these viruses all erties of dRNASEK-depleted cells (Fig. S1C). require endocytic uptake and acidification for entry (28–30).

A 1.2 B 1.2 C 1.2 dRNASEK WNV (KUN) DENV2 (S2) 1 1 1

0.8 0.8 0.8

0.6 0.6 0.6 * 0.4 0.4 * 0.4 Relative Fold Change Fold Relative Relative Fold Change Fold Relative Relative Fold Change Fold Relative 0.2 0.2 0.2 * *** 0 0 0 Control dRNASEK Control dRNASEK WNV Control dRNASEK D RNAi E RNAi F RNAi

1.2 hRNASEK 1.2 WNV (KUN) 1.2 DENV2 (NGC)

1 1 1

0.8 0.8 0.8

0.6 0.6 0.6 *** 0.4 0.4 0.4 Relative fold change fold Relative

* change fold Relative Relative Fold Change Change Fold Relative 0.2 *** 0.2 0.2 *** *** *** 0 0 0 Control RNASEK(1) RNASEK(2) Control RNASEK(1)RNASEK(2) Control RNASEK(1) RNASEK(2) siRNA siRNA siRNA

Fig. 1. RNASEK promotes flavivirus infection in both Drosophila and human cells. Drosophila DL1 cells were treated with the indicated dsRNAs for 3 d and subsequently infected with WNV (KUN) (MOI 1, 48 h) or DENV2 (S2) (MOI 0.5, 72 h), and RNA was subject to RT-qPCR for dRNASEK (A), WNV (B), or DENV2 (C). Mean ± SEM; n = 3; *P < 0.05, ***P < 0.001. Human U2OS cells were transfected with the indicated siRNAs for 3 d and subsequently infected with WNV (KUN) (MOI 0.5, 24 h) or DENV2 (NGC) (MOI 0.5, 40 h), and RNA was subject to RT-qPCR for hRNASEK (D), WNV (E), or DENV2 (F). Mean ± SEM; n = 3; *P < 0.05, ***P < 0.001.

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1424098112 Hackett et al. Downloaded by guest on September 27, 2021 1.2 1.2 A RVFV (MP12) B SINV (HRsp) 1 1

0.8 0.8

0.6 0.6

0.4 0.4 * Relative fold change fold change Relative Relative fold change fold change Relative * * 0.2 0.2 * 0 0 siControl siRNASEK1 siRNASEK2 siControl siRNASEK1 siRNASEK2 1.2 1.2 C 1.2 IAV (CA/09) D IAV (Bris/10) E IAV (PR8) 1 1 1

0.8 0.8 0.8

0.6 0.6 0.6

0.4 0.4 0.4 * ** Relative foldchange Relative fold changeRelative Relative fold change fold change Relative 0.2 ** 0.2 ** 0.2 ** ** 0 0 0 Control RNASEK(1) RNASEK(2) Control RNASEK(1) RNASEK(2) siControl siRNASEK1 siRNASEK2 F 1.4 G H PIV5 1.2 VACV 1.2 CVB 1.2 1 1 1 0.8 0.8 0.8 0.6 0.6 0.6

0.4 0.4 0.4 Relative fold change fold change Relative Relative fold change fold change Relative 0.2 0.2 change Fold Relative 0.2

0 0 0 siControl siRNASEK1 siRNASEK2 siControl siRNASEK1 siRNASEK2 siCon siRNASEK-1 siRNASEK-2 MICROBIOLOGY Fig. 2. RNASEK promotes infection of a broad spectrum of RNA viruses in human cells. Human U2OS cells were transfected with either the indicated siRNAs for 3 d and subsequently infected with (A) RVFV (MP12) (MOI 1, 24 h), (B) SINV (HRsp) (MOI 0.5, 24 h), (C) IAV (CA/09) (MOI 0.05, 24 h), (D) IAV (Bris/10) (MOI 0.05, 24 h), (E) IAV (PR8) (MOI 0.05, 2 4 h), (F) PIV5 (MOI 0.1, 24 h), or (G) CVB (MOI 3, 8 h) and processed for RT-qPCR analysis of viral RNA. Mean ± SEM; n = 3; *P < 0.05, **P < 0.005. (H) Human U2OS cells were transfected with either the indicated siRNAs for 3 d and subsequently infected with VACV (MOI 0.5, 8 h) and processed for automated microscopy with the percentage of infected cells quantified. Mean ± SEM; n = 3.

Therefore, we tested whether RNASEK was required for in- to test whether RNASEK is required for steps in the entry fection of influenza virus, an unrelated pH-dependent virus that pathway downstream of cellular attachment. For these studies, is not transmitted by insects. Influenza viruses are diverse, causing we bound virus [WNV (KUN), SINV (HRsp), or IAV (CA/09)] 3–5 million cases of severe illness and about 250,000–500,000 deaths in the cold to siRNA-transfected cells for 1 h. Next, we incubated yearly, and even more during pandemics (31, 32). We tested two the cells in either neutral or acidic PBS for 10 min to allow antigenically distinct influenza A viruses (IAV), H1N1 (A/Puerto plasma membrane fusion at the low pH. Cells were washed, Rico/8/1934 (PR8) and A/California/7/2009 (CA/09), and an H3N2 ammonium chloride was added to block secondary infections, strain (A/Brisbane/10/2007 (Bris/10). Again, depletion of hRNASEK and infection was monitored 24 hpi by automated microscopy. significantly attenuated infection, as measured by RT-qPCR and We found that depletion of hRNASEK at neutral pH leads to automated microscopy (Fig. 2 C–E and Fig. S4C). decreased infection of all three viruses (Fig. 3 A–C, Left). In All these viruses are enveloped and use endocytic mechanisms contrast, under acidic conditions that bypass endocytic requirem- to access low-pH compartments for entry (22). To determine ents, depletion of hRNASEK has no effect on infection (Fig. 3 whether the commonality between these viruses was their de- A–C, Right). These data suggest that hRNASEK is required pendence on intracellular modes of entry, we tested two viruses downstream of cellular attachment during entry. These data that penetrate at the plasma membrane: the paramyxovirus also demonstrate that hRNASEK is not required at a step down- parainfluenza virus 5 (PIV5) and the picornavirus Coxsackie B stream of entry, as infection was normal once the viruses fused at virus (CVB) (33, 34). We found that depletion of hRNASEK had the plasma membrane. no effect on the infection of either PIV5 or CVB measured by RT- qPCR (Fig. 2 F and G). We also tested vaccinia virus (VACV), as Viral Uptake Is Dependent on RNASEK. Our acid bypass assay sug- it is dependent on macropinocytosis for entry (35, 36). We found gests that RNASEK is required either for viral uptake or for the that RNASEK was dispensable for infection by VACV (Fig. 2H). acidification of endosomal compartments. Therefore, we set out Therefore, our data suggest that hRNASEK is required for virus to determine whether virus internalization was dependent on infection by viruses that undergo endocytic entry and fuse within RNASEK. For these studies, we used a RT-qPCR-based assay to internal compartments. monitor virus binding and internalization of WNV (KUN), SINV (HRsp), IAV (CA/09), and RVFV (MP12). Viruses were bound Requirements for RNASEK Can Be Bypassed by Fusion at the at 4 °C and then extensively washed to remove unbound virus. Membrane. If the requirement for hRNASEK was during in- Next, we either left the cells untreated or trypsinized the cells to tracellular entry, then we reasoned that we may be able to bypass remove bound virus and verified that we successfully removed the requirement for RNASEK by inducing virus fusion at the bound virus (Fig. S5). We quantified the cell-bound virus by plasma membrane. Many viruses, including WNV, SINV, and subtracting the background signal from the total binding quan- IAV, can enter cells at the membrane if the pH in the environ- tified by RT-qPCR and found that RNASEK-depleted cells had ment is acidified, allowing envelope triggering and cell fusion no statistically significant effect on binding, (Fig. 3 D–G), although (37–40). This has been termed an acid bypass assay, allowing us there was a trend toward increased binding. Next, we monitored

Hackett et al. PNAS Early Edition | 3of6 Downloaded by guest on September 27, 2021 2.5 n.s. A WNV (KUN) D E F G 2 WNV (KUN) SINV (HRsp) IAV (CA/09) RVFV (MP12) 4 6 1.5 * 8 45 3.5 7 5 1 40 3 6 35 Relative Infection 0.5 4 2.5 5 30 Virus Virus Virus 25 Virus 3 0 4 bound 2 bound bound bound siControl siRNASEK siControl siRNASEK (0h) (0h) 20 (0h) 1.5 3 (0h) 2 pH 7.4 pH 5.5 15

n.s. 1 2 fold changeRelative Relative fold change Relative 10

3 fold change Relative Relative fold changeRelative 1 B SINV (HRsp) 0.5 1 5 2.5 0 0 0 0 2 1.5 ** 1 1.2 1.2 Relative Infection Relative H 1.2 I J K 1.2 0.5 1 1 1 0 1 siControl siRNASEK siControl siRNASEK Fraction Fraction 0.8 0.8 0.8 internalized 0.8 pH 7.4 pH 5.5 internalized (2h) Fraction Fraction 1.6 0.6 (3hpi) 0.6 * 0.6 internalized 0.6 internalized C n.s. (3hpi) 1.4 IAV (CA/09) (2hpi) 0.4 0.4 0.4 1.2 * 0.4 * Relative fold changeRelative Relative fold change fold change Relative Relative fold changeRelative

1 fold Relative change 0.2 0.2 0.2 0.2 0.8 *** *** 0.6 0 0 0 0 0.4 Relative InfectionRelative 0.2 0 siControl siRNASEK siControl siRNASEK pH 7.4 pH 5.5

Fig. 3. RNASEK is required for virus internalization. (A–C) U2OS cells were transfected with indicated siRNA for 72 h and infected with (A) WNV (KUN) (MOI 5), (B) SINV (HRsp) (MOI 50), or (C) IAV (CA/09) (MOI 5) for 1 h at 4 °C. Cells were then treated for 10 min with PBS (pH 5.5 or 7.4), washed, and incubated at 37 °C for 24 h in the presence of ammonium chloride to block spread (added 2 hpi) and processed for automated microscopy. The normalized percentage of infected cells per well for four wells per experiment for three independent experiments is shown. Mean ± SEM; *P < 0.05, **P < 0.01. (D–K) U2OS cells were transfected with the indicated siRNAs for 3 d and infected with (D and H) WNV (KUN) (MOI 5), (E and I) SINV (HRsp) (MOI 5), or (F and J) IAV (CA/09) (MOI 5), or RVFV (MP12) (MOI 5) at 4 °C. Samples were untreated or trypsinized and processed for RT-qPCR at 0 h. The amount of bound virus (total minus signal remaining posttrypsinization) normalized to control is shown (D–G). Alternatively, the cells were shifted to 37 °C for 2–3 h as indicated, trypsinized to remove external virus, and processed for RT-qPCR, and the amount internalized normalized to control is shown (H–K). Mean ± SEM; n = 3; *P < 0.05, **P < 0.01, ***P < 0.001.

the amount of virus uptake by binding the viruses at 4 °C and RNASEK Is Dispensable for General Uptake. Because the uptake of washing unbound virus as earlier, but in this case, we subsequently WNV, SINV, and IAV at the plasma membrane is clathrin- shifted the cells to 37 °C for 2–3 h to allow internalization of dependent (2), we assayed whether clathrin-mediated endocy- bound virions. After internalization, cells were treated with trypsin tosis is generally disrupted by RNASEK depletion. If RNASEK to remove viral particles bound to the outside of cells. Here we was required for clathrin-mediated endocytosis, it would not be found that cells lacking hRNASEK are significantly attenuated in an ideal drug target because clathrin-mediated endocytosis is their ability to internalize all four viruses (Fig. 3 H–K). required for essential cellular processes. A classical assay for These data suggest that RNASEK is required for the initial clathrin-dependent endocytosis entails monitoring of transferrin stages of internalization from the plasma membrane for a diverse uptake (41). Transferrin binds to the transferrin receptor, which set of viruses that use different receptors for entry. To further verify requires clathrin-dependent endocytosis for internalization. this hypothesis, we performed a microscopy-based virus uptake Therefore, we assayed fluorescent transferrin uptake in control assay with WNV (KUN) and IAV (Bris/10) because we have an- and RNASEK-depleted cells. We treated cells with Alexa594- tibodies that efficiently recognize the viral glycoproteins for these Transferrin for 15 min and quenched extracellular transferrin, viruses. We incubated control or RNASEK-depleted cells with using an acid wash (42). We observed no difference in uptake virions at 4 °C, washed extensively, and fixed the cells (0 hpi) or between control and RNASEK-depleted cells (Fig. 4 B and C). shifted the cells to 37 °C for 3 h to allow internalization (3 hpi) and As a positive control, we treated cells with the dynamin inhibitor fixed the cells. Cells were then processed for microscopy in the dynasore, which completely blocks transferrin uptake (Fig. S7). absence of permeabilization to exclusively visualize extracellular Therefore, all clathrin-dependent endocytosis is not dependent virions, using antibodies against the viral glycoproteins. We found on RNASEK. Macropinocytosis is another major mechanism that there were similar levels of virus binding at 0 hpi comparing for viral uptake. VACV entry is dependent on this process, control with RNASEK-depleted cells for either WNV or IAV and in some situations, RVFV and IAV can enter cells via (Fig. 4A). At 3 hpi, we observed the loss of external virus staining macropinocytosis (8). Because VACV infection is RNASEK- in control cells, whereas extracellular virions were retained in independent, we expected that RNASEK is not required for RNASEK-depleted cells for both WNV and IAV infections (Fig. macropinocytosis, but nevertheless, we tested whether the up- 4A). This was quantified in Fig. S6. Furthermore, to verify that the take of dextran was affected by the loss of RNASEK. We treated loss in signal observed in the control siRNA-treated cells at 3 hpi control or RNASEK-depleted cells with fluorescently labeled was a result of internalization, we compared unpermeabilized with dextran for 30 min and observed no defect in the uptake in permeabilized cells and found that we readily detected viral an- RNASEK-deficient cells, whereas treatment with the macropino- tigens inside the cells on permeabilization (Fig. S6). Therefore, in cytosis inhibitor 5-(N-ethyl-N-isopropyl) amiloride (EIPA) blocked this microscopy-based assay, we observed that RNASEK has no uptake (Fig. S8). These data suggest that RNASEK is not required effect on viral attachment, but is required for virus uptake. for all clathrin-dependent endocytosis or macropinocytosis.

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1424098112 Hackett et al. Downloaded by guest on September 27, 2021 A 0HPI 3HPI internalization of viruses, and not canonical cargo. Such charac- siControl siRNASEK siControl siRNASEK teristics differentiate RNASEK from other recently identified factors, such as FUZ and TER94, which have been shown to be required for the uptake of both alphaviruses and cargo (38, 43). RNASEK is a ubiquitous protein that arose in metazoans. Most canonical endocytosis genes are conserved in yeast, further WNV (KUN) suggesting that RNASEK has more specialized functions. RNASEK is thought to encode a protein with two putative transmembrane domains that may anchor the protein at the plasma membrane for its function. How RNASEK promotes the internalization of virions, but not other cargo, such as transferrin, is IAV (Bris/10) IAV unclear, but there are a number of possibilities. First, it may be 100um required for uptake of large cargo. The typical biomolecule in- B 0 min 15 min C ternalized by clathrin-mediated endocytosis, including transferrin, 16 is small, whereas viruses are much larger. Second, it may be re- 14 0min 12 15min quired for signaling events that promote internalization. After at-

siControl tachment and receptor engagement, many viruses induce signaling 8 cascades that promote entry (3, 44). This is most clearly established 6 for viruses that engage receptors that induce macropinocytosis for 4 Transferrin puncta/cell uptake, including IAV (3, 8, 9, 44–46). Informatic analysis suggests 2 that RNASEK has an SH2 domain that may be interacting with 0 siRNASEK siRNASEK 50um si Control si RNASEK signaling events engaged by virus binding at the membrane. How- ever, many endogenous cargoes also induce signaling events in- Fig. 4. Viral uptake, but not transferrin uptake, is disrupted in RNASEK- cluding transferrin binding to its receptor (47, 48). depleted cells. (A) siRNA-transfected U2OS cells were incubated with WNV (KUN) (MOI 15) or IAV (Bris/10) (MOI 15) for 1 h at 4 °C, washed, and either Importantly, we found that RNASEK is not required for clathrin- immediately processed or incubated for 3 h at 37 °C for immunofluorescence mediated endocytosis or macropinocytosis of transferrin or dextran, under nonpermeabilized conditions. (B) siRNA-transfected U2OS cells were respectively. Therefore, targeted inhibition of RNASEK is not incubated with 50 μg/mL fluorescently labeled transferrin in the cold for 1 h expected to interfere with these normal cellular functions. Thus, MICROBIOLOGY and shifted to 37 °C for 15 min before an acid wash to quench extracellular RNASEK may present a previously unknown target for pan-antiviral transferrin and subsequent processing for confocal microscopy. (C) Auto- therapeutic interventions because major viral pathogens such as in- mated image analysis quantified the number of puncate per cell and was fluenza and dengue virus are dependent on RNASEK for infection. normalized to control at 0 min. Mean ± SEM; n = 3. Materials and Methods Discussion Cell Lines, Viruses, Antibodies, and Reagents. Human cells and Drosophila DL1 cells were grown and maintained as previously described (49). The WNV- By mining our RNAi screening data (18, 19), we found that KUNV isolate 550 (CH16532) was a gift of R. Tesh and propagated as pre- RNASEK was potentially required for infection of WNV and viously described (50). DENV2-S2 was provided by M. Garcia-Blanco and RVFV in insect cells. Given that RNASEK is conserved but propagated as described (51). GFP-expressing Sindbis virus, a gift from R. poorly characterized, we set out to dissect the role of RNASEK Hardy, University of Indiana, was propagated as detailed previously. Cox- in viral infection. We found that RNASEK promotes viral up- sackie B virus was provided by J. Bergelson, Children’s Hospital of Pennsyl- take of a diverse panel of viruses including flaviviruses, alpha- vania (CHOP), and influenza stocks [H1N1 (California/7/2009), H1N1 (PR8), viruses, bunyaviruses, and orthomyxoviruses. These viruses all and H3N2 (Brisbane/10/2007)] were grown at the Wistar Institute. DENV2 (New Guinea C) and PIV5 stocks were purchased (ATCC) and propagated in enter cells using endocytic routes and are dependent on acidi- BHK (DENV2) or Vero (PIV5) cells. VACV-YFP was a gift of S. Isaacs, University fication for entry. We tested whether viruses that fuse at the of Pennsylvania. Monoclonal antibodies against flavivirus NS1 and E proteins membrane, including a paramyxovirus and a picornavirus, were (4G2 and 9NS1, respectively) were provided by M. Diamond (Washington

dependent on RNASEK, and found that they were not. We also University). RVFV GN monoclonal antibody (4D4) was provided by R. Doms found that VACV, which enters by macropinocytosis, was not (CHOP). Influenza A antibody to NP was provided by BEI Resources (NR-4282). dependent on RNASEK. By performing a series of assays to Other antibodies were obtained as follows: GFP (Santa Cruz Biotechnology narrow down the requirement, we found that RNASEK promotes [SCBT]), beta-actin (SCBT), and tubulin (Sigma). Fluorescent secondary an- the earliest steps of virus uptake but has no effect on virus binding tibodies and transferrin were obtained from Invitrogen, and Horse radish to cells. RNASEK was dispensable once viruses delivered their peroxidase-conjugated secondary antibodies were from Amersham. Other payload to the cytoplasm, suggesting it has no additional roles in reagents were from Sigma. viral infection downstream of entry. RNAi. Drosophila cells were treated with dsRNA for 3 d, as described (52). The panel of viruses sensitive to RNASEK uses disparate re- U2OS cells or 293T cells were transfected with 30 nM either negative control ceptors and endocytic pathways for entry. WNV, DENV, SINV, siRNA (Ambion) or one of two independent siRNAs targeting RNASEK and IAV are thought to be largely dependent on clathrin-mediated (Ambion s54102, s45103), both of which target the major isoform of RNASEK, endocytosis (2). RVFV and IAV are thought to use macro- but only s54102 targets the poorly conserved isoform 2, using HiPerfect pinocytosis at least under some conditions; however, VACV, which according to the manufacturers protocol (Qiagen) and infected at 72 hours is dependent on macropinocytosis, was RNASEK-independent (7, post transfection (hpt), as indicated. 8, 12). Therefore, our data suggest that RNASEK selectively affects clathrin-mediated endocytosis of viruses. Thus, we tested RT-qPCR Assays. Total RNA was extracted using TRIzol reagent (Invitrogen). whether RNASEK is required for the uptake of other cargo de- cDNA was generated using random hexamers to prime reverse transcription pendent on clathrin-mediated endocytosis or macropinocytosis. As reactions using M-MLV reverse transcriptase. qPCR was performed with the cDNA, using Power SYBR Green PCR Master Mix and a StepOne Plus RT-PCR expected, we found that RNASEK was dispensable for the mac- system (Applied Biosystems). Reactions were initially run at initial 95 °C for ropinocytic uptake of dextran. However, surprisingly, we found that 5 min and then 40 cycles of 95 °C for 20 s, 52 °C for 30 s, and 72 °C for 30 s RNASEK was dispensable for the uptake of the classic cargo Analysis by ΔΔCT method. Relative copy numbers were generated by nor- transferrin, which is dependent on clathrin-mediated endocytosis. malizing to cells to rp49 for Drosophila cells or 28S RNA for human cells and Therefore, RNASEK is seemingly uniquely required for the compared with control siRNA. Primers are described in Table S1.

Hackett et al. PNAS Early Edition | 5of6 Downloaded by guest on September 27, 2021 Acid Bypass Assay. U2OS cells were transfected with siRNAs and infected with point, fixed in PBS/4% (vol/vol) formaldehyde (15 min, room temperature), SINV (HRsp) (multiplicity of infection [MOI] 5), WNV (KUN) (MOI 5), or IAV and blocked in PBS/2% (wt/vol) BSA and processed for confocal microscopy (CA/09) (MOI 5) in the cold for 1 h to allow binding. Cells were then washed and using anti-E for WNV or anti-HA for IAV (53). incubated for 10 min in either PBS at pH 5.5 or PBS at pH 7.2. Culture media was then replaced and cells were incubated at 37 °C with 20 mM NH4Cl and pro- Transferrin Uptake Assay. siRNA-transfected U2OS cells were treated with cessed for automated microscopy at 24 hpi. Cells were fixed with PBS/4% 50 μg/mL Alexa594-labeled transferrin for 1h at 4 °C and then incubated at (vol/vol) formaldehyde for 15 min and washed 3× with PBS/1% Triton X-100. 37 °C for 15 min. After washing, cells were placed in acid strip buffer, as de- Cells were processed and imaged, using an automated microscope (ImageXpress scribed, to monitor intracellular transferrin (42). Automated image analysis Micro) and automated analysis, as previously described (53). Four wells per was used to quantify the number of punctae per cell of 10 fields per con- condition with four sites per well were collected and quantified (MetaXpress). dition in three independent experiments. Viral Entry Assays. siRNA-transfected cells were incubated with WNV (KUN, MOI 5 for qPCR, MOI 15 for microscopy), SIN (HRsp, MOI 5), IAV (H1N1, MOI 5 ACKNOWLEDGMENTS. We thank members of the S.C. laboratory for helpful discussions and technical advice throughout. We thank BEI. This work for qPCR assay, MOI 15 for microscopy), or RVFV (MP12, MOI 5) in the cold (t = 0) – = was supported by National Institutes of Health Grants R01AI074951 or incubated at 37 °C for 2 3h(t 2 or 3 h). WNV (KUN) and RVFV (MP12) (to S.C.), U54AI057168 (to S.C.), R21AI103441 (to S.C.), R01AI095500 were spinoculated onto the cells for the qPCR assay. For the qPCR assay, cells (to S.C.), R01AI113047 (to S.E.H.), and R01AI108686 (to S.E.H.). S.C. is a were also left untreated or treated with trypsin for 3 min, as indicated. For recipient of the Burroughs Wellcome Investigators in the Pathogenesis of the microscopy assay, cells were washed twice with PBS at the end of each Infectious Disease Award.

1. Mercer J, Helenius A (2009) Virus entry by macropinocytosis. Nat Cell Biol 11(5): 28. Smit JM, Moesker B, Rodenhuis-Zybert I, Wilschut J (2011) Flavivirus cell entry and 510–520. membrane fusion. Viruses 3(2):160–171. 2. Mercer J, Schelhaas M, Helenius A (2010) Virus entry by endocytosis. Annu Rev Bio- 29. Sánchez-San Martín C, Liu CY, Kielian M (2009) Dealing with low pH: Entry and exit of chem 79:803–833. alphaviruses and flaviviruses. Trends Microbiol 17(11):514–521. 3. Yamauchi Y, Helenius A (2013) Virus entry at a glance. J Cell Sci 126(Pt 6):1289–1295. 30. Guu TS, Zheng W, Tao YJ (2012) Bunyavirus: Structure and replication. Adv Exp Med 4. Cureton DK, Massol RH, Saffarian S, Kirchhausen TL, Whelan SP (2009) Vesicular Biol 726:245–266. stomatitis virus enters cells through vesicles incompletely coated with clathrin that 31. Anonymous; Centers for Disease Control and Prevention (CDC) (2010) Estimates of depend upon actin for internalization. PLoS Pathog 5(4):e1000394. deaths associated with seasonal influenza — United States, 1976-2007. MMWR Morb 5. Kirchhausen T (2000) Clathrin. Annu Rev Biochem 69:699–727. Mortal Wkly Rep 59(33):1057–1062. 6. Rust MJ, Lakadamyali M, Zhang F, Zhuang X (2004) Assembly of endocytic machinery 32. World Health Organization (2003) Influenza. Available at www.who.int/mediacentre/ around individual influenza viruses during viral entry. Nat Struct Mol Biol 11(6): factsheets/2003/fs211/en/. Accessed November 3, 2014. 567–573. 33. Bergelson JM, Coyne CB (2013) Picornavirus entry. Adv Exp Med Biol 790:24–41. 7. van der Schaar HM, et al. (2008) Dissecting the cell entry pathway of dengue virus by 34. Chang A, Dutch RE (2012) Paramyxovirus fusion and entry: Multiple paths to a single-particle tracking in living cells. PLoS Pathog 4(12):e1000244. common end. Viruses 4(4):613–636. 8. Mercer J, Helenius A (2012) Gulping rather than sipping: Macropinocytosis as a way of 35. Mercer J, et al. (2010) Vaccinia virus strains use distinct forms of macropinocytosis for – virus entry. Curr Opin Microbiol 15(4):490 499. host-cell entry. Proc Natl Acad Sci USA 107(20):9346–9351. 9. de Vries E, et al. (2011) Dissection of the influenza A virus endocytic routes reveals 36. Sandgren KJ, et al. (2010) A differential role for macropinocytosis in mediating entry macropinocytosis as an alternative entry pathway. PLoS Pathog 7(3):e1001329. of the two forms of vaccinia virus into dendritic cells. PLoS Pathog 6(4):e1000866. 10. Edinger TO, Pohl MO, Stertz S (2014) Entry of influenza A virus: Host factors and 37. Liao M, Kielian M (2005) Domain III from class II fusion proteins functions as a dom- – antiviral targets. J Gen Virol 95(Pt 2):263 277. inant-negative inhibitor of virus membrane fusion. J Cell Biol 171(1):111–120. 11. de Boer SM, et al. (2012) Acid-activated structural reorganization of the Rift Valley 38. Panda D, et al. (2013) Genome-wide RNAi screen identifies SEC61A and VCP as con- – fever virus Gc fusion protein. J Virol 86(24):13642 13652. served regulators of Sindbis virus entry. Cell Reports 5(6):1737–1748. 12. Filone CM, et al. (2010) Rift valley fever virus infection of human cells and insect hosts 39. Stauffer S, et al. (2014) Stepwise priming by acidic pH and a high K+ concentration is is promoted by protein kinase C epsilon. PLoS ONE 5(11):e15483. required for efficient uncoating of influenza A virus cores after penetration. J Virol 13. Harmon B, et al. (2012) Rift Valley fever virus strain MP-12 enters mammalian host 88(22):13029–13046. cells via caveola-mediated endocytosis. J Virol 86(23):12954–12970. 40. Thompson BS, et al. (2009) A therapeutic antibody against west nile virus neutralizes 14. Merrifield CJ, Kaksonen M (2014) Endocytic accessory factors and regulation of cla- infection by blocking fusion within endosomes. PLoS Pathog 5(5):e1000453. thrin-mediated endocytosis. Cold Spring Harb Perspect Biol 3;6(11):a016733. 41. Mayle KM, Le AM, Kamei DT (2012) The intracellular trafficking pathway of trans- 15. Mooren OL, Galletta BJ, Cooper JA (2012) Roles for actin assembly in endocytosis. ferrin. Biochim Biophys Acta 1820(3):264–281. Annu Rev Biochem 81:661–686. 42. Chu JJ, Ng ML (2004) Infectious entry of West Nile virus occurs through a clathrin- 16. Kilcher S, Mercer J (2014) Next generation approaches to study virus entry and in- mediated endocytic pathway. J Virol 78(19):10543–10555. fection. Curr Opin Virol 4:8–14. 43. Ooi YS, Stiles KM, Liu CY, Taylor GM, Kielian M (2013) Genome-wide RNAi screen 17. Panda D, Cherry S (2012) Cell-based genomic screening: Elucidating virus-host in- identifies novel host proteins required for alphavirus entry. PLoS Pathog 9(12): teractions. Curr Opin Virol 2(6):784–792. e1003835. 18. Yasunaga A, et al. (2014) Genome-wide RNAi screen identifies broadly-acting host 44. Greber UF (2002) Signalling in viral entry. Cell Mol Life Sci 59(4):608–626. factors that inhibit arbovirus infection. PLoS Pathog 10(2):e1003914. 45. Eierhoff T, Hrincius ER, Rescher U, Ludwig S, Ehrhardt C (2010) The epidermal growth 19. Hopkins KC, et al. (2013) A genome-wide RNAi screen reveals that mRNA decapping restricts bunyaviral replication by limiting the pools of Dcp2-accessible targets for cap- factor receptor (EGFR) promotes uptake of influenza A viruses (IAV) into host cells. snatching. Genes Dev 27(13):1511–1525. PLoS Pathog 6(9):e1001099. 20. Suthar MS, Diamond MS, Gale M, Jr (2013) West Nile virus infection and immunity. 46. Meertens L, et al. (2012) The TIM and TAM families of phosphatidylserine receptors – Nat Rev Microbiol 11(2):115–128. mediate dengue virus entry. Cell Host Microbe 12(4):544 557. 21. Ikegami T, Makino S (2011) The pathogenesis of Rift Valley fever. Viruses 3(5): 47. Cao H, Chen J, Krueger EW, McNiven MA (2010) SRC-mediated phosphorylation of “ ” 493–519. dynamin and cortactin regulates the constitutive endocytosis of transferrin. Mol – 22. Marsh M, Helenius A (2006) Virus entry: Open sesame. Cell 124(4):729–740. Cell Biol 30(3):781 792. 23. Pirooz SD, et al. (2014) UVRAG is required for virus entry through combinatorial in- 48. Pelkmans L, et al. (2005) Genome-wide analysis of human kinases in clathrin- and teraction with the class C-Vps complex and SNAREs. Proc Natl Acad Sci USA 111(7): caveolae/raft-mediated endocytosis. Nature 436(7047):78–86. 2716–2721. 49. Cherry S, Perrimon N (2004) Entry is a rate-limiting step for viral infection in a Dro- 24. Brinton MA (2014) Replication cycle and molecular biology of the West Nile virus. sophila melanogaster model of pathogenesis. Nat Immunol 5(1):81–87. Viruses 6(1):13–53. 50. Hanna SL, et al. (2005) N-linked glycosylation of west nile virus envelope proteins 25. Economopoulou MA, Fragoulis EG, Sideris DC (2007) Molecular cloning and charac- influences particle assembly and infectivity. J Virol 79(21):13262–13274. terization of the human RNase kappa, an ortholog of Cc RNase. Nucleic Acids Res 51. Sessions OM, et al. (2009) Discovery of insect and human dengue virus host factors. 35(19):6389–6398. Nature 458(7241):1047–1050. 26. Kiritsi MN, Fragoulis EG, Sideris DC (2012) Essential cysteine residues for human RNase 52. Cherry S (2011) RNAi screening for host factors involved in viral infection using κ catalytic activity. FEBS J 279(7):1318–1326. Drosophila cells. Methods Mol Biol 721:375–382. 27. Rampias TN, Fragoulis EG, Sideris DC (2008) Genomic structure and expression analysis 53. Shelly S, Lukinova N, Bambina S, Berman A, Cherry S (2009) Autophagy is an essential of the RNase kappa family ortholog gene in the insect Ceratitis capitata. FEBS J component of Drosophila immunity against vesicular stomatitis virus. Immunity 30(4): 275(24):6217–6227. 588–598.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1424098112 Hackett et al. Downloaded by guest on September 27, 2021