bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
1
2
3
4
5
6
7 Resurrection of a viral internal ribosome entry site from a 700 year old ancient
8 Northwest Territories cripavirus
9
10
11 Xinying Wang and Eric Jan*
12
13
14 Department of Biochemistry and Molecular Biology, Life Sciences Institute, University of
15 British Columbia, Vancouver, BC Canada
16
17
18 *corresponding author: [email protected]
19
20
1 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
21 ABSTRACT
22 The dicistrovirus intergenic region internal ribosome entry site (IGR IRES) uses
23 an unprecedented streamlined mechanism whereby the IRES adopts a triple-
24 pseudoknot (PK) structure to directly bind to the conserved core of the ribosome and
25 drive translation from a non-AUG codon. The origin of this IRES mechanism is not
26 known. Previously, a partial fragment of a divergent dicistrovirus RNA genome, named
27 ancient Northwest territories cripavirus (aNCV), was extracted from 700-year-old
28 caribou feces trapped in a subarctic ice patch. Structural prediction of the aNCV IGR
29 sequence generated a secondary structure similar to contemporary IGR IRES
30 structures. There are, however, subtle differences including 105 nucleotides upstream
31 of the IRES of unknown function. Using filter binding assays, we showed that the aNCV
32 IGR IRES could bind to purified salt-washed human ribosomes and compete with a
33 prototypical IGR IRES for ribosomes. Toeprinting analysis using primer extension
34 pinpointed the putative start site of the aNCV IGR at a GCU alanine codon adjacent to
35 PKI. Using a bicistronic reporter RNA, the aNCV IGR IRES can direct internal ribosome
36 entry in vitro in a manner dependent on the integrity of the PKI domain. Lastly, we
37 generated a chimeric virus clone by swapping the aNCV IRES into the cricket paralysis
38 virus infectious clone. The chimeric infectious clone with an aNCV IGR IRES supported
39 translation and virus infection. The characterization and resurrection of a functional IGR
40 IRES from a divergent 700-year-old virus provides a historical framework in the
41 importance of this viral translational mechanism.
2 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
42 IMPORTANCE
43 Internal ribosome entry sites are RNA structures that are used by some positive-
44 sense monopartite RNA viruses to drive viral protein synthesis. The origin of internal
45 ribosome entry sites is not known. Using biochemical approaches, we demonstrate that
46 an RNA structure from an ancient viral genome that was discovered from a 700-year-old
47 caribou feces trapped in subarctic ice is functionally similar to modern internal ribosome
48 entry sites. We resurrect this ancient RNA mechanism by demonstrating that it can
49 support virus infection in a contemporary virus clone, thus providing insights into the
50 origin and evolution of this viral strategy.
3 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
51 INTRODUCTION
52 Due to their limited genomes, viruses do not encode ribosomes nor a full
53 complement of translation factors and as a result, they must hijack the host translation
54 machinery to drive viral protein synthesis (1). Most eukaryotic mRNAs utilize a core of
55 translation initiation factor (eIFs) to mediate the 5'cap scanning mechanism that involves
56 recruitment of the 40S ribosomal subunit and initiator Met-tRNAi, scanning of the pre-
57 initiation complex along the 5’untranslated region to locate the appropriate AUG start
58 codon and 60S ribosomal subunit joining to form an 80S elongation competent
59 ribosome. Some virus infections lead to shutdown of overall host cap-dependent
60 translation, either as an antiviral response or as a viral strategy (2). However, viruses
61 have evolved mechanisms to bypass the translation block and compete for the
62 recruitment of the ribosome to facilitate viral translation.
63 One such mechanism is through viral cis-acting elements, called internal
64 ribosome entry sites (IRESs), which enable mRNAs to recruit ribosome internally and
65 initiate translation (1). Among different types of IRESs, the dicistrovirus intergenic (IGR)
66 IRES uses the most autonomous and streamlined mechanism by directly recruiting the
67 ribosome and starting translation at a non-AUG codon (3, 4). Dicistroviruses are positive
68 sense, single-stranded RNA viruses that primarily infect arthropods (5). The
69 approximately 9 kb genome of dicistroviruses contains two open reading frames (ORFs)
70 encoding viral non-structural and structural proteins, both of which are driven by distinct
71 IRESs. Translation of the first ORF is directed by a type III-like IRES which is located at
72 the 5’ UTR (6).The IGR IRES controls translation of the downstream ORF encoding the
73 structural proteins.
4 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
74 The key to the IGR IRES mechanism is within its core secondary and tertiary
75 structures. Typically 150-200 nucleotides in length, the IGR IRES contains three main
76 pseudoknots (PKI-III) that direct distinct functions; PKII and PKIII form a core domain
77 that is responsible for recruiting 40S and 60S ribosomal subunits, while PKI contains a
78 tRNA-like anticodon:codon structure that occupies the ribosomal A-site (1, 7, 8).
79 Structural and biochemical analyses have revealed key contacts between the IGR IRES
80 and the ribosome that drive factorless translation. In the initial assembly of ribosomes
81 on the IGR IRES, stem-loops SLIV and SLV interact with uS7 and eS25 of the 40S
82 subunit, and loop L1.1 interacts with the L1 stalk of the 60S subunit (7). Mutations of
83 these terminal loop regions disrupt 40S and/or 60S recruitment. After ribosome
84 assembly, the IRES:ribosome complex undergoes a translocation event whereby the
85 PKI domain, which initially occupies the ribosomal A site, translocates to the P site and
86 allows delivery of an aminoacyl-tRNA to the empty ribosomal A site. Both of these
87 events are coordinated by elongation factors eEF1A and eEF2 and is called
88 pseudotranslocation as this does not involve peptide formation (9–11). The IRES
89 undergoes a second pseudotranslocation event leading to movement of the aminoacyl-
90 tRNA to the P site and delivery of the next aminoacyl-tRNA to the A site, where at this
91 point, peptide formation can occur. Cryo-EM analysis capturing the IRES translocating
92 through the ribosome has revealed that the IRES mediates interactions within all three
93 tRNA binding sites of the ribosome (11–13). Remarkably, the IRES undergoes a
94 dynamic conformational change after two translocation cycles whereby the PKI domain
95 in the ribosomal E site flips from an anticodon-codon mimic to a tRNA acceptor arm
96 mimic on the ribosome (8). IRES binding within and manipulation of the conserved core
5 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
97 of the ribosome explain how this IRES can function across species including yeast,
98 plant, insect and mammalian systems (14–16). The IRES can also recruit and direct
99 translation in bacteria, though the mechanism appears to be distinct (14).
100 Members of this family include cricket paralysis virus (CrPV) and Drosophila C
101 virus, which have been studied extensively and used as models for understanding
102 insect innate immunity and fundamental viral translation mechanisms including the IGR
103 IRES (5). Other members include the Taura syndrome virus, which has led to penaeid
104 shrimp outbreaks, plautia stali intestine virus, which infects aphids, and the honeybee
105 viruses, Israeli acute bee paralysis virus, acute bee virus and Kashmir bee virus, which
106 have been associated with bee disease (5, 17). The mechanism of IGR IRESs of CrPV,
107 TSV and IAPV have been studied at the biochemical and structural level (8, 18, 19).
108 The origin of the IGR IRES mechanism is not known, still analysis of present-day
109 IRESs may provide insights. Currently, the IGR IRESs studied to date can be divided
110 into two main types; Type I and II IGR IRESs are exemplified by the CrPV and IAPV
111 IGR IRES, respectively, with the main differences in the L1.1 nucleotides and that the
112 Type II IGR IRESs have an extra stem-loop within the PKI domain (20). Previous
113 studies have shown that the PKI domains of Type I and II can be functionally
114 interchanged and that the IRESs are modular, thus suggesting that the IRESs may have
115 evolved through recombination of modular domains (21, 22). Recent metagenomic
116 approaches have identified an increasing diversity of dicistro-like viral genomes and
117 IGR IRESs that may provide hints into the origins of the IRES (23, 24). Alternatively,
118 identifying ancient RNA viral genomes may provide historical context in the evolution of
119 viral strategies, however, this is challenging given the relatively labile nature of RNA.
6 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
120 However, some preserved RNA viral genomes have been discovered. The pioneering
121 benchmark in identifying ancient RNA viruses is the recovery of the influenza virus
122 genome from 1918-1919 (25, 26). Furthermore, a compete genome of ancient Barley
123 Stripe Mosaic Virus was identified from barley grain dated (~700 years) and a 1000-year
124 old RNA virus related to plant chrysoviruses was isolated from old maize samples with a
125 nearly complete genome (27, 28). Recently, Ng et al. recovered two novel viruses from
126 700-year-old caribou feces trapped in a subarctic ice patch (29), one of which is a
127 fragment of a divergent dicistrovirus RNA genome, named ancient Northwest territories
128 cripavirus (aNCV). The partial aNCV sequence that was recovered included the IGR
129 domain, thus providing an opportunity to characterize and compare the aNCV IGR to
130 contemporary dicistrovirus IGR IRESs. In this study, we examine the molecular and
131 biochemical properties of the aNCV IGR and demonstrate that the aNCV IGR directly
132 assembles ribosomes and can direct internal ribosome entry both in vitro and in vivo.
133 We show that the intact aNCV IGR including the IRES region and 105 nucleotides
134 upstream of the IRES can support viral translation and infection in a heterologous
135 dicistrovirus clone, thus highlighting the significance of this translational mechanism in
136 an ancient virus.
7 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
137 RESULTS
138 aNCV IGR IRES adopts a triple pseudoknot structure
139 Using computational and compensatory baseparing analysis, a secondary
140 structure model of the aNCV IGR was predicted (Fig. 1A). The aNCV IGR is predicted
141 to adopt an overall similar structure to modern dicistrovirus IGR IRESs, all possessing
142 the main PKI, PKII and PKIII structures and is classified as a Type I IGR IRES.
143 However, there were notable differences. An alignment analysis of the aNCV IGR with
144 classic Type I and II IGR IRESs showed that the aNCV IGR structure contains a
145 chimera mix of Type I and II features, especially at key domains that direct distinct steps
146 of IGR IRES translation (Fig. 1B). Specifically, the aNCV IGR shares sequence
147 similarities to Type I IGR IRESs within Loop 1.1B, Loop 3, and to Type II at Loop 1.1A.
148 Furthermore, whereas all Type I and II IGR IRESs contain an invariant AUUU within the
149 loop of SLIV, the aNCV contains an AUUA loop sequence. Strikingly, the aNCV IGR
150 IRES also includes an extra 105 nucleotides between the stop codon of ORF1 and the
151 start of the predicted IGR IRES structure. Several smaller stem-loops (SLIV, SLV and
152 SLVI) are predicted within this 105-nucleotide region. In summary, the aNCV IGR
153 adopts an overall secondary structure that is similar to other known dicistrovirus IGR
154 IRESs but is a chimera of Type I and II IGR IRESs at key loop domains and has an
155 extended upstream region.
156
157 aNCV IGR IRES binds tightly to human 80S ribosomes
158 Given the chimeric makeup of the aNCV IGR, we next examined whether the
159 aNCV IGR possesses properties similar to dicistrovirus IGR IRESs. The first property
8 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
160 tested was its ability to bind to purified ribosomes in vitro. The IGR IRESs directly binds
161 to purified ribosomes with high affinity (30, 31). Using an established filter-binding
162 assay, we measured 80S binding to the aNCV IGR and the CrPV IGR IRES by
163 incubating radiolabled IGR with increasing amounts of purified salt-washed human 40S
164 and 60S. The fraction of ribosome:IGR complexes were then resolved by monitoring the
165 amount of radioactivity on the nitrocellular and nylon membrane. As shown previously,
166 the wild-type but not the mutant (∆PKI/II/III) CrPV IGR IRES bound to 80S ribosomes
167 with high affinity (apparent KD 0.4 nM) (30, 31). The mutant CrPV ∆PKI/II/III IRES
168 contains mutations that disrupts all three PK basepairing (Fig. 2A). Similarly, the wild-
169 type aNCV IGR bound to purified ribosomes with similar affinity as the CrPV IGR IRES
170 (KD 0.7 nM). To confirm the binding specificity of ribosome:IGR complex interactions, we
171 performed competition assays by addition of excess unlabeled IGR RNAs to the
172 reaction prior to incubating with ribosomes. As expected, adding increasing amounts of
173 unlabeled wild-type but not mutant CrPV IGR decreased the levels of radiolabeled CrPV
174 IGR bound to 80S ribosomes (Fig. 2B, left). Similarly, adding increasing amounts of
175 unlabeled aNCV IGR also decreased the fraction of CrPV IGR IRES-ribosome complex
176 formation. In the reverse experiment, excess of unlabeled wild-type CrPV and aNCV
177 IGRs but not mutant CrPV IGR also competed for ribosomes from radiolabed aNCV
178 (Fig. 2B, right). The apparent KD measurements of the aNCV-ribosome complex binding
179 in the competition assay were similar to the direct filter binding assay (KD 0.8 nM).
180 Taken together, aNCV IGR RNA binds to purified ribosomes with high affinity and likely
181 occupies the same sites on the ribosome as the CrPV IGR IRES.
182
9 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
183 RNase protection analysis of aNCV IGR-ribosome complexes
184 We next examined whether aNCV IGR RNA binds within the intersubunit space
185 of the ribosome. We previously developed a novel ribosome protection assay, whereby
186 localization of the RNA in the intersubunit space or solvent side of the ribosome can be
187 inferred by its susceptibility to RNase I-mediated degradation (data not shown).
188 Radiolabeled CrPV IGR IRES in complex with purified ribosomes were incubated with
189 RNase I and then the RNA was isolated and resolved on an urea-PAGE gel. RNase I
190 treatment of radiolabeled CrPV IGR IRES prebound to the ribosome resulted in faster
191 migrating fragments, indicative of sequences that were protected by the ribosome from
192 degradation (Fig. 3). Specifically, compared to the full-length CrPV IGR IRES (208
193 nucleotides), RNase I treatment led to ribosome-protected fragments ranging from 50-
194 160 nucleotides. Sequencing of these IRES fragments and mapping back to the IRES
195 revealed a signature core domain consisting of PKII and PKIII structures that were
196 primarily protected by the ribosome from RNase I (unpublished work). Importantly,
197 RNase I treatment of the mutant PKI/II/III CrPV IGR IRES (TM) incubated with
198 ribosomes did not lead to protected fragments, indicating that the concentration of
199 RNase I is sufficient to degrade the unprotected RNA. Similar to that observed with the
200 CrPV IGR IRES, RNase I-treatment of aNCV IGR-ribosome complexes resulted in
201 faster migrating fragments (Fig 3). Compared to the full-length aNCV IGR IRES (291
202 nucleotides), several protected fragments ranging from 70 to 200 nucleotides were
203 detected. These results indicate that the aNCV IGR binds to the intersubunit core of the
204 ribosome.
205
10 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
206 Determination of the aNCV IGR IRES initiation site
207 We next determined the start site of the aNCV IGR, which is predicted to be at a
208 GCU codon and adjacent to the PKI domain (Fig. 1). To investigate this, we monitored
209 initiating ribosomes assembled on the aNCV IGR in rabbit reticulocyte lysates (RRL) by
210 toeprinting, an established primer extension assay (32). Briefly, when reverse
211 transcriptase encounters the ribosome assembled on the IRES, a truncated cDNA
212 product is generated, which can be detected on a urea-PAGE gel. As shown previously,
213 ribosomes assembled on the wild-type but not mutant CrPV IGR IRES in RRL in the
214 presence of cycloheximide resulted in toeprints at CC6232-3, which is +20-21
215 nucleotides from the PKI domain CCU triplet, given that the first C is +1 (Fig. 4B) (30).
216 This result showed that the ribosome can translocate on the IGR IRES two cycles in the
217 presence of cycloheximide (3, 9, 10). Ribosomes assembled on the aNCV IGR in the
218 presence of cycloheximide resulted in a prominent toeprint at U308, which is +20
219 nucleotides from the AUC codon, given that A is +1, which is consistent that ribosomes
220 have translocated two cycles (Fig.4B). To confirm that this toeprint is representative of
221 translocated ribosomes on the aNCV IGR, mutations within the PKI domain were
222 generated that disrupted PKI basepairing (mPKI #1 and #2) or compensatory (comp)
223 mutations that restored PKI basepairing (Fig. 4B). Toeprint U308 was only detected
224 when the PKI domain is intact, thus supporting the conclusion that the initiating codon is
225 the GCU alanine adjacent to the PKI of aNCV IGR.
226 To further confirm direct ribosome binding on the IRES, we performed toeprinting
227 analysis of purified ribosomes assembled on the aNCV IGR IRES. As shown previously,
228 purified ribosomes assembled on the wild-type but not mutant CrPV IGR IRES resulted
11 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
229 in toeprints at CA6226-7, which is +13-14 nucleotides from the PKI domain CCU triplet,
230 given that the first C is +1 (Fig. 4C). Purified ribosomes assembled on the aNCV IGR
231 resulted in a toeprint at C302, which is +13 nucleotides from the AUC codon, given that
232 A is +1 (Fig. 4C), thus supporting the conclusion that ribosomes initially assembled on
233 the aNCV IGR contains the AUC codon in the ribosomal A site and the adjacent GCU
234 codon is the initiating start site.
235
236 aNCV IGR IRES directs translation in vitro
237 To determine whether the aNCV IGR has IRES activity, we inserted the aNCV
238 IGR in a bicistronic reporter RNA construct and assessed translation in RRL (Fig. 5A,
239 5B). In vitro transcribed RNA was incubated in RRL in the presence of [35S]-
240 methionine/cysteine to monitor protein expression level. Wild-type but not mutant
241 (mPKI) aNCV IGR IRES was active in RRL in vitro (Fig. 5A lane 3). Importantly,
242 mutating the aNCV PKI region in the reporter abolished FLuc activity, indicating that
243 IRES activity is compromised (Fig. 5A, lanes 4 and 5). In contrast, compensatory
244 mutations that restored base-pairing rescued aNCV IGR IRES activity to ~50% of wild
245 type (Fig. 5A, lane 6). Unlike some dicistrovirus IGR IRESs that can direct +1 frame
246 translation (18, 33), the aNCV IGR IRES does not in RRL (data not shown). In
247 summary, the aNCV IGR is a bona fide IRES.
248 To investigate whether the upstream region of the aNCV IRES is important for
249 IRES activity, we monitored translation of wild-type and a mutant aNCV IRES, where
250 the upstream region has been deleted (1-99). In RRL, both wild-type and mutant (1-
12 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
251 99) aNCV can direct IRES activity to similar levels (Fig. 5B), indicating that this
252 upstream region does not contribute to IRES activity in vitro.
253
254 Chimeric CrPV clone containing the aNCV IGR is infectious
255 Having demonstrated that the aNCV IGR can bind to ribosomes directly and
256 drive IRES activity from a non-AUG codon, we next examined whether the aNCV can
257 support infection in a more physiological system. Because only a partial sequence of
258 the aNCV genome was recovered in the 700-year-old caribou feces (29), we used a
259 heterologous approach by generating chimeric dicistrovirus CrPV clones by replacing
260 IGR IRES with either the full-length aNCV (CrPV-aNCV) or the mutant aNCV (1-99),
261 where the upstream 99 nucleotides is deleted (34, 35) (Fig. 6A). We first addressed
262 whether the aNCV IGR can support translation in the infectious clone. In Sf21 extracts,
263 incubation of in vitro transcribed chimeric CrPV-aNCV led to translation of nonstructural
264 and structural proteins that are similar to that of the wild-type CrPV infectious clone (Fig.
265 6B). Of note, the chimeric CrPV-aNCV containing either the full-length or the mutant
266 aNCV (1-99) led to similar expression of viral proteins in vitro, consistent with previous
267 data showing that the upstream 99 nucleotides do not affect aNCV IRES activity.
268 Compared to the CrPV clone RNA, translation of structural proteins VP1 and VP2/3 was
269 compromised in reactions containing the full-length and the mutant chimeric CrPV-
270 aNCV RNA, indicating a defect in aNCV IGR IRES translation in vitro under these
271 conditions (Fig. 6B). The mutant CrPV containing a stop codon within ORF1 only
272 resulted in expression of the unprocessed ORF2 polyprotein (35).
13 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
273 To determine whether the CrPV-aNCV clone is infectious, we transfected in vitro
274 transcribed RNA into S2 cells and monitored viral protein expression by immunoblotting.
275 Transfection of the wild-type CrPV and CrPV-aNCV containing the full-length aNCV IGR
276 RNA resulted in expression of the CrPV nonstructural protein 1A and the structural
277 protein VP2 at 120 hours post-transfection (h.p.i) (Fig. 6C), which in previous
278 experience is indicative of productive infection (35). By contrast, viral proteins were not
279 detected in cells transfected with the the CrPV-aNCV (1-99) clone. To confirm these
280 results, we monitored viral replication by RT-PCR analysis. Mirroring the viral protein
281 expression, CrPV RNA was detected by RT-PCR after transfection of the CrPV-aNCV
282 and CrPV RNA but not the CrPV-aNCV (1-99) RNA (Fig. 6D). Note that the CrPV RNA
283 was detected at 72 hours h.p.t. whereas CrPV-aNCV was not detected until 144 h.p.t.,
284 suggesting that the replication of the chimeric virus is delayed.
285 We attempted to propagate the CrPV-aNCV virus from transfected cells by
286 reinfecting and passaging in naïve S2 cells. Despite multiple attempts, the CrPV-aNCV
287 (1-99) did not lead to productive virus as measured by viral titres. However, the CrPV-
288 aNCV yielded productive virus (1.15 X 10^10 FFU/mL). We sequenced verified that the
289 aNCV IGR was intact in viral RNA isolated from the propagated CrPV-aNCV (data not
290 shown). To further validate virus production, we infected S2 cells with CrPV-aNCV and
291 monitored viral expression by immunoblotting. Similar to that observed with CrPV
292 infection, the CrPV 1A protein produced from CrPV-aNCV was detected at 2-10 h.p.i.,
293 albeit VP2 expression was reproducibly detected until 4 h.p.i., thus likely reflecting the
294 decreased IRES activity of aNCV compared to CrPV (Fig. 6E). In summary, we have
14 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
295 demonstrated that the full-length aNCV IGR can support virus infection in a
296 heterologous infectious clone.
15 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
297 DISCUSSION
298 Tracking back and identifying ancient RNA viruses may provide insights into the
299 evolution and origin of present-day viruses including the mechanisms that permit virus
300 translation, replication and host virus interactions. In this study, we have examined the
301 IGR of an ancient dicistrovirus that was discovered from 700-year-old ice-preserved
302 caribou feces. To our knowledge, the aNCV genome has not be identified in RNA
303 metagenomic analysis; thus, this study is the first to characterize the oldest IRES to
304 date and provides insights into the origins of this type of IRES. Using molecular and
305 biochemical approaches, we demonstrate that the aNCV IGR possesses IRES activity
306 using a mechanism that is similar to that found in present day Type IV dicistrovirus
307 IRESs. The aNCV IGR can direct ribosome assembly directly and initiates translation
308 from a non-AUG codon. The aNCV IGR IRES can support virus infection in a
309 heterologous infectious clone; thus a functional aNCV IGR IRES has been resurrected
310 from an ancient RNA virus.
311 The secondary structure model of the core aNCV IGR resembles classic Type I
312 IGR IRESs containing all three PKs. However, there are subtle differences, including
313 the L1.1 bulge region, responsible for 60S recruitment, which is comprised of a mixture
314 of Type I and II IRESs elements. Additionally, aNCV IGR contains 105 nucleotides
315 upstream of the core IGR IRES, within which several stem-loops are predicted. We
316 showed that this upstream sequence is not required for aNCV IGR IRES translation (Fig
317 5, 6) but is required for virus infection using the CrPV-aNCV. RT-PCR analysis of the
318 viral RNA suggests that the upstream region has a role in replication; however, it may
319 have other roles in the viral life cycle such as RNA stability or replication. Of note, a few
16 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
320 dicistrovirus IGRs also contain sequences beyond the core IRES structure. For
321 example, the Rhopalosiphum padi virus (RhPV) contains an IGR that is over 500
322 nucleotides. Similar to the aNCV IRES, the extra sequences within the RhPV IGR do
323 not affect the Type I RhPV IGR IRES (36). It will be interesting to investigate more in
324 detail how these extra IGR sequences play a role in the viral life cycle.
325 To date, there are increasing numbers of dicistrovirus-like genomes identified via
326 metagenomic studies (23, 24). From limited analysis, it is clear that the IGR IRES
327 structures can be distinct (ie Type I and II) and can have novel functions (ie +1 reading
328 frame selection). For example, the Halastavi árva virus IRES uses a unique but similar
329 streamlined mechanism as the CrPV IRES, where it bypasses the first
330 pseduotranslocation event to direct translation initiation (37). Moreover, the IGR IRES
331 can direct translation in multiple species (14–16), suggesting that the IGR IRES
332 mechanism may be a molecular fossil of an RNA-based translational control strategy
333 that existed in ancient viruses. IGR IRESs likely evolved through recombination of
334 independent functional domains; the PKI domains can be functionally swapped between
335 Type I and II IRESs (21, 22). Furthermore, IRES elements in unrelated viruses appear
336 to have disseminated between these viruses via horizontal gene transfer (38), further
337 supporting the idea that dicistrovirus IGR IRESs may have originated via recombination
338 events.
339 Given that the aNCV IGR IRES structure resembles an IGR IRES found in
340 contemporary dicistrovirus genomes, it was not surprising that it functions by a similar
341 mechanism. Relatively speaking, the IGR IRES from a 700-year-old RNA viral genome
342 is far from the presumed primordial IRES. However, this study provides proof-in-concept
17 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
343 that this viral RNA translation mechanism can be resurrected and investigated to
344 provide context in the evolution of viral mechanisms. The challenge in identifying more
345 ancient RNA viral genomes and mechanisms has been and will always be in capturing
346 intact RNA viral genomes as RNA, compared to DNA virus counterparts, are unstable
347 apt to degrade over time. The remarkable discoveries by Ng et al and others of ancient
348 RNA viral genomes (27–29) provide hope in the pursuit of identifying more ancient
349 viruses that sheds light into the origins of contemporary viral mechanisms.
18 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
350 MATERIALS and METHODS
351 Cell culture
352 Drosophila Schneider line 2 (S2) cells were maintained and passaged at 25°C in
353 Shields and Sang M3 insect medium (Sigma-Aldrich) supplemented with 10% fetal
354 bovine serum.
355
356 Virus infection
357 S2 cells were infected at the desired multiplicity of infection in minimal phosphate-
358 buffered saline (PBS) at 25°C. After 30 min absorption, complete medium was added
359 and harvested at the desired time point. Virus titres were monitored as described (Au et
360 al. 2017) using immunofluorescence (anti-VP2).
361
362 DNA constructs
363 The aNCV IGR sequence (Accession KJ938718.1) was synthesized (Twist
364 Biosciences) and cloned into pEJ4 containing EcoRI and NcoI sites upstream of the
365 firefly luciferase (FLuc) open reading frame. For the bicistronic reporter construct, PCR-
366 amplified full-length aNCV IGR or truncated aNCV IGR (∆1-99) was ligated into the
367 standard bicistronic reporter construct (pEJ253) using EcoRI and NdeI sites.
368
369 CrPV/aNCV chimeric infectious clone constructs
370 The chimeric CrPV-aNCV infectious clone was derived from the full-length CrPV
371 infectious clone (pCrPV-3; Accession KP974707) (35) using Gibson assembly (New
19 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
372 England BioLabs), per the manufacturer's instructions. All constructs were verified by
373 sequencing.
374
375 In vitro transcription and translation
376 Monocistronic and bicistronic reporters were linearized with NcoI and BamHI,
377 respectively. pCrPV-3 and chimeric CrPV-aNCV clones were linearized with Eco53kI.
378 RNAs were in vitro transcribed using a bacteriophage T7 RNA polymerase reaction and
379 RNA was purified using an RNeasy Kit (Qiagen). Radiolabeled RNAs were bulk labelled
380 by incorporating α- [32P] UTP (3000 Ci/mmol). The integrity and purity of RNAs were
381 confirmed by agarose gel analysis. Uncapped bicistronic RNAs were first pre-folded by
382 heating at 65 °C for 3 min, followed by the addition of 1× Buffer E (final concentration:
383 20 mM Tris pH 7.5, 100 mM KCl, 2.5 mM MgOAc, 0.25 mM Spermidine, and 2 mM
384 DTT) and slowly cooled at room temperature for 10 min. The pre-folded RNAs (20-40
385 ng/μl) were incubated in RRL containing 8 U Ribolock inhibitor (Fermentas), 20 μM
386 amino acid mix minus methionine, 0.3 μl [35S]-methionine/cysteine (PerkinElmer,
387 >1000 Ci/mmol), and 75 mM KOAc pH 7.5 at 30 oC for 1 hr. For the infectious clones, 2
388 μg RNA was incubated at 30 oC for 2 hr in Spodoptera frugiperda (Sf21) extract
389 (Promega) in the presence of [35S]methionine-cysteine (Perkin-Elmer) and an additional
390 40 mM KOAc and 0.5 mM MgCl2. The translated proteins were resolved using SDS-
391 PAGE and analyzed by phosphorimager analysis (Typhoon, GE life sciences).
392
393 Purification of the 40S and 60S subunits
20 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
394 Ribosomal subunits were purified from HeLa cell pellets (National Cell Culture
395 Centre) as described (30). In brief, HeLa cells were lysed in a lysis buffer (15 mM Tris–
396 HCl (pH 7.5), 300 mM NaCl, 6 mM MgCl2, 1% (v/v) Triton X-100, 1 mg/ml
397 heparin). Debris was removed by centrifuging at 23,000 g and the supernatant was
398 layered on a 30% (w/w) cushion of sucrose in 0.5 M KCl and centrifuged at 100,000 g to
399 pellet crude ribosomes. Ribosomes were gently resuspended in buffer B (20 mM Tris–
400 HCl (pH 7.5), 6 mM magnesium acetate, 150 mM KCl, 6.8% (w/v) sucrose, 1 mM DTT)
401 at 4 oC, treated with puromycin (final 2.3 mM) to release ribosomes from mRNA, and
402 KCl (final 500 mM) was added to wash and separate 80S ribosomes into 40S and 60S.
403 The dissociated ribosomes were then separated on a 10%–30% (w/w) sucrose gradient.
404 The 40S and 60S peaks were detected by measuring the absorbance at 260 nm.
405 Corresponding fractions were pooled, concentrated using Amicon Ultra spin
406 concentrators (Millipore) in buffer C (20 mM Tris–HCl (pH 7.5), 0.2 mM EDTA, 10 mM
407 KCl, 1 mM MgCl2, 6.8% sucrose). The concentration of 40S and 60S subunits was
408 determined by spectrophotometry, using the conversions 1 A260 nm = 50 nM for 40S
409 subunits, and 1 A260 nm = 25 nM for 60S subunits.
410
411 Filter-binding assays
412 RNAs (final 0.5 nM) were preheated at 65 °C for 3 min, followed by the addition
413 of 1× Buffer E and slowed cooling in a water bath preheated to 60 °C for 20 min. The
414 prefolded RNAs, with 50 ng/μl of non-competitor RNA, were incubated with an
415 increasing amount of 40S subunits from 0.1 nM to 100 nM, and 1.5 fold excess of 60S
416 subunits for 20 min at room temperature. Non-competitor RNAs were in vitro transcribed
21 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
417 from the pcDNA3 vector 880-948 nucleotides. Reactions were then loaded onto a Bio-
418 Dot filtration apparatus (Bio-Rad) including a double membrane of nitrocellulose and
419 nylon pre-washed with buffer E. Membranes were then washed three times with buffer
420 E, dried, and the radioactivity was imaged and quantified by phosphorimager analysis.
421 The dissociation constant is determined by the formula,
422
[퐴퐵] [퐵] = 푓푚푎푥( ) [퐴]푡표푡푎푙 [퐵] + 퐾퐷
423
424 where [A] is the concentration of RNAs, [B] is the concentration of ribosomes, [AB] is
425 the concentration of RNAs bound to the ribosomes, fmax is the saturation point, and KD
426 is the dissociation constant.
427 Bulk-labeled RNAs (final 0.5 nM), IRES competitors, and non-competitor RNAs
428 (50 ng/μl) were pre-folded in buffer E. Unlabeled IRES competitors were added in
429 increasing concentrations from 2 nM to 250 nM. RNAs were then incubated with 6 nM
430 40S and 9 nM 60S subunits at room temperature for 20 min. Reactions were then
431 loaded onto the Bio-Dot filtration apparatus, and data were fitted to the Linn-Riggs
432 equation that describes competitive ligand binding to the target:
[푆](1 − 휃) 휃 = 퐾퐷{(1 + [퐶])퐾퐶} + [푅](1 − 휃)
433
434 Where [S] is 80S ribosome concentration, [R] is radiolabelled RNA concentration, θ is
435 fraction of radiolabelled RNAs bound to ribosomes, [C] is competitor RNA
22 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
436 concentration, KD and KC are dissociation constants of labelled RNAs and competitor
437 RNAs, respectively.
438
439 Ribosome protection assay
440 [32P]-labeled RNAs (final 0.1 μM) were pre-folded in buffer E and incubated with
441 0.6 μM 40S and 0.9 μM 60S at room temperature for 20 min. 1 μL of 1U/μL of RNase I
442 (Ambion) was added to the mixture and incubated at 20 oC for 1 hr. RNAs without
443 RNase I treatment were incubated at 20 oC with the same time length. RNAs from
444 mixtures with or without RNase I treatment were TRIZOL-extracted and loaded onto 6%
445 (w/v) polyacrylamide/8M urea gels to separate them. RNA Ladders (Ambion) were
446 synthesized per manufacturer’s instruction. Gels were dried and imaged by
447 phosphorimager analysis.
448
449 Toeprinting/primer extension analysis
450 Toeprinting analysis of ribosomal complexes in RRL was performed as previously
451 described (3). 0.4 µg of bicistronic WT or mutant CrPV IGR IRES RNAs and WT or
452 mutant aNCV IGR IRES RNAs were annealed to primer 5′-
453 GTAAAAGCAATTGTTCCAGGAACCAG-3′ and primer 5′-
454 GTTAGCAGACTTCCTCTGCCCTCTC-3′, respectively in 40 mM Tris (pH 7.5) and 0.2
455 mM EDTA by slow cooling from 65°C to 30°C. Annealed RNAs were added to RRL pre-
456 incubated with 0.68 mg/mL cycloheximide and containing 20 µM amino acid mix, 8 units
457 of Ribolock (Fermentas) and 154 nM final concentration of potassium acetate (pH 7.5).
458 The reaction was incubated at 30°C for 20 minutes. Toeprinting analysis using purified
23 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
459 ribosomes was performed as followed: 75 ng of RNAs were annealed to primers in 40
460 mM Tris·Cl, pH 7.5, and 0.2 mM EDTA by slow cooling from 65°C to 35°C. Annealed
461 RNAs were incubated in Buffer E (containing 100 mM KCl) containing 40S (final
462 concentration 100 nM); 60S (final concentration 150 nM); ribolock (0.02 U/μl) at 30°C for
463 20 mins. Following incubation, ribosome positioning was determined by primer
464 extension/reverse transcription using AMV reverse transcriptase (1 U/μl) (Promega) in
465 the presence of 125 µM of each of dTTP, dGTP, dCTP, 25 µM dATP, 0.5 µL of α- [32P]
466 dATP (3.33 µM, 3000 Ci/mmol), 8 mM MgOAc, in the final reaction volume. The reverse
467 transcription reaction was incubated at 30°C for 1 hour, after which it was quenched by
468 the addition of STOP solution (0.45 M NH4OAc, 0.1% SDS, 1 mM EDTA). Following the
469 reverse transcription reaction, the samples were extracted by phenol/chloroform (twice),
470 chloroform alone (once), and ethanol-precipitated. The cDNA was analyzed under
471 denaturing conditions on 6% (w/v) polyacrylamide/8M urea gels, which were dried and
472 subjected to phosphorimager analysis.
473
474 RNA transfection
475 3 μg of in vitro-transcribed RNA derived from the pCrPV-3 or chimeric clones was
476 transfected into 3 x 106 S2 cells using lipofectamine 2000 reagent (Life Technologies)
477 per the manufacturer’s instructions.
478
479 RT-PCR and sequence confirmation
480 Viruses were passaged three times after transfection of viral RNAs in S2 cells. Total
481 RNA was isolated from S2 cells using TRIzol reagent. RT was performed using 1 μg of
24 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
482 RNA using LunaScript™ RT SuperMix Kit (New England BioLabs) per manufacturer’s
483 instructions. For PCR amplification of the negative-sense strand of the CrPV genome to
484 detect replication, tagged primer (5’-CTATGGATCCATGGGAGAAGATCAGCAAAT-3′;
485 tag is underlined) and primer (5′-GTGGCTGAAATACTATCTCTGG-3′) were used. Rps6
486 was amplified using primers (5’-CGATATCCTCGGTGACGAGT-3’) and (5’-
487 CCCTTCTTCAAGACGACCAG-3’).
488
489 Western blotting
490 Cells were washed once using 1× PBS and harvested in lysis buffer (20 mM
491 HEPES, 150 mM sodium chloride, 1% Triton X-100, 10% glycerol, 1 mM EDTA, 10 mM
492 tetrapyrophosphate, 100 mM sodium fluoride, 17.5 mM β-glycerophosphate, protease
493 inhibitor cocktail [Roche]). Equal amounts of protein lysate were resolved by SDS-
494 PAGE and subsequently transferred onto polyvinylidene difluoride Immobilon-FL
495 membrane (Millipore). Following transfer, the membrane was blocked for 30 min at
496 room temperature in 5% skim milk in Tris-buffered saline containing 0.1% Tween 20
497 (TBST) and incubated for 1 h with rabbit polyclonal antibody raised against CrPV 1A
498 (1:1,000) or CrPV VP2 (1:5,000). Membranes were washed 3 times with TBST and
499 subsequently incubated with IRDye 800CW goat anti-rabbit IgG (1:20,000; Li-Cor
500 Biosciences) for 1 h at room temperature.
25 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
501 ACKNOWLEDGEMENTS
502 We thank Hilda Au and the Jan lab for critical analysis and reading of the manuscript.
503 This work was supported by CIHR (PJT-148761).
26 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
504 FIGURE LEGEND
505 Figure 1. A) Secondary structure model of the aNCV IGR (accession KJ938718.1)
506 Pseudoknots (PK) I, II, III are indicated. Orange-colored nucleotides at Loop (L)1.1B, L3
507 denote sequences conserved in Type I IGR IRES, and blue-colored nucleotides at
508 L1.1A indicate sequences conserved in Type II IGR IRES. Red-colored nucleotides at
509 stem loop (SL) IV are unique in sequence in aNCV IGR IRES. Green-colored
510 nucleotides represent IGR sequences outside the core IRES. The predicted start codon
511 is GCU adjacent to the PKI domain. Numbering refers to the nucleotide position in the
512 IGR IRES as the genome of aNCV is not complete. (B) Alignment of Type I and II IGR
513 IRESs with aNCV IGR.
514
515 Figure 2. Affinity of 80S-aNCV IGR IRES complexes. (A) Filter binding assays. [32P]-
516 aNCV IGR IRES, CrPV IGR IRES or mutant (PKI/II/III) CrPV IGR IRES (0.5 nM) were
517 incubated with increasing amounts of purified salt-washed 80S. The fraction bound
518 were quantified by phosphorimager analysis. (B) Competition assays. Quantification of
519 radiolabeled 80S-CrPV IGR IRES (left) or 80S-aNCV IGR IRES (right) complex
520 formation with increasing amounts of cold competitor RNAs (aNCV IGR IRES, CrPV
521 IGR IRES or mutant (PKI/II/III) CrPV IGR IRES.) Shown are the averages standard
522 deviation from at least three independent experiments.
523
524 Figure 3. aNCV IGR IRES bound to 80S is RNase I-resistant. Radiolabeled RNAs were
525 incubated with 80S (0.6 μM) for 20 min before adding RNase I (1 U) for 1 hr. RNAs from
526 RNase I treated/untreated reactions were TRIZOL-extracted and loaded onto urea-
27 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
527 PAGE and visualized by phosphorimager analysis. Shown is a representative gel from
528 at least two independent experiments.
529
530 Figure 4. Toeprinting analysis of 80S-aNCV IRES complexes. (A) Schematic of IGR
531 IRES PKI region. Mutations within the PKI region are annotated. (B) Primer extension
532 analysis was performed on in vitro transcribed bicistronic RNAs containing the indicated
533 wild-type or mutant CrPV IRES (left) or aNCV IRES or CrPV IRES (right) incubated in
534 rabbit reticulocyte lysate (RRL) pretreated with cycloheximide (0.68 mg/mL). For the
535 aNCV IRES, the sequencing ladder corresponds to nucleotides 279-331. The major
536 toeprint is indicated at right, which is twenty nucleotides downstream of AUC289-91 of PKI
537 given that the A is +1. (C) Primer extension analysis on in vitro transcribed bicistronic
538 RNAs containing the indicated wild-type or mutant aNCV IRES incubated with purified
539 ribosomes. Sequencing ladder corresponds to aNCV IRES nucleotides 269-319. The
540 major toeprint is indicated at right, which is 13 nucleotides downstream of AUC289-91.
541 Shown is a representative gel from at least three independent experiments.
542
543 Figure 5. In vitro translation assays in RRL. (A) Schematic of bicistronic reporter
544 construct containing the IRES within the intergenic region. (B) In vitro-transcribed
545 bicistronic reporter RNAs were incubated in rabbit reticulocyte lysate (RRL) for 60 min in
546 the presence of [35S]-methionine/cysteine. The reactions were loaded on a SDS-PAGE
547 gel, which was then dried and imaged by phosphorimager analysis. (top) Integrity of the
548 in vitro transcribed RNAs are shown. (bottom) Quantification of the radiolabeled Renilla
28 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
549 (RLuc) and firefly (FLuc) luciferase proteins. ns, P > 0.05; **, P < 0.005. Shown are the
550 averages standard deviation from at least three independent experiments.
551
552 Figure 6. Chimeric CrPV-aNCV clone is infectious in Drosophila S2 cells. (A) Schematic
553 of the chimeric CrPV-aNCV clone replacing CrPV IGR IRES with that of the aNCV
554 IRES. (B) In vitro translation of CrPV and CrPV-aNCV RNAs in Sf21 extracts. CrPV-
555 ORF1-STOP contains a stop codon within the N-terminal ORF1, thus preventing
556 expression of the non-structural proteins. Reactions were resolved by SDS-PAGE and
557 visualized by autoradiography. (C) Immunoblotting of CrPV VP2 structural protein and
558 1A nonstructural protein. D) RT-PCR of viral negative-strand RNA from Drosophila S2
559 cells transfected with the indicated viral clone RNAs at 72 and 144 hours after
560 transfection. (E) Immunoblotting of CrPV 1A and VP2 proteins from lysates of
561 Drosophila S2 cells infected with CrPV or CrPV-aNCV chimera (MOI 5) at the indicated
562 h.p.i.. Shown are representative gels from at least three independent experiments.
29 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
563 REFERENCES
564 1. Mailliot J, Martin F. 2018. Viral internal ribosomal entry sites: four classes for one
565 goal. Wiley Interdiscip Rev RNA 9:e1458.
566 2. Jaafar ZA, Kieft JS. 2018. Viral RNA structure-based strategies to manipulate
567 translation. Nat Rev Microbiol 17:110–123.
568 3. Wilson JE, Pestova T V, Hellen CU, Sarnow P. 2000. Initiation of protein
569 synthesis from the A site of the ribosome. Cell 102:511–20.
570 4. Sasaki J, Nakashima N. 2000. Methionine-independent initiation of translation in
571 the capsid protein of an insect RNA virus. Proc Natl Acad Sci 97:1512–1515.
572 5. Warsaba R, Sadasivan J, Jan E. 2020. Dicistrovirus-Host Molecular Interactions.
573 Curr Issues Mol Biol 34:83–112.
574 6. Gross L, Vicens Q, Einhorn E, Noireterre A, Schaeffer L, Kuhn L, Imler J-L, Eriani
575 G, Meignin C, Martin F. 2017. The IRES5′UTR of the dicistrovirus cricket paralysis
576 virus is a type III IRES containing an essential pseudoknot structure. Nucleic
577 Acids Res 45:8993–9004.
578 7. Kerr CH, Ma ZW, Jang CJ, Thompson SR, Jan E. 2016. Molecular analysis of the
579 factorless internal ribosome entry site in Cricket Paralysis virus infection. Sci Rep
580 6:37319.
581 8. Pisareva VP, Pisarev A V, Fernández IS. 2018. Dual tRNA mimicry in the Cricket
582 Paralysis Virus IRES uncovers an unexpected similarity with the Hepatitis C Virus
583 IRES. Elife 7:e34062.
584 9. Jan E, Kinzy TG, Sarnow P. 2003. Divergent tRNA-like element supports
585 initiation, elongation, and termination of protein biosynthesis. Proc Natl Acad Sci
30 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
586 U S A 100:15410–15415.
587 10. Pestova T V., Hellen CUT. 2003. Translation elongation after assembly of
588 ribosomes on the Cricket paralysis virus internal ribosomal entry site without
589 initiation factors or initiator tRNA. Genes Dev 17:181–186.
590 11. Fernández IS, Bai X-C, Murshudov G, Scheres SHW, Ramakrishnan V. 2014.
591 Initiation of translation by cricket paralysis virus IRES requires its translocation in
592 the ribosome. Cell 157:823–31.
593 12. Nikolay R, Hilal T, Qin B, Mielke T, Bürger J, Loerke J, Textoris-Taube K,
594 Nierhaus KH, Spahn CMT. 2018. Structural Visualization of the Formation and
595 Activation of the 50S Ribosomal Subunit during In Vitro Reconstitution. Mol Cell
596 70:881-893.e3.
597 13. Svidritskiy E, Demo G, Loveland AB, Xu C, Korostelev AA. 2019. Extensive
598 ribosome and RF2 rearrangements during translation termination. Elife 8:e46850.
599 14. Colussi TM, Costantino DA, Zhu J, Donohue JP, Korostelev AA, Jaafar ZA, Plank
600 TDM, Noller HF, Kieft JS. 2015. Initiation of translation in bacteria by a structured
601 eukaryotic IRES RNA. Nature 519:110–113.
602 15. Thompson SR, Gulyas KD, Sarnow P. 2001. Internal initiation in Saccharomyces
603 cerevisiae mediated by an initiator tRNA/eIF2-independent internal ribosome
604 entry site element. Proc Natl Acad Sci U S A 98:12972–12977.
605 16. Wilson JE, Powell MJ, Hoover SE, Sarnow P. 2000. Naturally occurring dicistronic
606 cricket paralysis virus RNA is regulated by two internal ribosome entry sites. Mol
607 Cell Biol 20:4990–9.
608 17. Bonning BC, Miller WA. 2010. Dicistroviruses. Annu Rev Entomol. Annu Rev
31 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
609 Entomol.
610 18. Ren Q, Wang QS, Firth AE, Chan MMY, Gouw JW, Guarna MM, Foster LJ, Atkins
611 JF, Jan E. 2012. Alternative reading frame selection mediated by a tRNA-like
612 domain of an internal ribosome entry site. Proc Natl Acad Sci U S A 109:e630-
613 639.
614 19. Acosta‐ Reyes F, Neupane R, Frank J, Fernández IS. 2019. The Israeli acute
615 paralysis virus IRES captures host ribosomes by mimicking a ribosomal state with
616 hybrid tRNAs. EMBO J 38:e102226.
617 20. Nakashima N, Uchiumi T. 2009. Functional analysis of structural motifs in
618 dicistroviruses. Virus Res 139:137–147.
619 21. Hertz MI, Thompson SR. 2011. In vivo functional analysis of the Dicistroviridae
620 intergenic region internal ribosome entry sites. Nucleic Acids Res 39:7276–7288.
621 22. Jang CJ, Eric J. 2010. Modular domains of the Dicistroviridae intergenic internal
622 ribosome entry site. RNA 16:1182–1195.
623 23. Shi M, Lin XD, Tian JH, Chen LJ, Chen X, Li CX, Qin XC, Li J, Cao JP, Eden JS,
624 Buchmann J, Wang W, Xu J, Holmes EC, Zhang YZ. 2016. Redefining the
625 invertebrate RNA virosphere. Nature 540:539–543.
626 24. Wolfe ND, Dunavan CP, Diamond J. 2007. Origins of major human infectious
627 diseases. Nature. Nature Publishing Group.
628 25. Taubenberger JK, Reid AH, Krafft AE, Bijwaard KE, Fanning TG. 1997. Initial
629 genetic characterization of the 1918 “Spanish” influenza virus. Science (80- )
630 275:1793–1796.
631 26. Taubenberger JK, Reid AH, Lourens RM, Wang R, Jin G, Fanning TG. 2005.
32 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
632 Characterization of the 1918 influenza virus polymerase genes. Nature 437:889–
633 893.
634 27. Smith O, Clapham A, Rose P, Liu Y, Wang J, Allaby RG. 2014. A complete
635 ancient RNA genome: Identification, reconstruction and evolutionary history of
636 archaeological Barley Stripe Mosaic Virus. Sci Rep 4:4003.
637 28. Peyambari M, Warner S, Stoler N, Rainer D, Roossinck J. 2019. A 1,000-Year-
638 Old RNA Virus. J Virol 93:e01188-18.
639 29. Ng TFF, Chen LF, Zhou Y, Shapiro B, Stiller M, Heintzman PD, Varsani A,
640 Kondov NO, Wong W, Deng X, Andrews TD, Moorman BJ, Meulendyk T, Mackay
641 G, Gilbertson RL, Delwart E, Palese P. 2014. Preservation of viral genomes in
642 700-y-old caribou feces from a subarctic ice patch. Proc Natl Acad Sci U S A
643 111:16842–16847.
644 30. Jan E, Sarnow P. 2002. Factorless Ribosome Assembly on the Internal Ribosome
645 Entry Site of Cricket Paralysis Virus. J Mol Biol 324:889–902.
646 31. Jang CJ, Lo MCY, Jan⁎ E. 2008. Conserved Element of the Dicistrovirus IGR
647 IRES that Mimics an E-site tRNA/Ribosome Interaction Mediates Multiple
648 Functions. J Mol Biol 387:42–58.
649 32. Nilsen TW. 2013. Toeprinting. Cold Spring Harb Protoc 2013:896–899.
650 33. Kerr CH, Wang QS, Moon KM, Keatings K, Allan DW, Foster LJ, Jan E. 2018.
651 IRES-dependent ribosome repositioning directs translation of a +1 overlapping
652 ORF that enhances viral infection. Nucleic Acids Res 46:11952–11967.
653 34. Au HHT, Elspass VM, Jan E. 2017. Functional Insights into the Adjacent Stem-
654 Loop in Honey Bee Dicistroviruses That Promotes Internal Ribosome Entry Site-
33 bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
655 Mediated Translation and Viral Infection. J Virol 92:e01725-17.
656 35. Kerr CH, Wang QS, Keatings K, Khong A, Allan D, Yip CK, Foster LJ, Jan E.
657 2015. The 5′ Untranslated Region of a Novel Infectious Molecular Clone of the
658 Dicistrovirus Cricket Paralysis Virus Modulates Infection. J Virol 89:5919–5934.
659 36. Domier LL, McCoppin NK, D’arcy CJ. 2000. Sequence requirements for
660 translation initiation of Rhopalosiphum padi virus ORF2. Virology 268:264–271.
661 37. Abaeva IS, Vicens Q, Bochler A, Soufari H, Simonetti A, Pestova T V., Hashem Y,
662 Hellen CUT. 2020. The Halastavi árva Virus Intergenic Region IRES Promotes
663 Translation by the Simplest Possible Initiation Mechanism. Cell Rep 33:108476.
664 38. Arhab Y, Bulakhov AG, Pestova T V., Hellen CUT. 2020. Dissemination of
665 Internal Ribosomal Entry Sites (IRES) Between Viruses by Horizontal Gene
666 Transfer. Viruses 12:612.
667
34 WANG_FIGURE 1
(A) SLVIII SLVII SLVI a u u uc a aa ua g u g a ug c a g c au g u u au a a u au au au au g c cg au au cg 1 au cg au 100 gc g c ua 5’ auuucugauuuuacgua gc uuccgga a c ua ccuuaaaaau a u SLIV u u u u Upstream region 189 a a a u a a u u a L 3 a u PKIII u a c g a u u u a u a a u a c c c g u a a u SLV c g c c g g g c a a g g u a g u u g g g c c u u u g 222 g a u u u c c c a u u a a c c c g g a a u c a c a g a Secondary Structure of the u u L 1.1B a u c g u 159 a a g c Ancient Northwest Terroritories cripavirus g a g c L 1.1A g u a a c u c g a u u u a PKII Internal Ribosome Entry Site a c c u c a u u g a g u u u a (aNCV IRES) a c g 127 u u a u a a a a a c u g c 251 u cg cg u au a au a au g gc u VLR a gc c a au 285 a ua c u c u u a g g a a u c gcuacgucaaca 3’ PKI Ala Thr
(B)
L1.1A L1.1B SL IV L 3 SL V
Type I
Type II
bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
WANG_FIGURE 2
A 80S Binding B [32P] CrPV IGR [32P] aNCV IGR N=3 1.0 1.0 1.0 N=3 N=3
0.5 0.5 0.5 Fraction bound Fraction Fraction bound Fraction bound Fraction 0.0 0.0 0.0 0 50 100 0 100 200 0 100 200 80S concentration (nM) Cold RNA Concentration (nM) Cold RNA Concentration (nM)
CrPV Competitor IGR RNA CrPV ΔPKI/II/III CrPV aNCV CrPV ΔPKI/II/III aNCV bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint WANG_FIGURE(which was not certified by peer review) is the author/funder. 3 All rights reserved. No reuse allowed without permission.
CrPV CrPV WT TM aNCV RNase I - + - + - +
nts 150
100 80
60 50 bioRxivWANG_FIGURE preprint doi: https://doi.org/10.1101/2021.01.28.428736 4 ; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
A gc aNCV IGR IRES u gc CrPV IGR IRES a a u cg a cg a a u ua u au mPKI #1 g ua a au a a u a a u a au u u a a ua gc mPKI #2 g c g u u u a a a g g a c a a u u gc mPKI gc comp u a g g a au u u a u u a ua g u c u a u g g a 6232 c u u a g 308 u a c c u g c aua u u u cc a a g a u a c c a u g 3´ g a a u c gcuacgucaaca acaauggccguu 3’ PKI PKI 298 +1 +1 Ala Thr Ala Thr +13 +20 6213 +13-14 +20-21 Major toeprints
B Toeprinting of IRES in RRL + cycloheximide
CrPV IGR IRES aNCV IGR IRES mPK1 mPK1 AGCT WT mPKI AGCT WT1 # #2 comp
AUC 289-91 {
CCU 6213-5 {
U308 U308 (+20)
C6233 CC6232-3 (+20-21)
C Toeprinting of purified 80S:IRES
CrPV IGR IRES aNCV IGR IRES WT WT mPKI WT WT mPKI #1mPKI #2 AGCT - + + 80S AGCT - + + + 80S
AUC { CCU 289-91 6213-5 {
C302 C302(+13) C6226 CA6226-7 (+13-14) bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
WANG_FIGURE 5
A B CrPV aNCV T7 IGR IRES WTmPKI WT Δ1-1 0 0 RLuc FLuc Bicistronic RNA CrPV aNCV mPKI mPKI PKI WTmPKI WT # 1 # 2 co mp kDa 75 FLuc Bicistronic 60 RNA 45
kDa 75 FLuc 35 60 RLuc 45 25
35 RLuc ns N=3 1.5 25 **
1 2 3 4 5 6 1.0
N=4 1.5 0.5 **
1.0 (normalized) FLuc/RLuc 0.0 T WT W mPKI 1-100 0.5 to WT CrPV) WT to
FLuc (normalized FLuc CrPV aNCV
0.0 T W WT mPKI comp mPKI_1 mPKI_2
CrPV aNCV bioRxiv preprint doi: https://doi.org/10.1101/2021.01.28.428736; this version posted January 29, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
WANG_FIGURE 6 aNCV IGR IRES A CrPV ORF1 CrPV ORF2 5’ IRES 2C2 B1A 3A 3C RdRp VP2 VP3 VP1 VPg AAA VPg VP4 n
1-100 UAA UAA
OR
CrPV-aNCV CrPV-aNCV (D1-100)
Chimera
B C
CrPV CrPV-ORF1-STOPCrPV-aNCVCrPV-aNCV (Δ1-100) CrPV CrPV-ORF1-STOPCrPV-aNCVCrPV-aNCV- (Δ1-100) Bicistronic kDa 100 75 RNA 60 45 35 kDa 100 ORF2 Polyprotein VP2 75 VP1/2 +VP4+VP3 25 60 20 45 VP3 +VP4(VP0) 15 35 VP2/3 VP1 25 20 20 1A
15 Tubulin
D E
CrPV CrPV-aNCV
Negative strand -100 kDa mock 2 4 6 8 10 2 4 6 8 10 h.p.i RT-PCR untransfectedCrPV Δ1 aNCV IGR 45 CrPV 35 VP2 rpS6 25 72 144 20 h.p.t h.p.t 1A
Tubulin