Expression of RNautophagy/DNautophagy-related genes is regulated under control of an innate immune receptor

ABSTRACT

Double-stranded RNA (dsRNA) is a molecular pattern uniquely produced in cells infected with various viruses as a product or byproduct of replication. Cells detect such molecules, which indicate non-self invasion, and induce diverse immune responses to eliminate them. The degradation of virus-derived molecules can also play a role in the removal of pathogens and suppression of their replication. RNautophagy and DNautophagy are cellular degradative pathways in which RNA and DNA are directly imported into a hydrolytic organelle, the lysosome. Two lysosomal membrane proteins, SIDT2 and LAMP2C, mediate nucleic acid uptake via this pathway. Here, we showed that the expression of both SIDT2 and LAMP2C is selectively upregulated during the intracellular detection of poly(I:C), a synthetic analog of dsRNA that mimics viral infection. The upregulation of these two gene products upon poly(I:C) introduction was transient and synchronized. We also observed that the induction of SIDT2 and LAMP2C expression by poly(I:C) was dependent on MDA5, a cytoplasmic innate immune receptor that directly recognizes poly(I:C) and induces various antiviral responses. Finally, we showed that lysosomes can target viral RNA for degradation via RNautophagy and may suppress viral replication. Our results revealed a novel degradative pathway in cells as a downstream component of the innate immune response and provided evidence suggesting that the degradation of viral nucleic acids via RNautophagy/DNautophagy contributes to the suppression of viral replication.

KEYWORDS:

• Lysosome
• RNautophagy
• DNautophagy
• PAMPs
• Double-stranded RNA
• PRRs
• immune response
• innate immunity
• virus infection
• virus replication

Introduction

Discrimination between self and non-self is a fundamental event in detecting and eliminating infectious agents through various immune responses. Double-stranded RNA (dsRNA) is a unique molecular pattern that is rarely, if ever seen in non-infected cells, but is produced intracellularly upon viral infection as a product or byproduct of virus replication, irrespective of the viral genome type [1]. Such molecular patterns that help cells detect non-self invasion are called pathogen-associated molecular patterns (PAMPs), and dsRNA is a typical PAMP that indicates viral infection [2]. Cells utilize PAMPs to detect non-self and initiate various immune responses. In addition, the elimination or degradation of molecules produced upon virus replication may also help host cells suppress their reproduction.

We previously identified an intracellular degradation pathway in which lysosomes, which are degradative organelles, directly import nucleic acids into their lumen in an ATP-dependent manner for degradation [3–6]. We termed this pathway RNautophagy for RNA degradation and DNautophagy for DNA degradation (RNautophagy/DNautophagy). In RNautophagy/DNautophagy, two lysosomal membrane proteins, SIDT2 and LAMP2C, mediate the direct uptake of nucleic acids into the lysosomes. To the best of our knowledge, RNautophagy/DNautophagy is the first reported pathway directly targeting nucleic acids for lysosomal degradation [7]. We speculate that RNautophagy/DNautophagy may function as part of the immune response against viral infections by degrading virus-derived nucleic acids; as such, the two RNautophagy/DNautophagy-related genes may be induced by the detection of PAMPs.

PAMPs, including dsRNAs, are detected by various sensor proteins known as pattern recognition receptors (PRRs) [2]. dsRNAs are detected intracellularly by three distinct PRRs: melanoma differentiation-associated gene 5 (MDA5), retinoic acid-inducible gene-I (RIG-I), and Toll-like receptor 3 (TLR3) [2]. All three PRRs reportedly recognize a synthetic dsRNA analog, poly(I:C), and induce divergent antiviral immune responses [8–11]. Therefore, treating cells or animals with poly(I:C) can mimic viral infections and provide an effective and generalized model for inducing antiviral immune responses [12]. Aberrant or irregular activation of antiviral immunity by poly(I:C) is also known to produce valid models for diseases other than infectious diseases, such as autoimmune and neurodevelopmental disorders [13–16].

In this study, we analysed whether two RNautophagy/DNautophagy-related genes, SIDT2 and LAMP2C, are regulated downstream of poly(I:C) induction and examined their relationship with PRRs that recognize dsRNA.

Results

RNautophagy/DNautophagy-related genes are transiently upregulated upon poly(I:C) transfection

First, we transfected HeLa cells with poly (I: C) and analysed the expression levels of SIDT2 and LAMP2C mRNA 3 h following after transfection. Interestingly, the expression levels of both SIDT2 and LAMP2C were upregulated upon poly(I:C) treatment compared to those in mock-transfected cells (Figure 1A,B). In contrast, no significant increase in SIDT2 or LAMP2C expression was observed in cells transfected with total mouse RNA, suggesting that the expression of these two genes was specifically induced by poly(I:C) (Figure 1A,B). We also observed that transfection of poly(A:U), a dsRNA harbouring a sequence other than poly(I:C), did not show as much of an effect on the levels of SIDT2 and LAMP2C as poly(I:C), although it tended to induce the expression of the two gene products to a lesser extent (Fig. S1A, B). In addition, expression levels of SIDT2 and LAMP2C were not affected in this time scale when poly(I:C) was added directly to the medium without lipofection reagent, even at 10 times the amount introduced when poly(I:C) was transfected (Figure 1C,D). These data suggest that the expression of SIDT2 and LAMP2C is upregulated upon the intracellular detection of poly(I:C).

Figure 1. Transient increase in expression of RNautophagy/DNautophagy-related genes by poly(I:C). A, B. relative levels of SIDT2 (A) or LAMP2C (B) mRNA in cells transfected with either poly(I:C) or mouse total RNA. C, D. relative levels of SIDT2 (C) or LAMP2C (D) mRNA in cells with poly(I:C) added directly into culture media. E, F. temporal change in expression levels of SIDT2 (E) or LAMP2C (F) mRNA in cells transfected with poly(I:C). (n = 3). ** p < 0.01, N.S., not significant.

Poly(I:C) is detected intracellularly by PRRs, and transiently induces the expression of various genes related to immune responses. To investigate the temporal changes in the effect of poly(I:C) transfection on the expression of SIDT2 and LAMP2C, we examined the expression levels of SIDT2 and LAMP2C in HeLa cells 3, 6, and 9 h after poly(I:C) transfection. The expression levels started to decrease after 6 h, and both SIDT2 and LAMP2C mRNA levels returned to levels similar to those in the poly(I:C)-untransfected cells after 9 h (Figure 1E,F). This indicates that the increase in SIDT2 and LAMP2C expression by poly(I:C) was transient.

Poly(i:c) selectively induces the expression of RNautophagy/DNautophagy-related genes

We observed that the induction of both SIDT2 and LAMP2C expression upon poly(I:C) transfection was dose-dependent in three distinct cell lines: HeLa, HEK-293 FT, and A549 (Figure 2A,B, Fig. S2A-D). Interestingly, the expression levels of LAMP2A and B, the two LAMP2 variants other than LAMP2C [17], were downregulated when LAMP2C expression was upregulated six- to seven-fold with a high dose of poly(I:C) (Figure 2C,D). As the expression levels of LAMP2 mRNA showed only a slight tendency to increase under these conditions (Figure 2E), the upregulation of LAMP2C upon poly(I:C) introduction may be regulated through alternative splicing machinery [17]. However, we cannot exclude the possibility that other mechanisms, such as the degradation and stability of mRNAs, are involved in alterations in the levels of LAMP2 isoforms. We also examined the mRNA expression levels of another major lysosomal membrane protein, LAMP1, finding no significant changes (Figure 2F). Taken together, these data indicate that the expression of SIDT2 and LAMP2C was selectively induced upon poly(I:C) transfection. We also observed a selective increase in SIDT2 and LAMP2C protein levels (Figure 2G–K).

Figure 2. Poly(i:c) selectively upregulates the expression of RNautophagy/DNautophagy-related genes. A, B. dose-dependency in the expression levels of SIDT2 (A) or LAMP2C (B) mRNA in cells transfected with poly(I:C). C–F. relative levels of LAMP2A (C), LAMP2B (D), all LAMP2 isoforms (E), or LAMP1 (F) mRNA in cells transfected with poly(I:C). G. immunoblotting of SIDT2, LAMP2C, all LAMP2 isoforms, and LAMP1 protein in cells transfected with poly(I:C). H–K. quantification of respective gene products shown in G. (n = 3). * p < 0.05, ** p < 0.01, N.S., not significant.

Induction of RNautophagy/DNautophagy-related gene expression by poly(I:C) is MDA5-dependent

MDA5, RIG-I, and TLR3 are known PRRs that recognize poly(I:C) and dsRNAs as part of the intracellular innate immune system. Because the expression of all three PRRs has been reported in HeLa cells, the involvement of either PRR in the upregulation of SIDT2 or LAMP2C upon poly(I:C) introduction is plausible. To elucidate the mechanism underlying the induction of SIDT2 and LAMP2C expression upon poly(I:C) introduction, we analysed the effect of knocking down each PRR on the poly(I:C)-dependent induction of the two genes. We introduced siRNAs targeting MDA5, RIG-I, or TLR3, followed by poly(I:C) transfection, and analysed the induction levels of SIDT2 and LAMP2C in each group. The levels of both SIDT2 and LAMP2C after poly(I:C) introduction were decreased by transfection with all three siRNAs (Figure 3A,B). However, the expression levels of PRRs other than the target gene were also downregulated in some samples; the expression of MDA5 and TLR3 decreased in cells transfected with siRNAs targeting any of the three PRRs (Fig. S3A–C). This is consistent with previous reports showing that knockdown of one PRR results in downregulation of the other PRRs, including MDA5, RIG-I, and TLR3 [18,19].

Figure 3. MDA5 mediates the upregulation of RNautophagy/DNautophagy-related genes expression by poly(I:C). A, B. relative levels of SIDT2 (A) or LAMP2C (B) mRNA in cells transfected with respective siRNA, followed by mock- or poly(I:C)-transfection. C, D. relative levels of SIDT2 (C) or LAMP2C (D) mRNA in cells transfected with respective expression vectors, followed by mock- or poly(I:C)-transfection. (n = 3). ** p < 0.01, N.S., not significant.

Next, we analysed the effect of MDA5, RIG-I, or TLR3 overexpression on the upregulation of SIDT2 and LAMP2C expression upon poly(I:C) transfection. We first transfected HeLa cells with the expression vectors for each PRR and confirmed the specific overexpression of each PRR (Fig. S3D). Cells overexpressing each PRR were transfected with poly(I:C) and the expression levels of SIDT2 and LAMP2C were analysed. The induction levels of both SIDT2 and LAMP2C upon poly(I:C) transfection significantly increased only in cells overexpressing MDA5 (Figure 3C,D), indicating that the induction of these two genes is at least dependent on MDA5.

Lysosomes directly take up and degrade viral RNA in an ATP-dependent manner

Taken together, our data show that the expression of SIDT2 and LAMP2C, both of which are involved in the lysosomal RNA/DNA degradation pathway RNautophagy/DNautophagy, is selectively induced by MDA5, a PRR involved in innate immunity, in response to introducing poly(I:C) into cells. SIDT2 reportedly functions as part of the immune response by releasing viral RNA from late endosomes into the cytoplasm, enabling cytosolic PRRs to recognize viral infections [20]. However, LAMP2C was not involved in this mechanism. We speculated that the uptake and degradation of viral RNA by lysosomes via RNautophagy may also be involved in the response to viral infection. However, direct uptake and degradation of viral RNA by lysosomes via RNautophagy has not yet been determined. We prepared an RNA virus, the Japanese encephalitis virus (JEV), and examined whether virus-derived RNA could be degraded via RNautophagy. First, we observed that viral RNA production was induced in Sidt2 knockout cells upon JEV infection compared to that in wild-type cells (Figure 4A,B Fig. S4), suggesting that the viral RNA production was suppressed by SIDT2. In addition, the production of viral protein and viral titre increased in Sidt2 KO cells upon JEV infection (Figure 4C–E, Fig. S5). Because SIDT2 plays a pivotal role in the mechanism of RNautophagy/DNautophagy [21], we chose Sidt2 for the gene knockout experiment. To determine whether viral RNA can be directly imported into lysosomes and degraded, we performed an in vitro reconstruction assay using isolated lysosomes. Lysosomes were isolated from mouse brain and incubated with RNA derived from JEV-infected cells in the presence or absence of ATP at 37°C. Following incubation, the lysosomes were pelleted by centrifugation and RNA was purified from the solution outside the lysosomes and from the lysosome pellet. The amount of viral RNA in each sample was quantified by quantitative PCR (qPCR). The total amount of viral RNA in the samples was quantified to measure the degradation of viral RNA by lysosomes. Viral RNA almost completely disappeared from the extralysosomal solution and relocated to the lysosomes in the presence of ATP (Figure 4F). In contrast, most viral RNA remained in the external solution in the absence of ATP (Figure 4F). In addition, viral RNA was only directly degraded by lysosomes in the presence of ATP (Figure 4G). Taken together, these results strongly suggest that viral RNA can be a target for RNautophagy, and that RNautophagy can function as an antiviral mechanism in cells.

Figure 4. Viral RNA can be targeted for degradation via RNautophagy. A, B. relative levels of extracellular (A) and intracellular (B) viral RNA produced in WT or Sidt2 KO cells infected with JEV at MOI = 1. C. immunoblotting of extracellular viral protein produced by WT or Sidt2 KO cells infected with JEV at MOI = 1. D. a representative image of plaque assays using cultured media derived from WT or Sidt2 KO cells infected with JEV at MOI = 1. Plaques are surrounded by pen strokes. E. virus titre of cultured media derived from WT or Sidt2 KO cells infected with JEV at MOI = 1. F. relative levels of viral RNA in extralysosomal solution or lysosomal pellet following the incubation with isolated lysosomes in the absence or presence of ATP. G. relative levels of total viral RNA following the incubation with isolated lysosomes in the absence or presence of ATP. (n = 3). ** p < 0.01.

Discussion

In this study, we showed that the expression of two distinct RNautophagy/DNautophagy-related genes, SIDT2 and LAMP2C, is selectively upregulated by the intracellular detection of poly(I:C), and that viral RNA can be a target for degradation by RNautophagy (Figure 5). We also showed that MDA5, a PRR that recognizes poly(I:C) in the cytoplasm, was involved in the induction of both SIDT2 and LAMP2C (Figure 5). However, because the contribution of MDA5, RIG-I, and TLR3 to the initiation of the immune response by poly(I:C) recognition varies by cell type, as well as by other experimental conditions, such as the length of poly(I:C) used [22–24], PRRs other than MDA5 May also plausibly participate in the induction of these two genes. For instance, although this was observed to a lesser extent than poly(I:C) and was not significant in the present conditions, transfection with poly(A:U) also slightly increased the levels of both SIDT2 and LAMP2C. Considering that poly(A:U) is a molecular pattern that stimulates immune responses mainly through TLR3 and TLR7 [25], the upregulation of RNautophagy/DNautophagy-related genes through endolysosomal TLRs may also be possible, which warrants further investigation (Figure 5). The involvement of PRRs that recognize PAMPs other than dsRNA, such as TLR7 and TLR8 for single-stranded RNA, TLR9 and absent in melanoma-2 (AIM2)-like receptors (ALRs) for double-stranded DNA, or even those that detect PAMPs other than nucleic acids, is also an intriguing issue. As shown in Figure 3A,B the expression levels of SIDT2 and LAMP2C, even in the absence of poly(I:C), varied after siRNA transfection. This may be a consequence of sequence-dependent induction of innate immune responses by siRNAs. Endosomal detection of siRNAs has been reported to induce immune responses dependent on TLR7 and TLR8, and the levels of this induction vary according to the siRNA sequences [26]. We did not examine the levels of RNautophagy activity upon poly(I:C) transfection in this study, but considering that the overexpression of SIDT2 and LAMP2C can induce RNautophagy activity both in vitro and at the cellular levels [3–6,21,27], the upregulation of RNautophagy activity is also plausible.

Figure 5. A schematic of our current model. Pattern recognition receptors (PRRs), such as MDA5, recognize pathogen-associated molecular patterns (PAMPs) such as dsRnas and upregulate RNautophagy/DNautophagy-related genes, to activate RNautophagy/DNautophagy. RNautophagy/DNautophagy contributes to clearance of viral nucleic acids, thereby suppressing virus replication. RNautophagy/DNautophagy may also contribute to the detection of PAMPs via endolysosomal TLRs by taking up viral nucleic acids into their recognition sites, which leads to induction of various anti-viral responses that may also include the activation of RNautophagy/DNautophagy as a positive feedback loop.

A previous study by Nguyen et al. reported a mechanism by which SIDT2 mediates the export of poly(I:C) from late endosomes to the cytoplasm, thereby enabling cytosolic PRRs to induce cytokine production [20]. Here, we revealed that a cytoplasmic PRR, MDA5, induces the expression of RNautophagy/DNautophagy-related genes, including SIDT2. Endocytosis is a major pathway for viral entry into host cells, and the export of viral genomic nucleic acids to the cytoplasm is an essential step for the replication of many viruses [28]. Viruses encode various genes that hijack the endosomes of their host, enabling the viral genome or virus particles to escape from the endocytosed compartment into the cytosol [28]. Through such pathways called ‘endosomal escape’, viruses can release their genome to the site of their replication before the endosomes mature into degradative compartments, lysosomes. If the function of SIDT2 in innate immunity is only to export viral RNA to the cytoplasm for recognition of viral infection, such a mechanism may benefit host cells at the beginning of infection. However, it can also help viruses efficiently escape from degradative compartments. A model in which SIDT2 exports viral RNA to the cytosol for recognition by PRRs may not be sufficient to thoroughly describe the function of SIDT2 in innate immunity. Our data show that SIDT2 expression is upregulated together with LAMP2C upon poly(I:C) introduction, and that viral RNA can be a target for degradation via RNautophagy, suggesting that lysosomal import and degradation of viral nucleic acids may also be important functions of SIDT2 and LAMP2C in innate immunity. It may be possible that the export of viral RNA from endosomes occurs at an earlier step of infection to detect viral invasion, and then the uptake of viral nucleic acids into degradative lysosomes via RNautophagy/DNautophagy occurs as a part of the process to suppress viral replication.

In addition to cytosolic PRRs such as MDA5 and RIG-I, TLRs are composed of membrane proteins that localize to the plasma or endosomal/lysosomal membrane [2]. Interestingly, all these endosomal/lysosomal TLRs recognize nucleic acids, but none recognize other molecular patterns or detect PAMPs in the endosomal/lysosomal lumen [2]. In addition to the degradation of viral nucleic acids, RNautophagy/DNautophagy can also play a role in the recognition of viral infections via endosomal/lysosomal TLRs by delivering viral nucleic acids into endosomes/lysosomes (Figure 5). If the expression of these two genes is also induced downstream of TLRs, a positive feedback loop could be formed between RNautophagy/DNautophagy and endosomal/lysosomal TLR-dependent immune responses (Figure 5). In contrast, the aberrant recognition of self-derived nucleic acids by endosomal/lysosomal TLRs has been implicated in autoimmune diseases [29], and the maternal induction of immune responses via poly(I:C) administration is known to produce models for neurodevelopmental disorders in animals [16]. The aberrant upregulation of RNautophagy/DNautophagy-related genes may also be implicated in the pathogenesis of autoimmune diseases and neurodevelopmental disorders.

Materials and methods

siRNAs

siRNAs used in this study were purchased from JBioS (Japan). The targeting sequences of the siRNAs were as follows: For EGFP: 5′-CAGCACGACUUCUUCAAGUCC-3′, for human MDA5:5′-CGAAUGAUAGAUGCGUAUACU-3′, for human RIG-I: 5′-CGGAUUAGCGACAAAUUUAAA-3′ and for human TLR3: 5′- CUCACUAUGCUCGAUCUUUCC-3′.

Plasmids

All plasmid vectors used in this study were generated by subcloning PCR products from the cDNA of the genes of interest into the pCI-neomammalian expression vector (Promega, E1841). The EGFP expression vector was prepared as previously described. For the MDA5, RIG-I, and TLR3 expression vectors, the cDNA of each gene was purchased from Open Biosystems and amplified by PCR using KOD FX Neo (ToYoBo). The PCR products and pCI-neomammalian expression vector were digested with XhoI and NotI restriction enzymes (TaKaRa) and subcloned using the DNA Ligation Kit Mighty Mix (TaKaRa). Sequences were confirmed for all the prepared vectors. Because two point mutations were observed in MDA5, revertant mutations were introduced using Pfu Turbo DNA Polymerase (Agilent Technologies) and the sequence was confirmed again. The clone ID for each gene was as follows: MDA5: BC111750, RIG-I: BC136610 and TLR3: BC096333. The following primer sequences were used for PCR: For MDA5: 5′-AAAACTCGAGCCGCCACCATGTCGAATGGGTATTCCACAGAC-3′ and 5′-AAAAGCGGCCGCCTAATCCTCATCACTAAATAAACAGC-3′, for RIG-I: 5′-AAAACTCGAGCCGCCACCATGACCACCGAGCAGCGACGC-3′ and 5′-AAAAGCGGCCGCTCATTTGGACATTTCTGCTGGATCA-3′, and for TLR3: 5′-AAAACTCGAGCCGCCACCATGAGACAGACTTTGCCTTGTATCT-3′ and 5′-AAAAGCGGCCGCTTAATGTACAGAGTTTTTGGATCCAAG-3′. The sequences of primers used for mutagenesis were as follows: pair 1: 5′-CAGGAGTTATCGAACATGAGACAGTTAATGA-3′ and 5′-TCATTAACTGTCTCATGTTCGATAACTCCTG-3′, and pair 2: 5′-GTAAGAGAAAACAAAGCACTGGAAAAGAAGT-3′ and 5′-ACTTCTTTTGCAGTGCTTTGTTTTCTCT TAC-3′.

Antibodies

The primary antibodies used in this study are as follows: goat polyclonal anti-SIDT2 (sc -54,151, Santa Cruz), mouse monoclonal anti-GFP (sc-9996, Santa Cruz), rabbit polyclonal anti-MDA5 (SAB3500356, Sigma Aldrich), mouse polyclonal anti-DDX58 (SAB1400461, Sigma Aldrich), rabbit monoclonal anti- Toll-like Receptor 3 (6961S, Cell Signaling), mouse monoclonal anti-β-actin (A5441, Sigma Aldrich) and rabbit polyclonal anti-JEV envelope protein (GTX125867, GeneTex). Rabbit polyclonal anti-LAMP2C antibodies were prepared as described previously [3]. The secondary antibodies used were horseradish peroxidase-conjugated rabbit anti-goat IgG (H+L) (305-036-003, Jackson ImmunoResearch), HRP-conjugated goat anti-mouse IgG (H+L) (31430, Thermo Scientific), and HRP-conjugated goat anti-rabbit IgG (H+L) (31460, Thermo Scientific).

Cell culture

HeLa, HEK-293 FT, A549, wild-type and Sidt2 knockout mouse embryonic fibroblasts, and BHK-21 cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco) supplemented with 10% foetal bovine serum (Sigma Aldrich). Sidt2 knockout cells were generated using the CRISPR/Cas9 System. The sgRNA sequence was designed using the CRISPR DESIGN programme (http://crispr.mit.edu), targeting a site in exon 13 of Sidt2 (5′-ACTGGACTCCATGAGCTCCG-3′). The DNA sequence corresponding to the sgRNA sequence was cloned into the p×330plasmid (Addgene) and transfected into wild-type mouse embryonic fibroblasts [5]. Following plasmid transfection, the cells were cloned and the expression levels of Sidt2 were confirmed by quantitative RT-PCR.

Transfection

For transfection of poly(I:C) or poly(A:U) into cells, unless otherwise preceded by prior experimental manipulations, 7.0 × 10⁴ cells/well of HeLa, HEK-293 FT, or A549 cells were seeded in 24-well plates, and poly(I:C) (P9582, Sigma Aldrich) was transfected the following day using Lipofectamine 2000 Reagent (Invitrogen) according to the manufacturer’s instructions and incubated for the indicated time. Unless otherwise stated, the concentration of poly(I:C) introduced was 0.5 µg/well. For the experiment in Figure 2G–K, 1.0 × 10⁶ cells/well of A549 cells were seeded in 6-well plates, and 7.5 µg/well of poly(I:C) was transfected the following day and incubated for 14 h. For siRNA transfection, HeLa at 2.0 × 10⁴ cells/well was seeded in 24-well plates, and 0.06 µg/well of siRNA was transfected by reverse transfection using Lipofectamine RNAiMAX Reagent (Invitrogen). The cells were then incubated for 2 d prior to transfection with poly(I:C). For transfection of the plasmid vector, 2.0 × 10⁴ cells/well of HeLa cells were seeded in 24-well plates, and 0.1 µg/well of plasmid was transfected the following day using Lipofectamine LTX with Plus Reagent (Invitrogen) and incubated for one day prior to poly(I:C) transfection.

Quantitative RT-PCR

For quantitative RT-PCR, RNA was purified from samples using TRIzol Reagent (Ambion), and cDNA was synthesized using the QuantiTech Reverse Transcription Kit (QIAGEN), followed by real-time quantitative PCR using SYBR Premix Ex Taq II (Tli RNaseH Plus) (TaKaRa) and the Mx3000P real-time quantitative PCR system (Stratagene) or CFX96 Real-Time PCR Detection System (Bio-Rad). The expression levels of all endogenous gene products were quantified relative to GAPDH, and the levels of viral RNA were quantified by reference to the levels of the same viral RNA in samples obtained from viral solutions of known titres. The sequences of the primers used were as follows: For human GAPDH: 5′-CAATGACCCCTTCATTGACC-3′ and 5′-GACAAGCTTCCCGTTCTCAG-3′, for human SIDT2: 5′-TCTTCCCTCTGACCTGTGCT-3′ and 5′-ATCAGTCTGGGAGCTGATGG-3′, for human LAMP2C: 5′-GTATTCTACAGCTGAAGAATGTTCTG-3′ and 5′-ACACCCACTGCAACAGGAAT-3′, for human LAMP2A: 5′-GGGTTCAGCCTTTCAATGTG-3′ and 5′-CAGCATGATGGTGCTTGAGA-3′, for human LAMP2B: 5′-GGGTTCAGCCTTTCAATGTG-3′ and 5′-CCTGAAAGACCAGCACCAAC-3′, for human LAMP2 (all variants): 5′-AATGCCACTTGCCTTTATGC-3′ and 5′-CAGTGCCATGGTCTGAAATG-3′, for human LAMP1: 5′-GTGTTAGTGGCACCCAGGTC-3′ and 5′-GGAAGGCCTGTCTTGTTCAC-3′, for human MDA5: 5′-CCAACTGCTGAACCTCCTTC-3′ and 5′-GCAATCCGGTTTCTGTCTTC-3′, for human RIG-I: 5′-ATATCCGGAAGACCCTGGAC-3′ and 5′-GGCCCTTGTTGTTTTTCTCA-3′, for human TLR3: 5′-TGTTTTCACGCAATTGGAAG-3′ and 5′-CCGAATGC TTGTGTTTGCTA-3′, and for JEV NS3: 5′-AGAGCACCAAGGGAATGAAATAGT-3′ and 5′-AATAGGTTGTAGTTGGGCACTCTG-3′. The sequences of specific primers for detecting LAMP2C, LAMP2A, LAMP2B, and JEV NS3 have been reported previously [30,31]. Sequences of the primers for detection of other gene products were generated using Primer3Plus (http://www.bioinformatics.nl/cgi-bin/primer3plus/primer3plus.cgi/) and specificity was confirmed by BLAST searches.

Immunoblotting

Samples in SDS-PAGE sample buffer consisting of 10 mM Tris-HCl (pH 7.8), 3% SDS, 5% glycerol, 0.02% bromophenol blue, and 2% 2-mercaptoethanol were separated by SDS-PAGE and transferred onto PVDF membranes (1620177, Bio-Rad). The membranes were blocked with blocking buffer containing 3% bovine serum albumin and 0.01% sodium azide in PBS, and then treated with primary antibodies diluted in the same blocking buffer overnight at 4°C. The following day, the membranes were washed with PBST containing 0.1% Tween 20 in PBS and probed with respective HRP-linked secondary antibodies diluted in PBST. The membranes were then treated with SuperSignal West Pico extended-duration substrate (Pierce 34,080), SuperSignal West Dura extended-duration substrate (Pierce 34,076), or SuperSignal West Femto extended-duration substrate (Pierce 34,095) to visualize the signals. Signals were detected using the FluorChem Chemiluminescence Imaging System (Alpha Innotech).

Inoculation of Japanese encephalitis virus in wild-type or Sidt2 knockout cells

For comparisons of replication levels of RNA, protein, and JEV, 5.0 × 10⁵ cells/well of wild-type or Sidt2 knockout mouse embryonic fibroblasts were seeded on 6-well plates. The cells were then inoculated with JEV (JaGAr01 strain produced in BHK-21 cells) at a multiplicity of infection (MOI) of 1 the following day, and the culture media were collected at 3, 6, 12, 24, 48, and 72 h post-inoculation. The cultures were centrifuged at 1, 760 × g at room temperature for 5 min to remove debris, and the supernatants were subjected to quantitative RT-PCR, immunoblotting, or plaque assays. Three days post-infection, RNA was collected from the infected cells, and the levels of intracellular viral RNA were analysed by quantitative RT-PCR.

Determination of viral titer by plaque assay

A plaque assay was performed to determine viral vitals. Briefly, BHK-21 cells were seeded at 1.0 × 10⁶ cells/well in 6-well plates; the following day, they were treated with culture media from wild-type or Sidt2 knockout cells inoculated with JEV. Viral samples were serially diluted 10-fold and inoculated into the cells in each well for 1 h. The culture media were removed and replaced with a medium containing 3% agarose. The cells were incubated for 3 d at 37°C, viable cells were stained with neutral red, and neutral red-negative plaques were counted to calculate the number of plaque forming units (PFUs).

Direct uptake and degradation of viral RNA using isolated lysosomes

Assays using isolated lysosomes were performed as previously described with slight modifications [3–6,21,27,32–34]. Briefly, lysosomes were freshly isolated from the whole brains of 11-week-old mice using the Lysosome Enrichment Kit for Tissues and Culture Cells (89839; Thermo Fisher Scientific). The lysosomes isolated using this procedure were confirmed to contain minimal amounts of other organelle contaminants [3,5,21]. The isolated lysosomes (~25 µg/assay of protein) were then incubated at 0.5 µg/assay with purified total RNA derived from wild-type mouse embryonic fibroblasts infected with JEV in 0.3% sucrose containing 10 mM MOPS buffer, pH 7.0, with or without 10 mM ATP at 37°C for 10 min. For quantification of viral RNA in the lysosome pellet and solution outside the lysosomes, samples were centrifuged at 4°C 17,700 × g for 2 min, and RNA from each fraction was purified separately using TRIzol Reagent. RNA was purified from the whole samples to detect viral RNA degradation. Purified RNA from each sample was subjected to quantitative RT-PCR to analyse the viral RNA levels. All animal experiments were performed in strict accordance with the guidelines of the National Institute of Neuroscience, National Center of Neurology and Psychiatry, Japan, and approved by the Animal Investigation Committee of the Institute.

Statistical analyses

Statistical analysis was performed using Student’s t-test for comparisons between two groups and Tukey’s test for comparisons between three or more groups. All data in the graphs are expressed as mean ± standard error.

Acknowledgments

We thank Yoshiyuki Ohshima, Yoshiaki Furuya, Natsumi Takeyama, Aki Nagao, and Shizuka Hayashi for technical assistance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data generated or analysed in this study are available from the corresponding authors on reasonable request.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15476286.2023.2291610

Additional information

Funding

This study was supported by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (14J08223, 17J10610, 21K15367, and 23K14453 to Y.F.; 16H05146, 16H01211, and 19H05710 to T.K.), ACT-X from Japan Science and Technology Agency (JPMJAX222H to Y.F.), and research grants from the Takeda Science Foundation (to Y.F.).