Structural Perspective of the Double-Stranded RNA Transport Mechanism by SID-1 Family Proteins

Abstract

The systemic RNA interference defective 1 (SID-1) family proteins are putative double-stranded RNA (dsRNA) transporters. Two mammalian homologs, SIDT1 and SIDT2 have been linked to many functions such as innate immune responses, microRNA uptake and lysosomal degradation of RNA/DNA whereas Caenorhabditis elegans SID-1 is essential for systemic RNA interference. However, dsRNA uptake mechanism is largely unknown. In this review, we discuss our current understanding of the molecular functions of SID-1 family proteins at a structure level, which highlights recent structural studies.

1. INTRODUCTION

RNA uptake to cells is essential process for extracellular RNAs to exert their function such as virus-host interactions, cell-cell communications, and potent RNA therapeutics.^1,2) However, RNA uptake mechanism is largely unknown. A Caenorhabditis elegans (C. elegans) protein, systemic RNA interference defective protein 1 (SID-1), was initially suggested to serve as a double-stranded RNA (dsRNA) channel or transporter, essential for systemic RNA interference (RNAi) in which the effect of RNAi spreads from the introduced cells or tissues throughout the organism and its progeny.^3–5) SID-1 proteins are evolutionarily conserved although systemic RNAi phenomena are not apparent in mammals, which implies the functional divergence. In mammals, two SID-1 homologues SIDT1 and SIDT2 have the ability to transport dsRNA.^6–10) SIDT1 is required for cellular uptake of cholesterol conjugated small interfering RNAs (siRNAs) in liver HepG2 cells.¹¹⁾ In gastric pit cells of the stomach which is the primary site for dietary microRNA (miRNA) absorption, SIDT1 is required for the absorption of dietary miRNAs.¹⁰⁾ The uptake of miRNAs or double-stranded miRNA mimics by stomach cells is increased in an acidic condition, which is not affected in sidt1 knockout cells.⁹⁾ The SIDT1/2-mediated cellular dsRNA uptake in HEK293T cells is controversial.¹²⁾ SIDT2 is localized on endolysosomal membranes and is involved in the release of internalized dsRNA from endolysosome to cytoplasm.⁸⁾ Loss of SIDT2 impairs endosomal escape of internalized polyinosinic-polycytidylic acid (poly(I:C), a synthetic analog of viral dsRNA, resulting in the accumulation of poly(I:C) in endolysosome. SIDT2 is involved in Retinoic acid-inducible gene-I (RIG-I) like receptor mediated interferon production but not in toll-like receptor 3 mediated interferon production.⁹⁾ SIDT1 is also partly involved in endosomal escape of internalized dsRNA, but it contributes less than SIDT2.⁷⁾

Recently, the cryo-electron microscopy analyses have revealed structures of human SIDT1, SIDT2 and C. elegans SID-1, which have provided new insight into the molecular function of SID-1 family proteins^13–19) (Table 1). In this review, the available structural information of SID-1 family proteins will be summarized, and conformational change associated with cholesterol binding and the dsRNA recognition will be addressed in addition to the implications in the dsRNA transport mechanism.

Table 1. Structural Studies of SID-1 Family Proteins

Protein	Species	pH	Conformation of TM region	Binding lipid	PDB ID	Ref.
SIDT1	human	6.0	Closed	Cholesterol	8KCW	16)
SIDT1	human	5.0	Open		8KCX	16)
SIDT1	human	7.5	Open	Phosphatidic acid	8JUL	14)
SIDT1	human	7.5	Open		8K13	15)
SIDT1	human	7.5	Open	Cholesterol	8WOR	17)
SIDT1	human	5.5	Closed	Cholesterol	8WOT	17)
SIDT2	human	7.5	Closed		8K10	17)
SIDT2	human	7.4	Closed		7Y63	13)
SIDT2	human	5.5	Closed		7Y69	13)
SID-1	C. elegans	7.5	Closed		8HIP	18)
SID-1	C. elegans	7.5	Closed		8HKE	18)
SID-1	C. elegans	7.4	Closed	Cholesterol, phosphatidyl choline	8XBS	19)
SID-1	C. elegans	7.4	Closed	Cholesterol, phosphatidyl choline	8XC1	19)

Structural studies of SID-1 family proteins are summarized. The conformations of TM region in cholesterol-bound form and cholesterol-unbound form of SIDT1 are referred to as Closed and Open, respectively.

2. OVERALL STRUCTURE OF SID-1 FAMILY PROTEIN

The structure of human SIDT2 has been reported as the first structure of SID-1 family proteins.¹³⁾ Subsequently, the structures of human SIDT1 and SIDT2 bound or unbound to lipid have been reported.^14–17) These structures revealed a common domain architecture of SID-1 family proteins comprising the extracellular/luminal two β-sandwich domains (ECD1 and ECD2) and a unique transmembrane domain (TMD) with eleven transmembrane helices (Fig. 1A).

Fig. 1. Overall Structure of SID-1 Family Proteins

(A) Topology diagram of SID-1 family proteins. Disulfide bonds are shown with orange lines. SIDT1 lacks β0 strand in ECD1. C. elegans SID-1 lacks the disulfide bond that links ECD1 to ECD2 whereas SID-1 contains extra helix in TM2-TM3 loop. (B) Overall structure human SIDT2 (PDB: 7Y63). Glycans (gray) and disulfide bonds (magenta) are shown with a stick model whereas zinc ions are shown with red spheres. (C) Zinc ion is coordinated with the conserved three histidine residues located at TM4 or TM11.

ECD1 and ECD2 adopt a jelly-roll fold comprising two four-stranded β-sheets and ECD1 additionally contains an extra β-hairpin which contacts with ECD2. The inter-domain disulfide bond and the intra-domain disulfide bond in ECD2 are formed, of which cysteine residues are conserved in vertebrate SID-1 family proteins, suggesting the evolutionary conserved contribution of disulfide bonds for the stabilization of inter-domain packing and structural rigidity (Fig. 1B).

TMD consists of eleven TM helices. Except for TM2, TMs are arranged in a clockwise from the cytoplasmic view. The disulfide bonds link extracellular TM2-TM3 loop to TM4-TM5 loop or TM10-TM11 loop. The short helix located in TM10-TM11 loop packs with ECD2 along with TM2-TM3 loop. The long cytoplasmic TM1-TM2 loop is largely disordered (Fig. 1B). On the extracellular side of the TM helices, a zinc ion is tetrahedrally coordinated with conserved three histidine (His) residues (Fig. 1C). The previous deep sequence similarity search suggested that these residues are conserved in a diverse superfamily of putative metal-dependent transmembrane hydrolases including 7 TM protein alkaline ceramidases (ACERs) and adiponectin receptors (AdipoRs).²⁰⁾ Although SIDT1/2 contains 11 TMs which is more than 7 TMs observed in the structures of ACERs²¹⁾ and AdipoRs,^22,23) 7 TMs except for TM1, TM2, TM7, and TM8 of SIDT1/2 are fitted to TMDs of ACER3 and ADIPOR2 with a small root mean square deviation. The zinc ion coordinating residues are also structurally conserved. The zinc ion binding sites of ACER3 and ADIPOR2 face the cytoplasm whereas that of SIDT2 faces the extracellular/luminal side.

Additionally, the structure of Caenorhabditis elegans SID-1 has been determined, which shows a similar domain architecture to SIDT1/2 although there are some conformational shifts between SID-1 and SIDT1/2.^18,19) This observation further supports the notion that the domain architecture of SID-1 family protein is evolutionally conserved.

3. DIMER INTERFACE

SIDT1 and SIDT2 form a side-by-side homodimer along the C2 symmetry axis, which is mediated by extensive interactions (Fig. 1B). The ECD1s of two protomers pack with each other via facing β-sheets while the ECD2s of two protomers pack with each other via loop region. SIDT2 has an additional N-terminal β-strand involved in interactions at the dimer interface, while the corresponding region in SIDT1 adopts a flexible loop. In the TM region of SIDT2, TM2 of one protomer packs with TM6 and the extracellular/luminal end of TM5 of the other protomer, incorporating hydrophobic interactions and hydrogen bonds. The initially reported structure of SIDT1 has a different dimer interface in TMD, in which TM5 packs with TM2 of the other protomer and TM6 is pushed out of the interface. Following reported structure of SIDT1 bound to cholesterol possesses a similar TMD conformation to SIDT2, which suggests that cholesterol induces the conformational change of SIDT1. Cholesterol serves as allosteric regulator for SIDT1, which will be discussed later. C. elegans SID-1 share the dimer interface with SIDT2.

4. CHOLESTEROL BINDING TO SIDT1 INDUCES A CONFORMATIONAL CHANGE OF TMD

SID-1 family proteins share sequence similarity with C. elegans CHUP-1, a protein involved in dietary cholesterol uptake.²⁴⁾ CHUP-1 is also suggested to mediate protective immune response by linking cholesterol metabolism and the immune response.²⁵⁾ Additionally, CRAC motif characterized by the sequence (L/V)-X₁₋₅-(Y)-X₁₋₅-(K/R),²⁶⁾ often observed in transmembrane proteins, is found in ECD1 and TM7 of SIDT1/SIDT2 and binds to cholesterol.¹¹⁾ Structures of SIDT1 and SID-1 containing cholesterol have been determined,^16,17,19) whereas the structure of SIDT2 containing cholesterol has not been available. However, most of those structural studies lack experimental evidence, and binding has been discussed only based on the EM density maps. Additionally, those structures were obtained from protein sample to which cholesteryl hemisuccinate was added during purification process. The only exception is cholesterol-bound structure revealed by Hirano et al.,¹⁶⁾ in which mass spectrometry analysis identified cholesterol in purified hSIDT1. Since they did not add cholesterol during purification, endogenous cholesterol was co-purified with SIDT1. Although it is not clear why SIDT1 sample is co-purified with cholesterol or not depending on the studies, the protein purification procedures including solubilizing detergent for protein extraction from cell membranes and/or expression cells are different, which can affect the lipid binding. Intriguingly, cholesterol-bound structure is obtained in SIDT1 at pH 6.0 while two conformations, cholesterol-bound and unbound structures are obtained at pH 5.0, suggesting that cholesterol binding to SIDT1 is pH dependent.

The two cholesterol molecules bind to a small concave formed by TM5, TM6, and TM7 at the intracellular side, suggesting that the cholesterols serve as molecular glue that stabilizes the conformation of TM5, TM6, and TM7 (Fig. 2A). Part of the previously identified CRAC motif in TM7 contacts with cholesterol, but the rest was disordered. Instead, another newly identified CRAC motif in TM5 interacts with cholesterol molecules more extensively than TM7.

Fig. 2. Conformational Change of SIDT1 upon Cholesterol Binding

(A) The structure of cholesterol-bound SIDT1 (PDB 8KCW). Two cholesterol molecules are docked into the cavity formed by TM5, TM6 and TM7. CRAC motifs of TMD are enclosed by green boxes. (B) TMD of cholesterol-bound structure (light-blue) is overlayed on TMD of cholesterol-unbound form (gray) from cytoplasmic view. Cholesterol binding induces the large conformational changes in TM5 (pink), TM6 (red) and TM7 (magenta) and disordered TM8 in cholesterol-unbound form is observed in cholesterol-bound form. (C) as (B), but side-view. For clarity, TM5, TM6, TM7and TM8 are shown. (D) PA-bound structure of SIDT1 (PDB 8JUL). PA (magenta) is captured at the catalytic site and Glu555 interacts with the carbonyl group of an acyl chain. (E) Overlay of the catalytic residues of SID-1 family proteins. The conserved aspartic residues at TM5 is shown in cyan (cholesterol-bound form of SIDT1), green (cholesterol-unbound form of SIDT1), pink (SIDT2) or orange (C. elegans SID-1).

Upon cholesterol binding, the TM region of SIDT1 undergoes a large conformational change while the extracellular region remains in the apo form structure (Fig. 2B). In the cholesterol-bound form, TM5 which interacts with TM2 in the cholesterol-unbound form rotates about 25° and its cytoplasmic side greatly moves from the dimer interface to inside the membrane spanning region (approx. 14 Å shift at C-terminal end). TM6 moves to the dimer interface (approx. 16 Å shift at N-terminal end) and forms a longer helix on the N-terminal region to interact with TM2 of the opposing protomer (Figs. 2B, 2C). These conformational changes of the transmembrane region are primarily caused by helix rotation, resulting in shifts of the cytoplasmic side with less changes in the extracellular side. The exception is TM7, where the extracellular side is shifted (approx. 6 Å shift at N-terminal end).

Cholesterol uptake in HEK293T cells is enhanced when SIDT1 is overexpressed while the cholesterol reduction in cells induces the relocalization of SIDT1 to the plasma membrane in a clathrin-dependent manner.¹¹⁾ Thus, the observed conformational change of SIDT1 may reflect the effect of releasing cholesterol in the transport process. SIDT2 also binds to cholesterol, but it localized intracellular compartment, which may suggest that SIDT2 is involved in the intracellular cholesterol transport. Further study is required to verify the detailed mechanism in SID-1 family proteins.

5. LIPID HYDROLASE ACTIVITY

Structure similarity between SIDT1/2 and ACERs inspires that SIDT1/2 has a ceramidase activity. Indeed, SIDT1 exhibits lipid hydrolase activity for ceramide,^15,16) dihydroceramide¹⁶⁾ and phosphatidylcholine (PC)¹⁴⁾ whereas SIDT2 exhibits lipid hydrolase activity for ceramide.¹³⁾ Mass spectrometry analysis revealed that SIDT1 cleaves PC at the distal phosphodiester bond and produces phosphatidic acid (PA). Furthermore, SIDT1 was co-purified with PA, and PA is bound to the catalytic pocket containing zinc ion coordinating residues as revealed by the structure bound to PA¹⁴⁾ (Fig. 2D). Interestingly, the mutations of Glu555 (the residue capturing carbonyl group at one acyl chain of PA) and His795 (residue coordinating zinc ion) dampened the lipid hydrolase activity and also the dsRNA uptake in HEK293T cells.¹⁴⁾ However, endogenous substrates and physiological function as a lipid hydrolase are unknown.

Intriguingly, cholesterol binding to SIDT1 attenuates ceramidase activity, which is associated with the conformational change¹⁶⁾ described above. In cholesterol bound form of SIDT1, Asp574 at TM6 coordinates the zinc ion, whereas in cholesterol-unbound form of SIDT1, Asp574 is away from zinc ion (Fig. 2E). In the catalytic mechanism of ACER3, zinc ion is directly coordinated by three His residues and a water molecule that forms a hydrogen bond with aspartic acid (Asp), in which the activated water molecule undergoes nucleophilic attack on the ceramide amide bond.²¹⁾ In cholesterol bond form, there is no space for a water molecule to be activated through the coordination with zinc ion. On the other hand, in the cholesterol-unbound form, Asp574 moves away from zinc ion, which creates a gap for a water molecule to be inserted. These observations suggest that cholesterol serves as an allosteric regulator for SIDT1. In the structure of SIDT2, the corresponding aspartic acid (Asp579) is away from zinc ion, which is consistent with the observation that SIDT2 shows ceramidase activity. In the available structures of SID-1, the corresponding aspartic acid (Asp551) coordinates the zinc ion as same as the cholesterol-bound form of SIDT1 (Fig. 2E). Although lipid hydrolase activity of SID-1 has not yet been investigated, its structural features may indicate that SID-1 possess a cryptic lipid hydrolase activity that can be modulated by ligand binding.

6. DSRNA RECOGNITION

The structure of SID-1 bound to dsRNA has been determined,^18,19) which provides the basis of dsRNA recognition by SID-1. The dsRNA runs at an angle of approx. 10° to the membrane surface and approx. 25 bp of dsRNA contact to the lateral side of the extracellular region of SID-1, which is consistent with the observation that SID-1 has greatly weaker affinity for 20 bp dsRNA than for 50 bp dsRNA²⁷⁾ (Fig. 3A). The lateral side of SID-1 exhibits a positively charged surface and binds to dsRNA primarily by electrostatic interactions on three surfaces; one surface located in the loop of ECD1 inserting into a major groove of the dsRNA, one in the linker between ECD1 and ECD2 inserting into a minor groove of the dsRNA and one mediated both ECD1 and ECD2 inserting their charged side chains into the minor and major grooves of the dsRNA (Fig. 3A). The helical rise of A-form dsRNA is approx. 2.5 Å whereas that of B-form dsDNA is 3.4 Å. The major groove is narrowest for dsRNA, wider for DNA-RNA, and widest for dsDNA. Thus, SID-1 specifically recognizes dsRNA via these structural difference from dsDNA. In addition, these interactions are almost through phosphate-ribose backbone, which is consistent with sequence-independent recognition by SID-1. Interestingly, SID-1 residues interacting with dsRNA are not conserved in mammalian SIDT1/2 and the electrostatic surface potential is also different between SID-1 and SIDT1/2, which partly explains why SIDT1/2 exhibit a 1/5 weaker affinity to dsRNA²⁷⁾ (Fig. 3B). In addition, the dsRNA binding of SIDT1/2 is low pH dependent,^14–17) which may indicate the divergent dsRNA recognition by SID-1 family proteins.

Fig. 3. The dsRNA Recognitions by C. elegans SID-1 and SIDT1

(A) Overall structure of C. elegans SID-1 bound to dsRNA (PDB: 8HKE). (B) SID-1 recognizes dsRNA mediated by three surfaces (I, II, III). (C) The electrostatic surface potential of human SIDT1 (PDB: 8KCW). (D) Docking model of SIDT1 bound to dsRNA. Based on the mutational analysis, mutants which greatly (red), modestly (orange) or less (blue) decreased the dsRNA binding are mapped on the molecular surface of SIDT1. dsRNA is placed across the lateral side of SIDT1 ECD.

The structures of SIDT1/2 bound to dsRNA are not available to date despite many attempts. Additionally, low sequence identities between SID-1 and mammalian SIDT1/2 hampered the dsRNA recognition by SIDT1/2. The extensive mutational study has been suggested that several polar residues located in the lateral side of SIDT1 are involved in the dsRNA binding.¹⁶⁾ Those residues are clustered on two distant surfaces and dsRNA can be placed on the line connecting those two surfaces with only a slight steric clash, which suggests that SIDT1 recognizes dsRNA in a similar manner to SID-1 (Fig. 3B). The dsRNA binding affinity is greatly reduced in PNGase treated SIDT1 to remove N-glycans, suggesting that N-glycans of SIDT1 are involved in dsRNA binding²⁸⁾ whereas the N-glycans of SID-1 is not involved in dsRNA binding.

SIDT1/2 binds to dsRNA at the cytoplasmic region via their arginine-rich region in the TM1-TM2 loop and their deletion or the mutation of arginine residues decreases dsRNA binding affinity.^16,29) SIDT2 can also bind to DNA and single-stranded oligonucleotide.

7. CURRENT PERSPECTIVE ABOUT DSRNA TRANSPORT BY SID-1 FAMILY PROTEINS

Structural studies revealed that SID-1 family proteins do not possess tunnels that pass-through dsRNA, suggesting that they do not serve as dsRNA channel or transporter. Thus, it is currently plausible that SID-1 serves as dsRNA receptor in the process of receptor-mediated endocytosis. Interestingly, the deletion of TM1-TM2 loop of SID-1 do not affect dsRNA localization or plasma membrane localization, but greatly decreases dsRNA uptake in S2 cell and systemic RNAi in C. elegans, which suggests that TM1-TM2 loop may be involved in the recruitment of vesicular trafficking-related proteins to facilitate endocytic pathway. However, the detailed mechanism for the SID1-mediated dsRNA uptake or internalization requires further investigation.

SIDT1 is localized at the plasma membrane and is involved in microRNA or double-stranded miRNA mimic uptake by stomach cells. Thus, SIDT1 may function as a dsRNA receptor in endocytic pathway as suggested in SID-1. However, SIDT2 is largely localized on endolysosomal membranes and is suggested to release internalized dsRNA to cytoplasm. Poly(I:C) accumulates in endolysosome in sidt2 knock out bone marrow-derived cells while poly(I:C) is diffused to cytoplasm in wild-type cells. Intriguingly, dsRNA uptake is not altered between wild-type and knockout cells. The released dsRNA activates innate immunity through RIG-I like receptor pathway. Sidt2 −/− knockout mice exposed to Encephalomyocarditis virus (EMCV), or herpes simplex virus 1 (HSV-1) show impaired production of inflammatory cytokines and reduced survival. SIDT1 is also at least partly involved in the same pathway although the contribution by SIDT1 is less than by SIDT2, based on the observations of knock-out mice. The delivery of internalized dsRNA to cytoplasm mediated by SIDT1/2 is not able to be explained by the function as a dsRNA receptor in endocytosis. Moreover, nucleic acid binding to the cytoplasmic region of SIDT2 is suggested to be important for incorporate NAs to lysosomes and their degradation in lysosomes.^29–31) These pathways seem not to be consistent with the receptor mediated endocytosis, which needs to be further investigated.

8. CONCLUSION

RNA uptake by cells has become a hot topic in recent years both for its biological and medical relevance. Past studies have suggested that SID family of proteins (SID-1 in C. elegans, SIDT1 and SIDT2 in human) are key players for RNA uptake. The structures of these proteins provide insight into the architecture of SID-1 family proteins, homo-dimer state, and the conformational change. The pore size of the tunnel is not acceptable for dsRNA transport, suggesting that SID-1 may serve as a dsRNA receptor leading to internalization of dsRNA by the cell, possibly via endocytosis. It is not clear how dsRNA transport is linked to the cholesterol-binding induced conformational change. Further structures that capture the progressive states of RNA binding and cytoplasmic release may enable the mechanism of the RNA transport to be determined.

Acknowledgments

This work was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science Grant Nos. 22K06110 (Y.H.); 23H00366 (T.S.); CREST, JST (Grand No. JPMJCR21E4) (T.S.); JP22H05182 (T.S.).

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Gruner HN, McManus MT. Examining the evidence for extracellular RNA function in mammals. Nat. Rev. Genet., 22, 448–458 (2021).
• 2) Ressel S, Rosca A, Gordon K, Buck AH. Extracellular RNA in viral-host interactions: thinking outside the cell. Wiley Interdiscip. Rev. RNA, 10, e1535 (2019).
• 3) Winston WM, Molodowitch C, Hunter CP. Systemic RNAi in C. elegans requires the putative transmembrane protein SID-1. Science, 295, 2456–2459 (2002).
• 4) Feinberg EH, Hunter CP. Transport of dsRNA into cells by the transmembrane protein SID-1. Science, 301, 1545–1547 (2003).
• 5) Shih JD, Hunter CP. SID-1 is a dsRNA-selective dsRNA-gated channel. RNA, 17, 1057–1065 (2011).
• 6) Duxbury MS, Ashley SW, Whang EE. RNA interference: a mammalian SID-1 homologue enhances siRNA uptake and gene silencing efficacy in human cells. Biochem. Biophys. Res. Commun., 331, 459–463 (2005).
• 7) Elhassan MO, Christie J, Duxbury MS. Homo sapiens systemic RNA interference-defective-1 transmembrane family member 1 (SIDT1) protein mediates contact-dependent small RNA transfer and microRNA-21-driven chemoresistance. J. Biol. Chem., 287, 5267–5277 (2012).
• 8) Nguyen TA, Smith BRC, Elgass KD, Creed SJ, Cheung S, Tate MD, Belz GT, Wicks IP, Masters SL, Pang KC. SIDT1 localizes to endolysosomes and mediates double-stranded RNA transport into the cytoplasm. J. Immunol., 202, 3483–3492 (2019).
• 9) Nguyen TA, Smith BRC, Tate MD, Belz GT, Barrios MH, Elgass KD, Weisman AS, Baker PJ, Preston SP, Whitehead L, Garnham A, Lundie RJ, Smyth GK, Pellegrini M, O’Keeffe M, Wicks IP, Masters SL, Hunter CP, Pang KC. SIDT2 transports extracellular dsRNA into the cytoplasm for innate immune recognition. Immunity, 47, 498–509.e6 (2017).
• 10) Chen Q, Zhang F, Dong L, et al. SIDT1-dependent absorption in the stomach mediates host uptake of dietary and orally administered microRNAs. Cell Res., 31, 247–258 (2021).
• 11) Wolfrum C, Shi S, Jayaprakash KN, Jayaraman M, Wang G, Pandey RK, Rajeev KG, Nakayama T, Charrise K, Ndungo EM, Zimmermann T, Koteliansky V, Manoharan M, Stoffel M. Mechanisms and optimization of in vivo delivery of lipophilic siRNAs. Nat. Biotechnol., 25, 1149–1157 (2007).
• 12) Méndez-Acevedo KM, Valdes VJ, Asanov A, Vaca L. A novel family of mammalian transmembrane proteins involved in cholesterol transport. Sci. Rep., 7, 7450 (2017).
• 13) Qian D, Cong Y, Wang R, Chen Q, Yan C, Gong D. Structural insight into the human SID1 transmembrane family member 2 reveals its lipid hydrolytic activity. Nat. Commun., 14, 3568 (2023).
• 14) Sun CR, Xu D, Yang F, Hou Z, Luo Y, Zhang CY, Shan G, Huang G, Yao X, Chen Y, Li Q, Zhou CZ. Human SIDT1 mediates dsRNA uptake via its phospholipase activity. Cell Res., 34, 84–87 (2024).
• 15) Zheng L, Yang T, Guo H, Qi C, Lu Y, Xiao H, Gao Y, Liu Y, Yang Y, Zhou M, Nguyen HC, Zhu Y, Sun F, Zhang CY, Ji X. Cryo-EM structures of human SID-1 transmembrane family proteins and implications for their low-pH-dependent RNA transport activity. Cell Res., 34, 80–83 (2024).
• 16) Hirano Y, Ohto U, Ichi I, Sato R, Miyake K, Shimizu T. Cryo-EM analysis reveals human SID-1 transmembrane family member 1 dynamics underlying lipid hydrolytic activity. Commun. Biol., 7, 664 (2024).
• 17) Liu W, Tang M, Wang J, Wang F, Song G, Zhang X, Wu S, Ru H. Structural insights into cholesterol transport and hydrolase activity of a putative human RNA transport protein SIDT1. Cell Discov., 10, 21 (2024).
• 18) Zhang J, Zhan C, Fan J, Wu D, Zhang R, Wu D, Chen X, Lu Y, Li M, Lin M, Gong J, Jiang D. Structural insights into double-stranded RNA recognition and transport by SID-1. Nat. Struct. Mol. Biol., 31, 1095–1104 (2024).
• 19) Wang R, Cong Y, Qian D, Yan C, Gong D. Structural basis for double-stranded RNA recognition by SID1. Nucleic Acids Res., 52, 6718–6727 (2024).
• 20) Pei J, Millay DP, Olson EN, Grishin NV. CREST—a large and diverse superfamily of putative transmembrane hydrolases. Biol. Direct, 6, 37 (2011).
• 21) Vasiliauskaité-Brooks I, Healey RD, Rochaix P, Saint-Paul J, Sounier R, Grison C, Waltrich-Augusto T, Fortier M, Hoh F, Saied EM, Arenz C, Basu S, Leyrat C, Granier S. Structure of a human intramembrane ceramidase explains enzymatic dysfunction found in leukodystrophy. Nat. Commun., 9, 5437 (2018).
• 22) Tanabe H, Fujii Y, Okada-Iwabu M, et al. Crystal structures of the human adiponectin receptors. Nature, 520, 312–316 (2015).
• 23) Vasiliauskaité-Brooks I, Sounier R, Rochaix P, Bellot G, Fortier M, Hoh F, De Colibus L, Bechara C, Saied EM, Arenz C, Leyrat C, Granier S. Structural insights into adiponectin receptors suggest ceramidase activity. Nature, 544, 120–123 (2017).
• 24) Valdes VJ, Athie A, Salinas LS, Navarro RE, Vaca L. CUP-1 is a novel protein involved in dietary cholesterol uptake in Caenorhabditis elegans. PLOS ONE, 7, e33962 (2012).
• 25) Otarigho B, Aballay A. Cholesterol regulates innate immunity via nuclear hormone receptor NHR-8. iScience, 23, 101068 (2020).
• 26) Fantini J, Barrantes FJ. How cholesterol interacts with membrane proteins: an exploration of cholesterol-binding sites including CRAC, CARC, and tilted domains. Front. Physiol., 4, 31 (2013).
• 27) Li W, Koutmou KS, Leahy DJ, Li M. Systemic RNA interference deficiency-1 (SID-1) extracellular domain selectively binds long double-stranded RNA and is required for RNA transport by SID-1. J. Biol. Chem., 290, 18904–18913 (2015).
• 28) Yang T, Xiao H, Chen X, Zheng L, Guo H, Wang J, Jiang X, Zhang CY, Yang F, Ji X. Characterization of N-glycosylation and its functional role in SIDT1-Mediated RNA uptake. J. Biol. Chem., 300, 105654 (2024).
• 29) Hase K, Contu VR, Kabuta C, Sakai R, Takahashi M, Kataoka N, Hakuno F, Takahashi SI, Fujiwara Y, Wada K, Kabuta T. Cytosolic domain of SIDT2 carries an arginine-rich motif that binds to RNA/DNA and is important for the direct transport of nucleic acids into lysosomes. Autophagy, 16, 1974–1988 (2020).
• 30) Aizawa S, Contu VR, Fujiwara Y, Hase K, Kikuchi H, Kabuta C, Wada K, Kabuta T. Lysosomal membrane protein SIDT2 mediates the direct uptake of DNA by lysosomes. Autophagy, 13, 218–222 (2017).
• 31) Aizawa S, Fujiwara Y, Contu VR, Hase K, Takahashi M, Kikuchi H, Kabuta C, Wada K, Kabuta T. Lysosomal putative RNA transporter SIDT2 mediates direct uptake of RNA by lysosomes. Autophagy, 12, 565–578 (2016).