CORD-19:bcef0894cf382a83c2aa4d9c803978780c2809db JSONTXT 8 Projects

The MHC class-II HLA-DR receptor mediates bat influenza A-like H17N10 virus entry 1 into mammalian cells 2 Abstract Bats are notorious reservoirs of diverse, potentially zoonotic viruses, exemplified by the 20 evolutionarily distinct, influenza A-like viruses H17N10 and H18N11 (BatIVs). The surface 21 glycoproteins [haemagglutinin (H) and neuraminidase (N)] of BatIVs neither bind nor cleave 22 sialic acid receptors, which suggests that these viruses employ cell attachment and entry 23 mechanisms that differ from those of classical influenza A viruses (IAVs). Identifying the 24 cellular factors that mediate entry and determine susceptibility to infection will help assess 25 the host range of BatIVs. Here, we investigated a range of cell lines from different species for 26 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint their susceptibility to infection by pseudotyped viruses (PV) bearing bat H17 and/or N10 27 envelope glycoproteins. We show that a number of human haematopoietic cancer cell lines 28 and the canine kidney MDCK II (but not MDCK I) cells are susceptible to H17-pseudotypes 29 (H17-PV). We observed with microarrays and qRT-PCR that the dog leukocyte antigen DLA-30 DRA mRNA is over expressed in late passaged parental MDCK and commercial MDCK II cells, 31 compared to early passaged parental MDCK and MDCK I cells, respectively. The human 32 orthologue HLA-DRA encodes the alpha subunit of the MHC class II HLA-DR antigen-binding 33 heterodimer. Small interfering RNA-or neutralizing antibody-targeting HLA-DRA, drastically 34 reduced the susceptibility of Raji B cells to H17-PV. Conversely, over expression of HLA-DRA 35 and its paralogue HLA-DRB1 on the surface of the unsusceptible HEK293T/17 cells conferred 36 susceptibility to H17-PV. The identification of HLA-DR as an H17N10 entry mediator will 37 contribute to a better understanding of the tropism of the virus and will elucidate its zoonotic 38 transmission. 39 40 to accommodate SA 2, 15, 17 . The cell receptors for the BatIVs are as yet unidentified, but they 79 are clearly not SA moieties, a conclusion reached by several studies 15, 18, 19 . Furthermore, bat 80 N10 and N11 are structurally similar to classical NAs but lack conserved amino acids for SA 81 binding and cleavage 2, 16, 17 and do not exhibit typical neuraminidase activity 18, 20 . Initial 82 efforts to isolate infectious BatIVs directly from bats have failed, mainly because the receptors 83 were unknown and susceptible cell lines were unavailable 21, 22, 23 . Attempts to circumvent 84 these limitations have included H17-or H18-pseudotyped vesicular stomatitis virus (VSV 19, 85 DLA-DRA is a well-conserved orthologue of the human leukocyte antigen class II DR α-chain 157 (HLA-DRA) (∼90% amino acid identity between canine, human and Desmodus rotundus bat 158 ectodomains; Supplementary material 3). In humans, MHC-II molecules occur as three highly 159 polymorphic isotypes (HLA-DR, HLA-DP and HLA-DQ) which are selectively expressed 160 under normal conditions on the surface of antigen presenting cells (APCs), including B, 161 dendritic and mononuclear phagocyte cells. These molecules are non-covalently associated 162 heterodimers of two glycosylated, transmembrane polypeptide chains, the monomorphic 163 35-kDa α-chain and the highly polymorphic 28-kDa β-chain 36 . Both chains have an 164 extracellular portion composed of two domains (α1 and α2, or β1 and β2) that is anchored 165 on the cell membrane by short transmembrane and cytoplasmic domains (Fig. 2c ). In the 166 classical scenario, the protease-derived foreign antigen peptides bind to MHC class II proteins 167 in the cleft formed by the α1 and β1 domains, and the complex is transported to the cell 168 surface 36, 37 . When antigenic peptides are not available, endogenous peptides such as the class 169 II associated invariant peptide (CLIP) substitute them and restore MHC class II dimer 170 stability 37 . The complex of HLA-DR and endocytosed peptides (usually 9-30 amino acids in 171 length), constitutes a ligand for the T-cell receptor (TCR) and plays a key role in the 172 presentation of foreign antigens to CD4 + T helper cells and immune surveillance 38, 39 . 173 Since the ultimate goal of our studies is to obtain insight on the zoonotic potential of H17N10, 175 we focused on the possible influence of HLA-DR on cellular susceptibility to H17. Hence, 176 taking advantage of the high expression of HLA-DR on certain human hematopoietic 177 carcinomas 40 , we further explored H17-PV tropism using a panel of human leukaemia and 178 lymphoma cell lines (Fig. 2d) . We found that the Burkitt's lymphoma-derived Raji, Ramos and 179 BJAB B-lymphocytes and the B lymphoblastoid cells (B-LCL) show decreasing susceptibility, 180 in that order, to H17-PV. Kasumi-1 leukaemic cells showed marginal susceptibility in terms of 181 luciferase activity, while Molt-4 and HL-60 leukaemic cells, Jurkat T-cells, pro-monocytic THP-182 1 and U-937 cells, and primary B cells showed no susceptibility to the pseudotypes (Fig. 2d, 183 left Y axis). We hypothesised that the different susceptibility of the various cell types by H17-184 PV were due to disparate expression of HLA-DR, confirmed by qRT-PCR analysis for a non-185 polymorphic region of the α chain of HLA-DR on the same samples (Fig. 2d , right Y-axis). The 186 presence of the HLA-DR heterodimer was also confirmed by flow cytometry with a FITC-187 conjugated monoclonal antibody (clone Tü36), which specifically binds to a monomorphic 188 epitope on the HLA-DR α/β complex and not the isolated α or β chains (Fig. 2e) cells correspondingly reduced the infection of H17-PV by 50% (Fig. 2f) . 203 To determine if blocking attachment of virus to the HLA-DR ectodomain can prevent its entry, 205 Raji cells were incubated with increasing concentrations of a monoclonal antibody (mAb 206 Clone 302CT2.3.2) targeting a monomorphic, extracellular region of the HLA-DRA antigen 207 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint (HLA-DRA epitope: amino acids 48-75). The presence of the antibody significantly reduced, in 208 a dose-dependent manner, infection with H17-PV but not VSV-G-PV (Fig. 2g) . 209 We next sought to ascertain whether ectopic expression of HLA-DR was sufficient to increase 211 the susceptibility of non-APC, HEK293T/17, cells to the H17-PVs. HEK293T/17 cells were 212 transiently transfected with the DRA expression vector, alone or in combination with DRB1. Taken together, HLA-DR is shown to function as a bona fide entry mediator for H17 but may 226 function with unknown factors that facilitate virus internalisation. Interaction between HLA-227 DR and H17 may trigger viral entry through canonical receptor-mediated endocytosis, but 228 could also trigger entry through an activation of cell signalling pathways that the virus 229 subverts to its advantage. Our finding therefore raises questions on the utility and possible 230 evolutionary advantage(s) that an APC-associated receptor would confer to H17N10 231 infectivity and broader fitness. Some viruses exploit cells of the immune system, such as 232 macrophages, B and dendritic cells, either as viral reservoirs or as "Trojan horses" to 233 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint penetrate the epithelial barriers 44, 45 . The measles virus for example, exploits macrophages or 234 dendritic cells, which traffic the virus to bronchus-associated lymphoid tissue or regional 235 lymph nodes, resulting in local amplification and subsequent systemic dissemination by 236 viremia 46 . A similar strategy employed by H17N10 could provide an explanation on why viral 237 RNA was detected in different organs and tissues in carrier S. lilium bats (i.e. lung, kidney, 238 liver, intestine) and why the virus fails to grow in vitro in cell lines developed from the same 239 In this study we did not establish the stoichiometry of the HLA-DR: H17 engagement, or 250 clarify how the virus moves to sub-membranous regions and might hijack the receptor-251 mediated signalling pathway to promote its internalization. It is likely that the determinants 252 of viral entry in vivo are more complicated. We cannot rule out the use by H17N10 (as by 253 other bat-borne viruses, e.g. SARS CoV 51, 52 ) of more than one molecular species as (co-) 254 receptors. Nevertheless, the implication of this study is that H17N10 has the capacity to 255 enter human HLA-DR + cells and our work provides substantial evidence that the H17N10 256 virus has zoonotic potential. The current finding not only sheds light on the understanding 257 of BatIV host range, but also provides additional information on the evolution of influenza 258 A viruses. 259 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint Cell lines (complete description in Supplementary material 1) were kindly provided as supplemented with heat-inactivated 15% fetal bovine serum (Life Technologies), penicillin 283 (100 U/ml) and streptomycin (100 μg/ml; Invitrogen). All cells were maintained in a 284 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint humidified incubator at 37°C and 5% CO2 and were found free of mycoplasma contamination 285 on repeated testing with the MycoFluor Mycoplasma Detection Kit (Life Technologies, UK). 286 were either treated with the endosomal acidification reagent ammonium chloride (1, 10 or 288 100 mM), or pre-treated with neuraminidase from Clostridium perfrigens for 2 h (1, 10 or 100 289 mM) or pronase (a mixture of endo-and exoproteases from Streptomyces griseus at a final 290 concentration of 5, 10 or 50 µg/ml; Calbiochem, UK) for 30 min, an N-glycosylation inhibitor 291 (tunicamycin from Streptomyces sp. at a final concentration of 0.01, 0.1 or 1 µg/ml; Sigma-292 Aldrich, UK) for 5 h. Pre-treated cells were washed with phosphate buffer saline (PBS) 3 293 times, and then incubated/infected as before with PVs for another 24 h. Cell viabilities were 294 assessed by a trypan blue exclusion test. 295 296 Pseudotypes expressing H17 and N10 genes were produced as described previously 32, 53 . 298 Briefly, the lentiviral packaging plasmid p8.91 54 , the pCSFLW firefly luciferase lentiviral 299 vector 55 or the GFP expressing vector pCSGW, the expression plasmids for H17 and/or N10 300 [vector pI.18 56 and the protease encoding plasmid pCAGGS-HAT (a kind gift by Eva Böttcher-301 Friebertshäuser, Philipps University of Marburg, Germany) were co-transfected using 302 polyethylenimine transfection reagent (Sigma Aldrich, UK) into HEK293T/17 cells, plated on 303 6-well Nunclon © plates (Thermo Fisher Scientific, UK). Supernatants were collected 48-72 h 304 post transfection and filtered through a 0.45 µm filter (Millipore, UK). To remove viral titer 305 bias between different PV stocks, pseudotypes were concentrated and (re-) titrated by serial 306 dilution. Concentration was carried out by ultra-centrifugation for 2 h at 25,000 rpm, 4°C in 307 the SW32 rotor of a L2-65B Beckman ultra-centrifuge. 308 Two-fold serial dilutions of PV-containing supernatant were performed as previously 309 described 32 using white 96-well Nunclon © plates (Thermo Fisher Scientific, UK). 310 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint Subsequently, approximately 1x10 4 (for adherent) and 3 x10 4 cells (for suspension) cells were 311 added in 50 µl of medium per well. Plates were incubated for 48 h, after which 50 µl of 312 Bright-Glo™ substrate (Promega, UK) was added. Luciferase readings were conducted with a 313 luminometer (FLUOstar OPTIMA, BMG Labtech) after a 5-minute incubation period and 314 luciferase reading recorded in relative luminescence units (RLU). Data were normalized using 315 Δ-env and cell-only measurements and expressed as RLU/ml. 316 Mammalian expression plasmids (pcDNA3.1 + /C -(K)DYK) for HLA-DRA (NM_019111) and Total RNA was isolated from biological triplicates of early and late passage MDCK from T25 331 flasks that had been seeded with 8 x 10 5 cells and allowed to become confluent and polarize 332 over 4 days in culture cells using a Ribopure kit (Ambion, Austin, TX). Acquired RNA was 333 precipitated with EtOH and subsequently purified employing columns, procedures and 334 reagents from an RNEasy kit (Qiagen, Germantown, MD) and resuspended in RNAse-free H2O. 335 Complementary DNA and RNA synthesis were performed according to Affymetrix Expression 336 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint Analysis protocols (see www.affymetrix.com). Briefly, double-stranded cDNA was synthesized 337 from 5 μg of total RNA using the Superscript double-stranded cDNA synthesis kit (Invitrogen). 338 Following phenol/chloroform extraction and ethanol precipitation, a biotin-labeled in-vitro 339 transcription reaction was carried out using the cDNA template (Enzo Life Sciences, 340 In order to evaluate the interaction of HLA-DR with H17 we used the HLA DRA mAb 361 (302CT2.3.2), which is generated from mice immunized with a KLH conjugated synthetic 362 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint peptide between 48-75 amino acids from human HLA-DRA. After a 1 h pre-incubation with Bio-Techne Ltd) targeting a monomorphic general framework determinant of HLA-DR Class II 386 antigen, diluted in 5% BSA/PBS for 1 hr at RT. The cover slips were then washed 3X with 387 0.02% Tween 20 and 1% BSA in PBS, followed by incubation with Alexafluor 488 conjugated 388 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint anti-mouse (Thermo Fisher Scientific, UK) for 30 minutes at RT. After washing 3X with 0.02% 389 Tween 20 and 1% BSA in PBS, the cover slips were mounted using Prolong Gold containing 390 DAPI (Invitrogen). Images were acquired on EVOS fluorescent microscope (EVOS FL imaging 391 system; Life Technology, USA). Experiments were carried out twice. Reverse: 5'-TCTGGAGGTACATTGGTGTTCG-3'), for canine DLA-DRB1 (Forward: 5'-414 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint AGCACCAAGTTTGACAAGC-3', Reverse: 5-AAGAGCAGACCCAGGACAAAG-3'). and RQ Manager v1.2 (Applied Biosystems). Gene expression data were normalized against 419 the housekeeping gene GAPDH, and compared with the mock controls using the comparative 420 CT method (also referred to as the 2 -ΔΔCT method 58 ). Absolute copy numbers of HLA-DRA in The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint The buffy coat residues for the isolation of CD19 + primary B cells were purchased from the UK 440 20. Garcia-Sastre A. The neuraminidase of bat influenza viruses is not a neuraminidase. 520 Proc Natl Acad Sci U S A 109, 18635-18636 (2012) . 521 author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint Ciminski K, Thamamongood T, Zimmer G, Schwemmle M. Novel insights into bat 523 influenza A viruses. J Gen Virol 98, 2393-2400 (2017 The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/507467 doi: bioRxiv preprint