key: cord-0747989-ois9fiuq
authors: Rahman, Shah Kamranur; Ansari, Mairaj Ahmed; Gaur, Pratibha; Ahmad, Imtiyaz; Chakravarty, Chandrani; Verma, Dileep Kumar; Chhibber, Sanjay; Nehal, Naila; Sellathanby, Shanmugaapriya; Wirth, Dagmar; Waris, Gulam; Lal, Sunil K.
title: The immunomodulatory CEA cell adhesion molecule 6 (CEACAM6/CD66c) is a candidate receptor for the influenza A virus
date: 2017-01-30
journal: bioRxiv
DOI: 10.1101/104026
sha: 68b45eb2b2741f583bd4772f08f47c7e81ad2215
doc_id: 747989
cord_uid: ois9fiuq

To establish a productive infection in host cells, viruses often use one or multiple host membrane glycoprotein as their receptors. For Influenza A virus (IAV) such a glycoprotein receptor has not been described, to date. Here we show that IAV is using the host membrane glycoprotein CD66c as a receptor for entry into human epithelial lung cells. Neuraminidase (NA), a viral spike protein binds to CD66c on the cell surface during IAV entry into the host cells. Lung cells overexpressing CD66c showed an increase in virus binding and subsequent entry into the cell. Upon comparison, CD66c demonstrated higher binding capacity than other membrane glycoproteins (EGFR and DC-SIGN) reported earlier to facilitate IAV entry into host cells. siRNA mediated knockdown of CD66c from lung cells inhibited virus binding on cell surface and entry into cells. Blocking CD66c by antibody on the cell surface resulted in decreased virus entry. We found CD66c is a specific glycoprotein receptor for influenza A virus that did not affect entry of non-IAV RNA virus (Hepatitis C virus). Finally, IAV pre-incubated with recombinant CD66c protein when administered intranasally in mice showed decreased cytopathic effects in mice lungs. This publication is the first to report CD66c (CEACAM6) as a glycoprotein receptor for Influenza A virus. Significance Statement Cells are enclosed by a semipermeable membrane that allows selective exchange of biomolecules between cells and their surroundings. A set of specialized proteins in this semipermeable membrane, work like gatekeepers to the cell and regulate entry of these biomolecules. One class of such surface proteins is termed as receptors. Viruses bind to one or more of these receptors and manipulate gatekeepers for their own successful entry into host-cells. A membrane protein that influenza A virus (Flu virus) uses for entry into the cells was not discovered till date. This study reports for the first time, a receptor for influenza A virus, that was sought after by researchers for decades. The viral receptor is a promising target that can be used to inhibit virus entry into host cells.

8 showed significant colocalization of NA (red) with CD66c (green) in merged fields (yellow) 182 (Figure 4a) . Altogether, we observed that cells with endogenous level of CD66c showed 183 significant virus binding at the cell surface however cells silenced for CD66c did not show any 184 visible virus binding on the cell surface due to absence of surface receptor CD66c (Figure 4a) . 185

To validate our immunofluorescence assay (IFA) data we performed a western blot analysis to 186 study the effect of siRNA-mediated-silencing of CD66c in virus entry (Figure 4b) hypothesis, we finally performed mAb mediated receptor blockade experiment in A549 lung 218 cells with endogenous (not overexpressed) levels of CD66c on the cell surface. In a control 219 experiment we also tested the effect of mock antibody (IgG isotype) binding on the A549 cell 220 monolayer as against CD66c binding (Figure 6a, b) . We observed a corresponding decrease in 221 virus entry when cells were treated with mAb anti-CD66c at the respective concentrations of 222 1.0 µg/mL, 1.5µg/mL, 2.0µg/mL, 4.0µg/mL and 8.0 µg/mL (Figure 6c ). Inhibition of virus 223 entry was demonstrated by measuring a corresponding reduction in expression levels of viral 224 NP protein in infected cells by flow cytometry (Figure 6d-g) . The dose dependence of mAb 225 anti-CD66c in the inhibition of viral entry was also confirmed by quantitating another viral 226 protein, M1 in the corresponding cells (Figure 6h , i). One set of the antibody mediated receptor 227 blockade experiment was also studied under confocal microscopy, which showed similar 228 inhibition of virus entry into A549 cells that were treated with 4.0 µg/mL of anti-CD66c mAb 229 

A few reports earlier showed that overexpression of two membrane proteins (EGFR and DC-233 SIGN) resulted an increase in virus binding and entry into mammalian cells. Therefore we 234 conducted an experiment to compare virus binding and entry in cells overexpressing CD66c, 235 EGFR and DC-SIGN respectively. For this experiment we checked the extent of virus binding 236 on the lung cells, at the endogenous and overexpressed levels of these host membrane 237 glycoproteins (EGFR, DC-SIGN and CD66c). We reasoned that a genuine receptor upon 238 overexpression in lung cells should exhibit significant increase in virus binding on the cell 239 surface, whereas a weak receptor candidate upon overexpression should display a modest 240 increase in virus binding. After allowing IAV virus to bind to cell monolayers, we examined 241 the cultures under a fluorescence microscope to monitor the membrane glycoproteins (green) 242 and viral NA (red). We did not notice significant increase in virus binding on lung cells 243 overexpressing EGFR as against cells with endogenous levels of EGFR (Figure 7a) . Also, the 244 Altogether, the corresponding increase in yellow spots in cells overexpressing CD66c signifies 252 the higher binding capacity of CD66c (green) towards Influenza A virus (NA) at the cell 253 surface (Figure 7c ) as compared to that of other two glycoproteins (EGFR and DC-SIGN). 254

An increase in virus binding to receptor leads to consequent virus entry into cells. 255 Therefore, after virus binding experiments, we tested and compared ability of other two 256 glycoproteins (EGFR and DC-SIGN) in virus entry with that of CD66c, at the same 257 experimental conditions. We observed, CD66c overexpression in lung cells resulted significant 258 increase in virus entry as monitored by expression levels of viral NP inside cells (Figure 8a) . 259

In contrast, overexpression of DC-SIGN and EGFR did not show much change in virus entry, 260 except for a modest increase in viral NP (Figure 8b, 8c) . We also found that transient 261 overexpression of CD66c did not affect the expression levels of glycoproteins EGFR and DC-262 SIGN (Figure 8a) . Similarly, the overexpression of these two glycoproteins (DC-SIGN and 263 EGFR) had no effect on CD66c expression (Figure 8b, 8c) . 264 265

The membrane glycoprotein DC-SIGN has been documented to serve as a low-specificity virus 267 receptor for IAV and is postulated to facilitate the entry of other viruses like HIV (Human 268 immunodeficiency virus) and HCV (Hepatitis C virus) (11). Critically, we argued that CD66c 269 being a membrane glycoprotein might also be expected to serve as a low-specificity receptor 270 for viruses other than IAV. To this effect, we sought to establish the specificity of CD66c 271 towards influenza virus against an unrelated RNA virus -HCV. We conducted these 272 experiments in human hepatoma Huh cells that were siRNA-mediated-silenced for CD66c 273 expression. We monitored HCV entry into these Huh cells by checking the expression levels of 274 the HCV NS3 protein (Figure 9 ). This data clearly showed that the absence of CD66c in Huh 275 cells had not inhibited entry of HCV, thus proving that CD66c was not a low-specificity 276 general viral receptor. 277

Since we had clearly shown that CD66c was capable of binding to the NA of IAV, we reasoned 280 if we could use heterologously expressed recombinant CD66c protein to bind IAV particles, 281 this should in principle bring down the infectivity of the virus in mice. Thus, we incubated 1µg 282 of biologically active recombinant CD66c (rCD66c), that was produced in mouse myeloma cell 283 lines, with 7.4 X 10 7 PFU IAV before intranasal infection of BALB/c mice. After ten days of 284 infecting mice with virus through intranasal inoculation, we noticed a considerable reduction in 285 alveolitis in the mice that were infected with rCD66c bound IAV ( Figure In our search for a receptor for IAV, we chose to follow the experimental path followed by 297 many other research groups to identify new viral receptors (13-20). We conducted similar 298 experiments in detail, which could validate the interaction between viral NA and host CD66c at 299 the outer cell surface, during IAV attachment and entry. The results thus obtained from these 300 experiments provided sufficient evidence that suggested CD66c as the glycoprotein receptor for 301 influenza virus. The validation of a protein receptor for influenza virus from this study provides 302 valuable insights into some unresolved problems of influenza entry. For instance, it was cited in 303 the earlier studies on influenza entry that although sialic acid was required for virus binding, a 304 specific subset of glycoprotein receptors was necessary for effective viral entry (8, 9) , which is 305 yet unknown. Therefore, with the results presented here, we suggest that CD66c is at least one, 306 of the possibly many, glycoprotein receptors. Additionally, this study may further lead to the 307 discovery of other glycoprotein receptors or co-receptors playing a role in virus entry besides 308

CD66c. The mechanism of virus entry is poorly understood and different alternative routes 309 were suggested for IAV entry, such as clathrin-mediated endocytosis, non clathrin-mediated, 13 caveolin-mediated endocytosis or macropinocytosis (21-28). We believe, CD66c, as a receptor 311 for the influenza virus will help in elucidating the precise route for virus-entry and pave the 312 way for discovery of other co-receptors and their mechanism. 313 Further, with this finding we noticed that IAV follows an infection pattern that is 314 similar to some other viruses wherein they take advantage of adhesive properties of hosts cell 315 adhesion molecules (CAMs), for their attachment and entry. For example, Coronavirus, Rabies, 316

Reovirus and Rhinovirus, employ the following cell adhesion molecules CEACAM1, NCAM-1, 317 JAM-A, ICAM-1 respectively, as their receptor for cellular entry (13-15). Additionally, viruses 318 often interact and utilize these cell adhesion molecules (CAMs) to foster a contact between 319 infected and uninfected target cells for an effective cell-to-cell spread (29). Carcinoembryonic 320

Cell Adhesion Molecule 6 (CEACAM6/CD66c) as the receptor for IAV opens opportunities for 321 further investigations on cell-to-cell spread for this virus as well. Our result showing prominent 322 NA-CD66c interaction at the site of cell-cell junction compared to the rest of the cellular 323 membrane is a preliminary indication in that direction (Figure 1c) . 324

Apart from cell adhesion other unique attributes of CD66c, such as GPI anchoring, lipid 325 raft association and heavy glycosylation (Sialyl-Lewis X ), make it a very suitable and strong 326 receptor candidate for IAV entry. Influenza binding on the cell surface causes lipid raft 327 mediated virus uptake (10), hence we suggest that this putative receptor CD66c being a 328 component of lipid rafts bears potential to further dissect and solve the enigma of the viral 329 internalization mechanism. In addition to that, Sialyl-Lewis X is reported as the common 330 receptor determinant of a number of influenza viruses of the terrestrial poultry (30). Therefore 331 presence of Sialyl-Lewis X on the CD66c molecules makes the latter a strong glycoprotein 332 receptor candidate for the IAV. More importantly, in human lungs there is abundant expression 333 of CEACAM6 by alveolar and bronchial epithelial cells, where it also demonstrates surfactant 334 association and secretion into lung-lining fluid (31). These features of CD66c with respect to 335 human lungs make this molecule vulnerable to respiratory pathogens like IAV. More 336 importantly, like other CEACAMs, which are receptors for respiratory pathogens (bacteria) 337

including Haemophilus influenzae and Moraxella catarrhalis (32, 33), CEACAM6 (CD66c) 338 from above results, serves as a receptor for yet another respiratory pathogen -IAV. 339

It is reported that when pathogens interact with the CEACAM receptors, there is 340 significant activation of PI3K signaling during internalization of the pathogen (34). In our 341 previous report, we validated the activation of PI3k/Akt pathways when CEACAM6 (CD66c) 342 interacts with influenza NA (12), here we demonstrate viral internalization upon NA-CD66c 343 interaction at the cell surface. Also, this new finding on CD66c provides support to the 344 viewpoint of a contentious argument made in the past on a role for NA in influenza entry (35, 345 36) . 346

More importantly, viruses frequently exploit chemokine receptors, some CD markers 347 and other membrane glycoproteins of IgSF as their receptors for entry and also for 348 manipulating the host defense mechanism (37). For this reason, a majority of these 349 immunomodulatory studies get direct reference to viral infections or immune evasion, at the 350 entry stage, which are to a great extent, centric to the interaction of the viral spike glycoproteins 351 with such cellular receptors and co-receptors. It is important to mention here that, the receptor 352 and their modulated co-expression by type I and type II interferon was reported recently (33, 358 41). Altogether these studies establish CD66c as an active immunomodulatory molecule 359 playing a significant role in innate and adaptive immunity in human. Accordingly, for being an 360 active immunomodulatory molecule and with presence on T cells and Macrophages, we took 361 elements of caution into consideration while designing animal experiments and interpreting 362 data thereof ( Figure S1 ). 363 Therefore we argue that immunomodulatory studies carried out earlier in abeyance of 364 any glycoprotein receptor for IAV were rather incomplete and CD66c at the helm of 365 immunomodulation and its interaction with virus during internalization has potential to unfold 366 the precise mechanism of influenza infection, consequent immune response and cell tropism. 367

Lack of any identified glycoprotein receptor during IAV attachment and entry, had greatly 368 limited influenza research discourse in this direction whereas similar questions had been 369 addressed well with other viruses (42) . representative snapshots of NP stained cells from flow cytometry. 786

Chemokine receptors as HIV-1 607 coreceptors: roles in viral entry, tropism, and disease

A novel method to produce Influenza A virus matrix 609 protein M1 Capsid Like Particles (CLPs)

Glycosylations in the globular head of the hemagglutinin protein 613 modulate the virulence and antigenic properties of the H1N1 influenza viruses

The abbreviations used are

SIA, sialic acid; IAV, influenza A virus; EGFR, epidermal growth factor receptor

Liver/lymph node-specific intercellular adhesion molecule-3-grabbing integrin

Dendritic cell-specific intercellular adhesion molecule-3-grabbing nonintegrin; GPI-anchored

Glycosylphosphatidylinositol-anchored

cell adhesion molecules; IgSF, 623 immunoglobulin superfamily; mAb, monoclonal antibody; CEACAM6, carcinoembryonic 624 antigen-related cell adhesion molecule 6; CD66c, cluster of differentiation 66c; m.o.i., 625 multiplicity of infection

DAPI, 4',6-diamidino-2-phenylindole

Lipid raft, Flu, Influenza A Virus, Receptor

IgG super family

The authors declare no competing interests.