Trans-endocytosis elicited by nectins transfers cytoplasmic cargo, including infectious material, between cells

Abstract

Here, we show that cells expressing the adherens junction protein nectin-1 capture nectin-4-containing membranes from the surface of adjacent cells in a trans-endocytosis process. We find that internalized nectin-1-nectin-4 complexes follow the endocytic pathway. The nectin-1 cytoplasmic tail controls transfer: its deletion prevents trans-endocytosis, while its exchange with the nectin-4 tail reverses transfer direction. Nectin-1-expressing cells acquire dye-labeled cytoplasmic proteins synchronously with nectin-4, a process most active during cell adhesion. Some cytoplasmic cargo remains functional after transfer, as demonstrated with encapsidated genomes of measles virus (MeV). This virus uses nectin-4, but not nectin-1, as a receptor. Epithelial cells expressing nectin-4, but not those expressing another MeV receptor in its place, can transfer infection to nectin-1-expressing primary neurons. Thus, this newly discovered process can move cytoplasmic cargo, including infectious material, from epithelial cells to neurons. We name the process nectin-elicited cytoplasm transfer (NECT). NECT-related trans-endocytosis processes may be exploited by pathogens to extend tropism. This article has an associated First Person interview with the first author of the paper.

Keywords: Cell adhesion; Cell communication; Cytoplasm transfer; Measles virsus; Nectin; Neuronal entry; Trans-endocytosis; Virus receptor.

Fig. 1.

Intercellular transfer of nectins. (A) Schematic showing the fluorescently tagged nectin constructs used here. Fluorescent tags were inserted upstream of the afadin-binding site located at the C-terminus of the cytoplasmic region (Kurita et al., 2011; Reymond et al., 2001). TM, transmembrane segment. (B) Interactions among nectins. Binding strengths were determined previously (Harrison et al., 2012). Gene name and main synonyms: N1, NECTIN1 (PVRL1, PRR1, HVEC, HigR, CD111); N2, NECTIN2 (PVRL2, PRR2, HVEB, CD112); N3, NECTIN3 (PVRL3, PRR3, CD113); N4, NECTIN4 (PVRL4, PRR4, LNIR). (C–F) Discovery of intercellular nectin transfer. N1–mCherry-expressing A431D cells co-cultured with A431D cells expressing N1–, N2–, N3– or N4–Dendra2, as indicated in the bottom right corners. Merged red and green signals are shown in yellow. White arrows point at colocalization signals at cell adhesion sites, blue arrows at internal colocalization signals. (G) Swap of fluorescent tags has no effect on direction of transfer. N1–Dendra2-expressing A431D cells co-cultured with N4-mCherry-expressing A431D cells. Scale bars: 40 µm.

Fig. 2.

Cellular localization of N1–N4 complexes. (A) Confocal microscopy analyses of cellular localization of N1–N4 complexes. The z-dimension is shown on both the top and right-side bars. N4–Dendra2 (green) and N1–mCherry (red) cells were co-cultured for 24 h; nuclei are stained with DAPI (blue). (B) Airyscan confocal microscopy of co-cultured N4 and N1 cells stained with anti-LAMP-1 antibody (white). (C) Airyscan confocal microscopy of co-cultured N4 and N1 cells stained with anti-VPS35 antibody. Association is pale yellow, denoted by white arrows. Scale bars: 20 µm. (D–K) Functions of the adhesive interface and of the cytoplasmic regions of nectins. All cell lines are derived from A431D cells. (D) Co-culture of adhesion mutant N1(F129D)–mCherry-expressing cells with N4–Dendra2-expressing cells. (E) Co-culture of N1–mCherry-expressing cells with adhesion mutant N4(F132D)–Dendra2-expressing cells. (G–K) Cells expressing N1 proteins with standard or altered cytoplasmic regions (all tagged with mCherry) co-cultured with cells expressing N4 proteins with standard or altered cytoplasmic regions (all tagged with Dendra2). The proteins expressed are indicated in the top-right corners. Color-coding: red, tagged with mCherry; green, tagged with Dendra2. Yellow: colocalization of red and green. Yellow signals in N1 cells, N4 cells and at the adherens junction are highlighted by blue, orange and white arrows, respectively. Scale bars: 40 µm.

Fig. 3.

Uptake kinetics of N4 and cytoplasmic proteins by N1-expressing cells. (A) Cell sorting analyses of N1- and N4-expressing cells. Fluorescence emitted by N1–mCherry cells (first panel), N4–Dendra2 cells (second panel), and by a 1:1 mixture of these two cells 12 h after co-culture (third panel). The fourth panel illustrates the gating strategy used to generate the graph shown in B. (B) One-hour interval kinetic analyses of N4–Dendra2 uptake by N1–mCherry cells. The mean±s.d. of three independent experiments, each one with three technical repeats, are shown. The primary data are in Fig. S3. Blue squares denote an early process causing double positivity in some cells. Since these cells are excluded by doublet discrimination, they are not canonical doublets. Since this effect reproducibly disappears at 2 h, it has no impact on data interpretation after that time point. (C) Cell sorting analysis of cytoplasmic protein uptake efficiency by N1-expressing cells. The first and third panels show N4–Dendra2 and N1–mCherry cells, respectively, without CTdR. The second panel shows CTdR staining of cytoplasmic proteins of N4–Dendra2-expressing cells. The fourth panel shows transfer of the CTdR-labeled proteins to N1–mCherry-expressing cells. The CTdR and mCherry emission spectra do not overlap (third panel). (D) Two-hour interval kinetic analyses of the percentage of N1–mCherry cells becoming CTdR positive. The mean±s.d. of three independent experiments, each one with three technical repeats, are shown. The primary data are in Fig. S4.

Fig. 4.

Transferred viral RNP remain functional in acceptor cells. (A) Schematic of the recombinant MeV–nCFP genome. The additional transcription unit expressing nCFP is inserted before the nucleocapsid (N) gene. (B) Nuclear localization of nCFP, which emits cyan fluorescence. H358 cells were infected with MeV–nCFP and imaged at 48 h post inoculation. Cytoplasmic N protein was visualized with a secondary antibody emitting red fluorescence. Scale bar: 20 µm. (C) MeV–nCFP infects only N4-expressing cells. N4–Dendra2 (left and center panel)- and N1–mCherry (right panel)-expressing cells were incubated with MeV–nCFP (MOI=2), and CFP expression was analyzed 12 h later. The center panel serves as a control for Dendra2 photoconversion in MeV–nCFP-infected cells. Emission maximum of the fluorescent proteins and filters are indicated. (D) Proposed steps in the RNP transfer process, and experimental approach to measure RNP transfer. (1) N4-expressing cells are shown in green, N1-expressing cells in red. An enveloped viral particle (top) is shown with the H and F glycoproteins inserted in its membrane and a helical RNP. (2) Viral particles enter only N4-expressing cells, where they produce cytoplasmic RNPs, glycoproteins that are transported to the cell membrane, and nCFP that is concentrated in the nucleus. (3) N1-expressing cells internalize membranes of N4-expressing cells that include viral glycoproteins and enclose the RNPs (top); double-membrane vesicles with an RNP cargo are formed (center); fusion between the two membranes occurs (close to bottom). (4) RNPs released in the cytoplasm of the acceptor cell are transcriptionally active, resulting in nCFP expression and nuclear concentration. (E) RNP transfer experiment. N4–Dendra2-expressing A431D cells were infected (MOI=2) and allowed to settle for 4 h; N1–mCherry-expressing cells were overlaid at a 1:1 ratio. Cells were analyzed either immediately after mixing (0 h, top row), or after 12 h of co-culture (bottom row). The three-color FACS analyses are shown in two-color combinations; for details see text. (F) Percentage of N1-expressing cells infected by MeV, as determined by nCFP expression. (G) Levels of MeV replication in N1-expressing cells, as determined by nCFP expression mean fluorescent intensity (MFI). The graphs in F and G show mean±s.d. of three independent experiments, each one with three technical repeats. The primary data are in Fig. S5.

Fig. 5.

Epithelial cell lines expressing physiological N4 levels can spread infection. (A) FACS analysis of N4 expression levels in epithelial cell lines H358 and HT1376, as compared to N4–Dendra2 A431D cells. Primary antibody 337516 (R&D Systems) was used to detect N4, with a secondary antibody emitting fluorescence in the PE channel. (B) N1–mCherry-expressing cells internalize N4 from H358 and HT1376 cells. N1–mCherry A431D cells were overlaid on N4–Dendra2-expressing A431D cells (first panel), H358 cells (second panel) or HT1376 cells (third panel), or plated alone (fourth panel), incubated for 12 h, fixed, permeabilized, antibody stained and analyzed by FACS. Antibody 337516 was used to detect N4, with a secondary antibody emitting fluorescence in the APC channel. (C) N1–mCherry expressing cells can internalize MeV RNP from H358 and HT1376 cells. Top: schematic drawing of the experiment. Center row of panels: nCFP expression in mixed cultures infected by MeV–nCFP. Panel order is as in B. Bottom row of panels: quantitative analysis of the double-positive (N1-mCherry and nCFP) cells. The areas selected for quantification are indicated by a red square.

Fig. 6.

MeV spreads from infected epithelial cells to primary neurons. (A) Primary mouse SCG neurons stained with mouse monoclonal phosphorylated neurofilament H-specific antibody (red) or rabbit polyclonal N1-specific antibody (green). Nuclei are contrasted with Hoechst staining (blue). While phosphorylated neurofilament H is only present in axons, N1 is observed in all neuronal structures. (B) Schematic of the recombinant MeV-RNPtracker genome. The additional transcription unit expressing eGFP–P is inserted after the H gene. (C) Viral RNP in association with neuronal axons after co-culture with H358 cells infected with MeV-RNPtracker. Left: GFP fluorescence associated with the MeV-RNPtracker structures. Right: merged phase-contrast image. The arrows indicate GFP positivity distal to cell contact sites. (D) Center panel: drawing of the compartmentalized neuronal culture system. Dissociated mouse SCG neurons are plated in the left compartment. The neurons extend neurite projections underneath two sequential barriers into the right compartment. Right panel: cells infected with MeV-RNPtracker are plated onto the neurites in the middle compartment; the picture was taken 48 h post co-culture. Left panel: SCG cell bodies 14 days after seeding infected H358 cells in the middle compartment. Infected cells emit green fluorescence. (E) Number of fluorescence-emitting SCG cell bodies in the left compartment after overlay of neurites with MeV-RNPtracker-infected H358 cells (red dots) or after direct inoculation with MeV-RNPtracker (black triangles). Six compartmentalized cultures were seeded for each condition. Results are mean±s.e.m.

Fig. 7.

Cell-associated MeV infection of neurons is N4 dependent. (A) Characterization of H358 N4-null SLAM+ cells. Analysis of N4 (left panel) and SLAM (right panel) surface expression. Color coding of cell lines: A431D, light blue; A431D N4–Dendra2, red; H358, magenta; H358 N4-null, orange; H358 N4-null SLAM+, green; VeroSLAM, black. (B) Fluorescent microscopy analysis of H358, H358 N4-null, and H358 N4-null SLAM+ cells 48 h post-infection with MeV-RNPtracker. (C) Time course of MeV–nCFP production by H358, H358 N4-null, and H358 N4-null SLAM+ cells (n=3, error bars, 3× s.d.). (D) Kinetic analysis of infection transfer by cells expressing or not expressing N4. Two experiments are plotted in the left and right panels. H358, H358 N4-null, or H358 N4-null SLAM+ cells were inoculated with MeV-RNPtracker. Vertical axis: number of fluorescence-emitting SCG cell bodies in the left compartment after overlay of neurites in the middle compartment. Each experiment used a minimum of six compartmentalized cultures seeded with infected cells. Results are mean±s.e.m. Asterisks indicate a non-zero number for H358 N4-null SLAM+ cells.