Toward the understanding of DSG2 and CD46 interaction with HAdV-11 fiber, a super-complex analysis

Abstract

The main limitation of oncolytic vectors is neutralization by blood components, which prevents intratumoral administration to patients. Enadenotucirev, a chimeric HAdV-11p/HAdV-3 adenovirus identified by bio-selection, is a low seroprevalence vector active against a broad range of human carcinoma cell lines. At this stage, there's still some uncertainty about tropism and primary receptor utilization by HAdV-11. However, this information is very important, as it has a direct influence on the effectiveness of HAdV-11-based vectors. The aim of this work is to determine which of the two receptors, DSG2 and CD46, is involved in the attachment of the virus to the host, and what role they play in the early stages of infection.

Keywords: adenoviruses; electron microscopy; receptors; virus-host interactions.

Fig 1

(A) Biolayer interferometry analyses of HAd11K binding to its receptors. Recombinant HAd11K, 100 µg/mL in 10 mM sodium acetate, pH 4, was coupled onto an Octet Amine Reactive Second-Generation (AR2G) biosensor. For kinetics analysis, rDSG2 (EC2-EC3 domains) or rCD46 (SCR1-SCR2 domains) were diluted in the running buffer (HBS-P) (Cytiva) with 3 mM CaCl₂. The concentrations tested were 0.7, 1.4, 2.7, 5.4, and 10.8 µM. All the experiments were performed at room temperature, including association for 180 s, and dissociation for 120 s. The dotted lines represent fits of the raw data (solid lines). They were used to obtain KD, kon, and koff values. (B) Competitive behavior between DSG2 and CD46 was determined using a premix assay. FcCD46 was first captured using Octet Protein A Biosensors (1st Association). HAd11K pre-complexed or not with rDSG2 or rCD46 was allowed to bind to the immobilized FcCD46 (2d Association). Black line, HAd11K; Blue line, HAd11K premixed with rDSG2; green line, HAd11K premixed with rCD46.

Fig 2

HAdV-5/11 or HAdV-3 infection competition studies with either soluble rDSG2 or rCD46 receptor. (A) Visualization of HAdV-5/11 encoding GFP infected cell 24H post infection using ZOE. (B) Quantification of HAdV-5/11 encoding GFP expressing cells by FACS. The percentage of infected cells is calculated using the cell infected by HAd5/11V w:o competitor as 100% of infection. (C) Control experiment with HAdV-3 encoding GFP and rDSG2 as competitor.

Fig 3

HAdV-5/11 entry in CHO-K1 and CHO-DSG2. (A) HAdV-5/11 was incubated at the indicated MOI with either CHO-K1 or CHO-DSG2. One day post infection, GFP expression was observed using ZOE. (B) Quantification of GFP-expressing cells was performed using FACS using non-infected cells to determine the threshold of GFP-positivity.

Fig 4

Analytical size exclusion chromatograms of HAd11K in the presence (dark line) and absence (dashed red line) of rDSG2. HAd11K was preincubated overnight with or without an equimolar concentration of DSG2, prior injection onto a Superdex S200 column. A chromatogram of rDSG2 alone is shown as a dashed blue line. Fractions corresponding to the black chromatogram were separated by SDS-PAGE, followed by Coomassie blue staining.

Fig 5

Structure of HAd11K/ rDSG2 complex and identification of critical binding regions. (A and B) Ribbon representation of the HAd11K trimer (orange, yellow, and blue subunits) in complex with two rDSG2 molecules (salmon and green subunits) in side (A) and top views (B). (C) Overview of the two interaction sites of rDSG2 on two different HAd11K monomers using the same color code as in A and B. Superimposed onto the rDSG2 molecule (salmon) is a ribbon representation of a rDSG2 molecule (gray), which was determined in complex with the HAd7 knob. (D and E) Zoomed views on the two contact zones between rDSG2 and HAd11K highlighted in (C). Putative hydrogen bonds and salt bridges are shown in dotted black lines. (F) Biolayer interferometry sensorgram showing mutated HAd11K (D265A) binding to different concentrations of DSG2 and KD determination. (G) Competition on HeLa cells of HAdV-5/11 preincubated with mutated HAd11K D265A. Visualization of GFP-infected cell 24H post infection using ZOE (left panel) and quantification of GFP-expressing cells by FACS (right panel). The percentage of infected cells is calculated using the cell infected by HAd5/11V w:o competitor as 100% of infection.

Fig 6

Analysis of complex formation between HAd11K and its receptors by size-exclusion chromatography. Mixtures of HAd11K with rDSG2 and rCD46 were examined in four different configurations: (A) HAd11K, rDSG2, and rCD46 were co-incubated at the same time; (B) pre-formed HAd11K/rDSG2 complex was co-incubated with rCD46; (C) pre-formed HAd11K/rCD46 complex was co-incubated with rDSG2; and (D) HAd11K, excess of rDSG2 (molar ratio of rDSG2 to rCD46, 3:1) and rCD46 were co-incubated at the same time. Eluted fractions corresponding to black chromatograms were analyzed by SDS-PAGE, followed by Coomassie blue staining. Dashed lines indicate elution of rDSG2 (blue) and rCD46 (green) injected alone on the same column.

Fig 7

(A) Ribbon representation of a side view of the HAd11K trimer (orange, yellow, and blue subunits) in complex with rDSG2 (salmon) and rCD46 (pink). (B) Overview of the three interaction sites of rCD46 on two different HAd11K monomers using the same color code as in A. Superimposed onto the rDSG2 molecule (salmon) is a ribbon representation of an rDSG2 molecule (gray), which was determined in complex with the HAd11 knob (see Fig. 5). Superimposed onto the rCD46 molecule (pink) is a ribbon representation of an rCD46 molecule (gray), which was determined in complex with the HAd11 knob by X-ray crystallography (PDB ID:2O39) (17). (C–E) zoomed views on the three contact zones between rCD46 and HAd11K highlighted in (B). Putative hydrogen bonds and salt bridges are shown in dotted black lines.