Concentrated A/FPV/Rostock/34 (H7N1) influenza virus was mixed together with fluo-Ad-DOPE to create labeled virus and analyzed by FACScan to determine probe labeling efficiency (Table 1). The intensity of the labeled virus increased proportionally by increasing the concentration of the FSL construct. We observed that fluo-Ad-DOPE fluorescence signal increased ~2.1-fold after 2 h incubation with the virus compared to the fluo-Ad-DOPE initial solution. We further removed an unconjugated dye by gel filtration, which allowed us to determine the real content of the label in the virions and, therefore, to measure the efficiency of the probe insertion (Table 1). In parallel, A/FPV/Rostock/34 (H7N1) influenza virus was labeled with FITC with a conventional direct method (Yoshimura A, et al., 1984), which exhibited 5-fold greater total fluorescence after purification than the same virus labeled with 25 μg/mL fluo-Ad-DOPE and ~2.1-times stronger than labeling with 100 μg/mL fluo-Ad-DOPE (Table 1). In contrast, the number of fluorescent residues incorporated into the virus particle was higher for fluo-Ad-DOPE used at all concentrations compared to FITC labeling, which showed similar results as reported earlier (Yoshimura A, et al., 1984). Notably, identical results were obtained from replicate experiments.
Fluorescence intensity of (×103)a Estimated number of Fluo residues per one virion (×103)b Initial labeling solution Labeling solution after 1-2 h incubation with the virus Labeled virus after purification fluo-Ad-DOPE (25) 3.5 ± 1.0 8.0 ± 1.0 3.0 ± 0.1 2.8 ± 0.1 fluo-Ad-DOPE (50) 6.0 ± 1.0 12.0 ± 1.0 5.0 ± 0.1 6.5 ± 0.1 fluo-Ad-DOPE (100) 8.5 ± 1.0 18.0 ± 1.0 7.0 ± 0.1 12.0 ± 0.1 FITC (100) 1500.0 ± 1.0 1500.0 ± 1.0 15.0 ± 0.1 1.0 ± 0.1 a: The value was normalized for equal concentration of virus particles; b: Since the amount of protein in the sample was known, the mass of a single virion was taken as 1×10-12 mg (Ruigrok R, et al., 1984).
Table 1. Fluorescent labelling efficiency of A/FPV/Rostock/34 (H7N1) influenza virus
To assess the stability of the fluorescent signal, an aliquot of the concentrated A/FPV/Rostock/34 (H7N1) virus labeled with fluo-Ad-DOPE (25, 50, and 100 μg/mL) was stored for 3 weeks at 4℃, and its fluorescent intensity was measured every week (Figure 2A). Our results showed a gradual increase up to ~6.4-fold in fluorescence over time. Re-purification of the virus using a Sephadex G-50 column after prolonged storage resulted in a considerable (2.5-fold) reduction in signal (data not shown), indicating the release of fluorescent molecules either as a native lipid and/or in a degraded form into the supernatant. Similar results (i.e., increased fluorescent signal over time) were observed after fluo-Ad-DOPE labeling of avian A/mallard/Pennsylvania/10218/84 (H5N2) and A/duck/France/46/82 (H1N1) viruses and human A/Puerto Rico/8/34 (H1N1) strains (data not shown). Conversely, we did not observe any changes in fluorescent signal of FITC-labeled A/FPV/Rostock/34 (H7N1) virus after 3 weeks (Figure 2A).
Figure 2. A: Changes in the fluorescence intensity of A/FPV/Rostock/34 (H7N1) influenza virus, labeled with fluo-Ad-DOPE (25, 50, or 100 μg/mL) or FITC during 3 weeks after infection. B: Changes in the absorbance signal of MDCK cells infected with A/California/07/09 (H1N1) virus labeled with increasing concentrations of biot-CMG2-DOPE (50-200 μg/mL) during 3 weeks after infection. Absorbance of the stained cultures 2 h post-infection is shown.
We further assessed whether biotin-labeled biot-CMG2-DOPE probe could be used to label influenza viruses. For this purpose, the human A/California/07/09 (H1N1) strain was modified with increasing concentrations (25-200 μg/mL) of biot-CMG2-DOPE, and the labeled virus aliquots were used to infect MDCK cells (Figure 2B). The cultures were fixed 2-h post-infection and stained with streptavidin-peroxidase. Our results clearly demonstrated concentration-dependent labeling (i.e., the higher concentration of biot-CMG2-DOPE probe was used to label H1N1 virus, the more absorbance intensity observed after MDCK cell infection) (Figure 2B). Similar results were observed for the biot-CMG2-DOPE labeled A/ Perth/16/09 (H3N2) and B/Brisbane/60/08 viruses (data not shown). To assess the stability of the biotin-labeled probe, aliquots of labeled A/California/07/09 (H1N1) and A/Perth/16/09 (H3N2) viruses were stored for 3 weeks at 4℃ and were used to infect MDCK cells every week. No changes in the absorbance signal of infected MDCK cells were observed during 3 weeks compared to the initial signal (Figure 2B, data for the A/Perth/16/09 (H3N2) virus are not shown).
We next studied the ability of A/Puerto Rico/8/34 (H1N1) influenza virus labeled with three different concentrations 25, 50, and 100 μg/mL of fluo-Ad-DOPE to react with MDCK cell suspensions. Our results showed that when labeled virus aliquots were incubated with MDCK cells, they retained their ability to bind these cells (Figure 3). Notably, the intensities of the fluorescent signal of MDCK cells were proportional to the initial fluorescence of the viral samples (Table 1, Figure 3). In parallel, FITC-labeled H1N1 virus was also found to efficiently bind MDCK cell line suspensions. We observed that direct FITC-labeled A/Puerto Rico/8/34 (H1N1) virus that had a higher starting fluorescence compared to the fluo-Ad-DOPE-labeled strain (Table 1) resulted in a higher fluorescence of infected MDCK cells (Figure 3). In addition, cells incubated with fluo-Ad-DOPE-labeled avian A/mallard/Pennsylvania/10218/84 (H5N2) and A/ duck/France/46/82 (H1N1) viruses showed fluorescence intensities similar to that of MDCK cells infected with fluo-Ad-DOPE-labeled A/Puerto Rico/8/34 (H1N1) virus (data not shown). In addition, fluorescent spots of fluo-Ad-DOPE-labeled A/Puerto Rico/8/34 (H1N1) virus were observed on fluorescence micrographs when adherent MDCK cells were infected with the virus (Figure 4).
Figure 3. Binding of fluo-Ad-DOPE-labeled A/Puerto Rico/8/34 (H1N1) influenza virus to MDCK cells. FACS histograms of MDCK cells incubated (1) without virus, (2) with FITC-labeled H1N1 virus, and with fluo-Ad-DOPE-labeled virus at concentrations of (3) 25, (4) 50, and (5) 100 μg/mL.
To study the effect of fluo-Ad-DOPE labeling on receptor specificity, we measured the receptor binding of A/FPV/Rostock/34 (H7N1) virus labeled with three different concentrations of the construct (25, 50, and 100 μg/mL) to "avian-type" sialyl receptor in comparison to the unlabeled virus (Figure 5A). FITC-labeled H7N1 virus was also included into the analysis. Our results clearly showed that neither FITC covalent attachment (100 μg/mL) nor fluo-Ad-DOPE insertion significantly affected H7N1 virus receptor affinity toward 3´SLNPAA-biot (PAA, polyacrylamide; P > 0.05, Figure 5A). No changes in receptor binding to the same sialyl receptor were observed for fluo-Ad-DOPE-labeled A/ mallard/Pennsylvania/10218/84 (H5N2) and A/duck/ France/46/82 (H1N1) viruses in comparison with the unlabeled respective strains (data not shown). Furthermore, fluo-Ad-DOPE labeling had no significant effect on receptor binding of A/Puerto Rico/8/34 (H1N1) strain to the "human-type" 6´SLN-PAA-biot receptor (data not shown).
Figure 5. Receptor binding of influenza viruses labeled with fluo-Ad-DOPE or biot-CMG2-DOPE. A: Binding of A/ FPV/Rostock/34 (H7N1) influenza virus labeled with FITC (100 μg/mL), fluo-Ad-DOPE (25, 50, or 100 μg/mL), or unlabeled with 3'SLN-PAA-biot sialyl receptor. B: Receptor binding of human A/California/07/09 (H1N1), A/ Perth/16/09 (H3N2), and B/Brisbane/60/08 viruses labeled with biot-CMG2-DOPE (50 μg/mL) in comparison with the respective unlabeled strains. Association constants (Ka) of virus complexes with synthetic sialylglycopolymers conjugated to 6´SL (N) are shown. Higher Ka values indicate stronger binding. Values are the means ± SD of three independent experiments.
We next studied the effect of biot-CMG2-DOPE labeling on receptor affinity of human influenza viruses to sialic substrates (fetuin and 6´SL/N) (Figure 5B). As shown by the Ka values, A/California/07/09 (H1N1) and A/Perth/16/09 (H3N2) strains labeled with 50 μg/mL biot-CMG2-DOPE exhibited identical affinity toward fetuin and "human-type" 6´SLN polymer as the respective unlabeled viruses. Similarly, receptor binding of biot-CMG2-DOPE-labeled B/Brisbane/60/08 virus to fetuin and 6´SL did not differ significantly (P > 0.05) from that of the unlabeled influenza B virus (Figure 5B).
To study the impact of fluo-Ad-DOPE and biot-CMG2-DOPE labeling on NA enzyme activity, we performed conventional fluorometric assay (Potier M, et al., 1979). We compared the fluorescent intensities of methylumbelliferone (MUF, λex/em = 360/445 nm), which is the result of NA cleavage of fluorogenic substrate 2'-(4-methylumbelliferyl)-α-D-N-acetylneuraminic acid (MUNANA) and which fluorescent signal does not overlap with that of fluo-Ad-DOPE (λex/em = 495/521 nm) between labeled and unlabeled viruses (Table 2). FITC-labeled viruses were also included in the analysis for comparison. Our results showed that covalent labeling with 100 μg/ mL FITC significantly decreased NA activities of three out of four influenza viruses tested (P < 0.05). Namely, we observed a reduction in ~44% in N1 NA activity of human A/Puerto Rico/8/34 (H1N1), avian A/duck/ France/46/82 (H1N1), and A/FPV/Rostock/34 (H7N1) viruses (Table 2), which could be partially explained by alkaline deactivation of NA enzyme activity due to labeling conditions, but not by the FITC conjugation itself. In contrast, neither fluo-Ad-DOPE nor biot-CMG2-DOPE labeling significantly affected the NA activities of the tested influenza viruses (P > 0.05), except the A/Puerto Rico/8/34 (H1N1) strain, which exhibited reduced N1 NA activities of ~82% and ~52% after labeling with fluo-Ad-DOPE and biot-CMG2-DOPE, respectively (Table 2).
Influenza virus Label (concentration, μg/mL) % of NA activity compared to the respective unlabeled virusa A/Puerto Rico/8/34 (H1N1) fluo-Ad-DOPE (25) 31.3 ± 4.1* fluo-Ad-DOPE (50) 18.4 ± 3.6* fluo-Ad-DOPE (100) 23.1 ± 3.8* biot-CMG2 -DOPE (25) 41.2 ± 5.9* biot-CMG2 -DOPE (50) 53.1 ± 7.3* biot-CMG2 -DOPE (100) 48.8 ± 4.0* FITC (100) 53.1 ± 4.0* A/duck/France/46/82 (H1N1) fluo-Ad-DOPE (50) 96.4 ± 4.4 FITC (100) 72.3 ± 5.1* A/mallard/Pennsylvania/10218/84 (H5N2) fluo-Ad-DOPE (50) 97.5 ± 3.8 FITC (100) 95.2 ± 4.8 A/FPV/Rostock/34 (H7N1) fluo-Ad-DOPE (25) 92.1 ± 6.1 fluo-Ad-DOPE (50) 95.4 ± 5.0 fluo-Ad-DOPE (100) 98.3 ± 4.5 FITC (100) 43.5 ± 3.7* A/California/07/09 (H1N1) biot-CMG2 -DOPE (50) 92.7 ± 9.9 A/Perth/16/09 (H3N2) biot-CMG2 -DOPE (50) 96.3 ± 5.6 B/Brisbane/60/08 biot-CMG2 -DOPE (50) 94.1 ± 6.4 a Unlabeled viruses were incubated with PBS instead of fluo-Ad-DOPE, biot-CMG2-DOPE, or FITC at the same labeling conditions that were used for labeling with the respective dyes, and their NA activities were taken as 100%.
* P < 0.05 compared to the respective unlabeled virus.
Table 2. NA activity of influenza viruses after labeling with fluo-Ad-DOPE, biot-CMG2-DOPE, or FITC
We next assessed whether labeling with fluo-Ad-DOPE and biot-CMG2-DOPE constructs affected influenza virus growth in vitro (Table 3, Figure 6). Human A/ California/07/09 (H1N1), A/Puerto Rico/8/34 (H1N1), A/Perth/16/09 (H3N2), and B/Brisbane/60/08 viruses labeled with 50 μg/mL fluo-Ad-DOPE or biot-CMG2-DOPE replicated to the same titers formed homogeneous plaques with similar sizes (diameter, 0.1-2.8 mm) in MDCK cells as their respective unlabeled viruses (Table 3). To further evaluate the replicative ability of the labeled influenza viruses, we assayed their viral yields in comparison with respective unlabeled strains after multiple replication cycles in MDCK cells (Figure 6, data for A/Puerto Rico/8/34 are not shown). We did not observe any significant differences in growth rates between viruses labeled with fluo-Ad-DOPE or biot-CMG2-DOPE and the respective unlabeled strains at any of the post-infection time points, indicating that non-destructive non-covalent modifications of viral membranes with FSL insertions did not affect their replicative fitness in an in vitro system.
Reciprocal HA titer Virus yield Plaque sizec log10 TCID50/mla log10 PFU/mLb A/California/07/09 fluo-Ad-DOPE (50) 640 7.9 ± 0.1 6.7 ± 0.2 0.5 ± 0.1 (H1N1) biot-CMG2 -DOPE (50) 320 7.7 ± 0.1 6.7 ± 0.4 0.6 ± 0.2 – 640 8.1 ± 0.3 6.6 ± 0.4 0.5 ± 0.2 A/Puerto Rico/8/34 fluo-Ad-DOPE (50) 1024 8.4 ± 0.1 8.8 ± 0.4 0.8 ± 0.1 (H1N1) biot-CMG2 -DOPE (50) 1024 8.1 ± 0.3 8.5 ± 0.3 0.9 ± 0.1 – 1024 8.0 ± 0.5 8.7 ± 0.4 0.8 ± 0.2 A/Perth/16/09 fluo-Ad-DOPE (50) 160 8.2 ± 0.2 7.8 ± 0.6 2.5 ± 0.4 (H3N2) biot-CMG2 -DOPE (50) 160 8.5 ± 0.1 8.1 ± 0.9 2.4 ± 0.5 – 160 8.3 ± 0.1 7.9 ± 0.5 2.8 ± 0.5 B/Brisbane/60/08 fluo-Ad-DOPE (50) 320 8.0 ± 0.4 6.8 ± 0.3 0.1 ± 0.1 biot-CMG2 -DOPE (50) 640 8.1 ± 0.2 6.5 ± 0.2 0.2 ± 0.1 – 640 7.8 ± 0.5 6.7 ± 0.2 0.2 ± 0.1 a: Values are the log10 TCID50/ml ± SD from three independent determinations. TCID50 values were determined in MDCK cells with 10-fold serial diluted viruses, incubated for 72h at 37℃ (or at 33℃ for influenza B viruses), and calculated by the Reed-Muench method (Reed LJ, et al., 1938); b: Values are the log10 PFU/mL ± SD from three independent determinations. PFU was determined in MDCK cells by plaque assay after 3days of incubation at 37℃ with 10-fold serial diluted viruses (or at 33℃ for influenza B viruses); c: Values are mean plaque diameter (mm) ± SD as measured by using the Finescale comparator.
Table 3. Growth characteristics of human influenza viruses labeled with fluo-Ad-DOPE or biot-CMG2-DOPE in vitro
Figure 6. Replication kinetics of human (A) A/ California/07/09 (H1N1), (B) A/Perth/16/09 (H3N2), and (C) B/Brisbane/60/08 influenza viruses labeled with 50 μg/mL fluo-Ad-DOPE or biot-CMG2-DOPE in comparison with respective unlabeled strains in MDCK cells. Confluent cells were infected with viruses at a multiplicity of infection of 0.1 PFU/cell. Virus yield (TCID50/mL) was titrated in MDCK cells 6-, 24-, 48-, and 72-h post-infection. Each data point represents the mean value from two independent experiments.