Grantee Research Project Results

 

Although we have already demonstrated the use of MB for the detection of infective coxsackievirues, we were interested in whether the same approach could result in similarly rapid detection of HAV, which do not usually produce plaque in less than one week. A molecular beacon (MB), HAV1, specifically targeting a 20-bp 5' non-coding region of HAV was designed and synthesized. These MBs were introduced into fixed and permeabilized fetal rhesus monkey kidney (FrhK-4) cells infected with HAV strain HM-175. Upon hybridizing with the viral mRNA, fluorescent cells were easily visualized using a fluorescence microscope. Discernible fluorescence was detected only in infected cells by using the specific MB HAV1. A non-specific MB, which is not complementary to the viral RNA sequence, produced no visible fluorescence signal (Figure 1a). To investigate whether this approach can be used to rapidly detect low doses of infectious HAV, cultures infected with 1 PFU of HAV were analyzed from 6 to 48 h P.I. to investigate the minimum time required to consistently detect a positive fluorescent signal. As shown in Figure 2, fluorescent cells can be visualized even within 6 h P.I. and the number of fluorescent cells increased with increasing infection time (Figure 1b). To investigate whether this 6-h infection window could be used to quantify the number of infectious HAV, cells were infected with 1 to 1000 PFU HAV and the average number of fluorescent cells was determined. A linear correlation was obtained by plotting the number of fluorescent cells vs. log PFU (Figure 2), indicating that this MB-based assay can be used as a quantification tool for detecting infectious HAV. A detection limit of 1 PFU was obtained at 6 h post infection, a 32-fold reduction in detection time compared with the 8-day conventional plaque assay.

 

Figure 1. a, Visualization of uninfected or infected FrhK-4 cells by introducing MBs. I. Uninfected cells with 5 μM MB HAV1. II. Highly infected (10⁷ PFU) cells at 72 hr P.I. with 5 μM nonspecific MB. III. Uninfected cells with 5 μM MB HAV1/oligonucleotide hybrids. IV. Highly infected (10⁷ PFU) cells at 72 hr P.I. with 5 μM MB HAV1. b, Visualization of FrhK-4 cells infected with 1 PFU at various P.I. time points. Scale bar = 20 μm.

 

Figure 2. a, A correlation between the fluorescent cells and the corresponding PFU at 6 hr P.I. b, Comparison of the conventional 8-day plaque assay and the fluorescence assay. FrhK-4 cells infected with unknown viral dosages for 6 hr were fixed and permeabilized before 1 hr incubation with 5 μM MB HAV1. The number of fluorescent cells of 27 fields within the chosen chamber well (y) was recorded and the logPFU values (x) were calculated by the calibration equation obtained above. In parallel, the viral dosages were independently determined using the 8-day plaque assay.

 

 

Experiments were conducted to evaluate if FACS was a sensitive method for detecting PV1 infection and to determine if FACS assay was able to detect quantitative differences in the number of infected cells in the sample using the FRET-based cellular reporter system for PV 2A^pro activity. BGM-PV cells were infected with ten-fold serial dilutions of PV1, with the lowest dilution having an MOI of 0.6. Quantification of infected cells was accomplished by FACS. Cells were gated based on size and granularity to include only intact cells. Figure 3 shows representative data plots generated from FACS analysis. After 12 hpi, 78% of the cells (counted from 30,000 events) (Figure 3A) showed a decrease in YFP intensity (indicating disruption of FRET), compared with the uninfected cells with less than 0.2% of the population were positive (Figure 3F). Also, the numbers of infected cells decreased as the infective viruses used to infect the cells become more diluted (Figure 3B to Figure 3E), suggesting that there is a clear correlation between the concentration of the infective virus dose and the number of infected cells. This further suggests that FACS can distinguish infected cells from non-infected cells based on the change in CFP/YFP intensity brought about by disruption of FRET resulting from viral protease activity.

 

Next, we determined the time course of PV1 infection, to identify the optimum incubation time for virus detection. BGM-PV cells were infected with PV1 at an MOI of 0.2, and cells were collected at 0, 2, 5, 8, 12, 16, and 24 h post infection (hpi). Infected cells were detected as early as 5 hpi (3% positives) and continued to increase until 8 hpi (4% positives). A second round of infection was observed between 8 and 12 hpi, and the number of infected cells continued to increase over time (Figure 4). This result demonstrated that PV 2A^pro activity was produced and active as early as 5 hpi. Progeny virions on the other hand were first released around 8 hpi, and that these particles initiated the second round of infection. Increasing numbers of infected cells were detected as the infection progressed indicating that the FACS assay was suitable for following the kinetics of infection.

 

Figure 3. Poliovirus 1 (PV1) infection of BGMPV cells. Confluent monolayers of BGMPV in 6-well plates were infected with ten-fold dilutions of PV1. After 12 h post infection (hpi), cells were trypsinized, washed with 19% FBS followed by 1X TBSS + 3 mM EDTA (pH 8.0), resuspended in the later, and then subjected to FACS. x axis, CFP intensity; y axis, YFP intensity from CFP excitation. The percentage of infected cells shown as % above each graph was determined by counting the number of cells without FRET signal divided by the total number of cells counted. A-E, 10^-1 to 10^-5 dilutions; F, uninfected cells.

 

 

Figure 4. Time course of PV1 infection of BGMPV at a Multiplicity of Infection (MOI) of 0.2 as determined by using FACS. Cells were grown and infected as described, and were harvested at different time points. Those cells that lose FRET due to the uncoupling of CFP-YFP fluorescent pairs brought about by PV1 infection were counted using FACS.

 

To determine the sensitivity of the FACS-based assay at the shortest time possible, we directly compared the FACS-based and plaque assays for the same dilution or preparation of PV1. After 12 hpi, cells were harvested and subjected to FACS analysis. A linear relationship (R² = 0.98) was observed between FACS-based assay and plaque assay results, indicating that the two assays were similar measures of infectious virus in the sample (Figure 5).

 

Figure 5. Direct comparison of FACS and plaque assays. Confluent monolayer of BGMPV in 12-well plates were infected with PV1 (20 min absorption). For FACS assay, cells were harvested after 12 h as described, and then subjected to FACS. For plaque assay, infected cells were overlaid with 1% CMC in MEM (+2% FBS), incubated for 48 h, then fixed/stained with 1% crystal violet in 3.7% formaldehyde solution

 

 

Figure 6. (A) A schematic representation of the TAT-modified nuclease-resistant MB. (B) Nuclease sensitivity assays utilizing ribonuclease-free DNase I. The fluorescence of the nucleaseresistant MB is shown in yellow and the fluorescence of an unmodified MB is shown in red. The background fluorescent signals (shown in black and green) without DNase I addition are also shown. (C) Kinetics of hybridization of TAT-modified MBs CVB6 with (green) or without (orange) complementary oligos.

 

In this study, we describe the use of nuclease-resistant molecular beacons (MBs) for the real-time detection of coxsackievirus B6 (CVB6) replication in living Buffalo green monkey kidney (BGMK) cells via TAT peptide delivery. A nuclease-resistant MB containing 2'-O-methyl RNA bases with phosphorothioate internucleotide linkages was designed to specifically target an 18-bp 5' noncoding region of the viral genome. For intracellular delivery, a cell-penetrating TAT peptide was conjugated to the MB using a thiolmaleimide linkage (Figure 6A). As expected, the modified MBs were highly resistant to nuclease cleavage by DNAase I (Figure 6B). In contrast, an unmodified MB was susceptible to nuclease degradation, resulting in almost instantaneous increase in fluorescence. The dual modifications had no effect on the hybridization kinetics of the MB, as a rapid increase in fluorescence was observed in the presence of a complimentary target (Figure 6C).

 

The intracellular delivery efficiency was tested by incubating 0.5, 1, or 2 μM of MB-target hybrids with a monolayer of Buffalo green monkey kidney (BGMK) cells. Fluorescence was detectable in 100% of the living cells as early as 15 min after introduction of the MB-target hybrids (Figure 7A). The time-lapse images showed that the degree of cellular uptake increased after 15 min and reached saturation after 1 h incubation. The fluorescence intensity was constant for up to 12 h, indicating that the MB-target hybrids were retained inside the cells after delivery and remained resistant to the intracellular RNase H. Unhybridized MB was also introduced into BGMK cells and no fluorescence was detected during the same 12 h period, again indicating the intracellular resistance of the modified MBs to nuclease attack (Figure 7B). In contrast, in the absence of TAT, no internalization of MBs was observed and fluorescence was detected only in the medium (Figure 7B), confirming the effectiveness of the TAT peptide for rapid intracellular delivery.

 

Figure 7. Intracellular delivery of (A) MB CBV6-TAT/target hybrids or (B) MB without TAT modification or without targets. BGMK cells were incubated with 1 μM MB for 12 h, and images were captured using a fluorescent microscope. Scale bar, 20 μm.

 

After validating the properties of TAT-modified MBs, their ability to detect viral RNA was tested. A confluent monolayer of BGMK cells was first incubated with 1 μM MB for 30 min before being infected with 10-fold serial dilutions of CVB6, followed by fluorescence microscopy. Compared with uninfected cultures (0 PFU), where no fluorescent cells were present independent of time, a greater number of fluorescent cells was detected at 2 h post infection (p.i.) for the culture infected with a viral dosage corresponding to one plaque-forming unit (PFU) (Figure 8A). The higher number of fluorescent cells compared to PFU is likely due to the fact that not all viruses that infect a cell are necessarily able to complete the replication cycle. This result is also consistent with the higher infectious virus titers observed using the quantal assay, which is based on the direct microscopic viewing of cells for virus-induced cytopathic effects, than the plaque assay. This 2-h detection window is substantially faster than a similar approach reported using cell fixation/permeabilization. It is possible that the rapid and non-invasive intracellular delivery enables hybridization with viral RNA to occur shortly after virus uncoating without the possibility of degradation caused by fixation/permeabilization. To our knowledge, detection of 1 PFU of CVB6, or any other enterovirus, at 2 h p.i. has never been reported.

 

Using the 2-h infection window, the utility of TAT-modified MBs to quantify infectious CVB6 dosages was tested. Cells were infected with 1 to 200 PFU CVB6 per well and the average number of fluorescent cells was recorded. A linear correlation was obtained by plotting the number of fluorescent cells versus PFU (Fig. 8B). The number of PFU can be easily determined using the correlation obtained after direct counting of fluorescent cells. More importantly, the method can be used to provide rapid quantification of infectious CVB6 dosages within 2 h p.i. compared to the minimum 48-h incubation period for the plaque assay.

 

 

Figure 8. (A) Visualization of BGMK cells infected with 0, 1 or 10⁵ PFU at 2 h p.i. (B) The correlation between the number of PFU and fluorescent cells at 2 h p.i. Error bars represent the standard deviation of three replicate experiments.

 

The ability to detect infected cells continuously should allow one to follow the spreading of infectious viruses on a real-time basis. To determine whether this was, indeed, possible, BGMK cells were infected at a very low infection dosage (multiplicity of infection: 0.01 pfu/cell) and monitored continuously using a fluorescence microscope in a fixed area for 12 h. Figure 9 shows the cell to cell progression of virus spreading at 6 representative time points. Several infected cells were observed at 15 min p.i., suggesting that the viruses entered the cells and started the uncoating process within 15 min. The number of fluorescent cells slowly increased with time suggesting continuous virus infection. By 6 h p.i., a further outward spread of fluorescent cells was observed, indicating the secondary spreading of infection from progeny virions to cells surrounding the initial infected cells. The number of fluorescent cells continued to increase with time and finally spread outward to the entire observation area by 12 h p.i. The majority of infected cells remained adherent and some fluorescence was visible outside the cells, indicating that the fluorescent hybrids with viral RNA entered the extracellular region as a result of the release of progeny virions during cell lysis.

 

 

Figure 9. Real-time detection of viral spreading. BGMK cells were first incubated with 1 μM MB, infected with CVB6 at an M.O.I. of 0.01 pfu/cell, and monitored using a fluorescent microscope.

 

 

In this study, we report the development of nuclease-resistant MBs using QD and Au NP as the FRET pair for real-time in vivo viral detection via Tat peptide delivery. We chose Coxsackievirus B6 (CVB6) as our virus model considering its importance in waterborne diseases. A nuclease-resistant MB targeting an 18-bp non-coding region of the CVB6 genome was designed similar to that reported previously, except for the inclusion of a thiol group at the 5' end and an amino group at the 3' end. A 12-bp linker sequence was inserted at the 3' end to act as a spacer between the MB and the Au NP. To synthesize the QD-MB-Au NP probes (Figure 10), a maleimide-modified hexahistidine (His₆) peptide linker was first conjugated with the free 5' thiol group of the MB to form a stable thioether. The presence of mono-sulfo-NHS esters on the surface of Au NPs enabled their facile attachment to the peptide-MB conjugates via the 3' amino group on the MB. Finally, self assembly of the Au NP-MB conjugates onto DHLA-capped QDs was accomplished via the strong metal-affinity coordination between the ZnS shell and the (His)₆ tag.

 

 

The QD emission at 540 nm was measured to follow the QD-MB-Au NP conjugation (Figure 11A). The effect of QD quenching was investigated by incubating a fixed concentration of QD (0.1 μM) with an increasing molar ratio of Au NP-labeled MB from 1 to 6. Nearly a 40% loss in QD emission was achieved even with an Au NP/QD ratio of 1. This result confirms the correct assembly of Au NP-MB conjugates onto QDs based on the His₆ interaction, resulting in the efficient quenching of the QD via donor-quencher FRET. A sequential increase in the QD quenching efficiency from 37% to 91% was observed when the Au NP MB to QD molar ratio was increased from 1:1 to 6:1. To ensure that the modifications with QD and Au NP had no effect on the hybridization kinetics, an excess amount of complementary oligos was added. Figure 11B shows the time-course recovery of QD emission in the presence of complementary oligos. The QD fluorescence intensity upon target binding was enhanced up to 7.3 times within 50 min.

 

 

Figure 11. Characterization of the QD-MB-Au NP nanoprobe. A) Fluorescence spectra of the QD-MB-Au NP complex at a ratio of MB-Au NP conjugate from 1 to 6. B) The time profile of QD emission in 10-fold molar excess of a complimentary oligo.

 

After confirming the expected properties of the QD-MB-Au NP probes, Tat peptides were appended to the QD surface via coordination with the His₆-tag at a ratio of 10:1. To investigate the intracellular delivery efficiency, QD-MB-Au NP probes appended with Tat peptides were first hybridized with an excess amount of complementary oligos before being added to a monolayer of Buffalo green monkey kidney (BGMK) cells. As depicted in Figure 12, intracellular delivery occurred within 1 h and the level of intracellular fluorescence continued to increase with time. Since the extracellular fluorescence signal continued to decrease within the same duration, this result indicates that the MB-target hybrids were retained inside the cells after delivery. In the absence of Tat peptide conjugation, there was no significant fluorescence detected inside the cells

 

 

Figure 12. Intracellular delivery of QD-MB-Au NP probes. BGMK cells were incubated with the QD-MB-Au NP conjugates (50 nM) for 6 h, and fluorescent images were captured at different time points.

 

To demonstrate the ability of the Tat-modified QD-MB-Au NP probes to monitor the infection state of individual cells, a confluent monolayer of BGMK cells was first incubated with 50 nM probes for 3 h before being infected with 0 to 10³ plaque forming unit (PFU) of CVB6. The number of fluorescent cells was followed by fluorescence microscopy after 4 h of infection. As shown in Figure 13, a significantly higher number of fluorescent cells was detected with increasing infection dosages, while the uninfected cultures (0 PFU) showed a negligible amount of fluorescence (background).

 

 

Figure 13. Detection of infectious viruses by QD-MB-Au NP probes. a) Fluorescent images of cells infected with 0, 1, 10² or 10³ PFU/well at 4 h post infection (p.i.).

 

 

A molecular beacon of MNV-1 (MNV-MB) was prepared by targeting a 23-bp region of the 5' untranslated region of the MNV-1 virus. The DNA backbone was modified with sulfur-substituted 2'O-methyl oligo-ribonucleotides for improved nuclease resistance. A modified TAT peptide was mixed with 1thiolated MB to form a stable thiol-maleimide linkage. The MNV-1 stock and RAW 264.7 cells were prepared according to published procedures.The TAT-modified MBs were tested to detect MNV-1 infection of RAW 264.7 cells by incubating the MBs with MB-complimentary oligonucleotide target hybrids. For MNV-1 experiments, the cells were incubated with MB MNV-Tat. The cells were then infected with MNV-1, followed by fluorescence microscopy. The infectious dose was independently confirmed by plaque assays. In the chamber well containing 1 PFU, fluorescent cells could be visualized within 6 hours post infection (p.i.). At 12 hours p.i, in the chamber well containing 100 PFU, 100% fluorescence was observed, the result of the onset of secondary infection due to the release of mature virions during cell lysis.

 

The MNV-1 MB was used to detect MNV-1 in groundwater samples. A 50-mL groundwater sample was seeded with MNV-1 at a concentration of 10⁵ pfu/mL, which was serially diluted, then inoculated onto RAW 264.7 cells for the fluorescent assay experiments. The slide was placed on the microscope stage and fluorescent images were taken over 12 hours (Figure 14). The plaque assay required 72 hours to detect MNV-1, compared to the 6-hour detection window for the TAT-modified MB assay. A linear correlation (r² = 0.9821) between the number of fluorescent cells versus plaque-forming units was observed (Figure 15). Due to the rapid and noninvasive intracellular delivery of the TAT-modified probe, the hybridization of viral RNA occurs shortly after virus uncoating, enabling virus detection early in its replication cycle. This study demonstrated the use of TAT peptide-linked molecular beacons as a rapid tool for detecting and quantifying infective murine noroviruses

 

 

Figure 14. Visualization of cell to cell spread of 0.01 MOI of MNV-1 in RAW 264.7 cells inoculated with dilutions of groundwater sample.

 

 

Figure 15. Comparison of plaque assay (pfu) vs fluorescent cells obtained by molecular beacon assay.