Effect of ambient temperature on respiratory tract cells exposed to SARS-CoV-2 viral mimicking nanospheres—An experimental study

Abstract

The novel coronavirus caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has reached more than 160 countries and has been declared a pandemic. SARS-CoV-2 infects host cells by binding to the angiotensin-converting enzyme 2 (ACE-2) surface receptor via the spike (S) receptor-binding protein (RBD) on the virus envelope. Global data on a similar infectious disease spread by SARS-CoV-1 in 2002 indicated improved stability of the virus at lower temperatures facilitating its high transmission in the community during colder months (December–February). Seasonal viral transmissions are strongly modulated by temperatures, which can impact viral trafficking into host cells; however, an experimental study of temperature-dependent activity of SARS-CoV-2 is still lacking. We mimicked SARS-CoV-2 with polymer beads coated with the SARS-CoV-2 S protein to study the effect of seasonal temperatures on the binding of virus-mimicking nanospheres to lung epithelia. The presence of the S protein RBD on nanosphere surfaces led to binding by Calu-3 airway epithelial cells via the ACE-2 receptor. Calu-3 and control fibroblast cells with S-RBD-coated nanospheres were incubated at 33 and 37 °C to mimic temperature fluctuations in the host respiratory tract, and we found no temperature dependence in contrast to nonspecific binding of bovine serum ablumin-coated nanospheres. Moreover, the ambient temperature changes from 4 to 40 °C had no effect on S-RBD-ACE-2 ligand-receptor binding and minimal effect on the S-RBD protein structure (up to 40 °C), though protein denaturing occurred at 51 °C. Our results suggest that ambient temperatures from 4 to 40 °C have little effect on the SARS-CoV-2-ACE-2 interaction in agreement with the infection data currently reported.

I. INTRODUCTION

The novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) continues to infect large parts of the human population globally (more than 15 × 10⁶ cases in the U.S. and 68 × 10⁶ globally—a figure that is growing rapidly).¹ A similar contagion, severe acute respiratory syndrome (SARS) infection, caused by another coronavirus (SARS-CoV-1), emerged in China in mid-November 2002, but unlike SARS-CoV-2, SARS-CoV-1 was largely eliminated by July 2003. The general consensus for decreased infection and spreading of SARS-CoV-1 was the increased ambient temperature during the summer.^2–5 As the present SARS-CoV-2 resembles SARS-CoV-1 in its genetic makeup and viral structure,⁶ several agencies and researchers in the scientific community were hoping for a similar effect, wherein the elevated temperatures and humidity of the summer months would reduce viral prevalence and infections.^7,8 Indeed, emerging epidemiological and mathematical models during the spring months of 2020 suggested that the viral stability fluctuates with ambient temperatures and may affect its transmission in the community.^9–16 However, these studies mainly relied on in silico modeling of patient infection or hospitalization data and did not provide experimental evidence on how or if an increase in the ambient temperature would affect cellular infection by SARS-CoV-2. In contrast, both epidemiology and laboratory studies confirmed that SARS-CoV-1 infection decreased with increasing ambient temperature for the virus on surfaces.⁴

Respiratory tract viruses that replicate in epithelial cells of the nasal cavity and the throat have evolved to survive at temperatures below 33 °C.^17,18 As a result, respiratory tract viral infections increase in winter as seasonal ambient temperatures fall,¹⁹ during which inhaling cold and dry air in the winter lowers the respiratory tract airway temperature to 31–33 °C.^20,21 Thus, these ambient conditions provide an optimal scenario for viral infection and transmission.^22,23 As ambient temperatures increase in the summer, the respiratory tract tracheal temperature increases to 37 °C, resulting in decreased viral susceptibility to host cells.²⁴ Several recent epidemiological studies on multiple countries have reported inconsistent correlations between ambient temperatures and daily new cases of COVID-19.^9,25,26 Epidemiological and meteorological data collected from 166 countries on daily new cases of COVID-19 until March 2020 suggested that rise in temperatures and relative humidity might prevent the rise in cases of COVID-19. The increase in the 1 °C ambient temperature was associated with a 3% reduction in daily new cases.⁹ On the other hand, epidemiological data collected from U.S. until April 2020 indicated that the decreased incidence of COVID-19 infections due to increased temperature was small and associated with an increase in the UV radiation index, as well as the countermeasures that were taken.²⁵ Global infection data over the last six months have shown us that viral infections have continued to increase despite temperature fluctuations; however, varying countermeasures taken by different states and municipalities have made it difficult to isolate the true relationship between the ability of SARS-CoV-2 to infect cells and the surrounding temperature. This information may help in planning for viral infection loads as colder months approach in the winter even as the newly developed vaccines^27,28 gain global approval and are administered in vaccine campaigns.

From a mechanistic perspective, it has been suggested that an increase in the ambient temperature promotes structural rearrangement of the virus surface proteins, affecting viral stability, viral enzymatic activity, and ligand-receptor binding efficiency to modulate viral trafficking into and out of host cells.^18,29,30 Campos et al. showed that viral envelope glycoprotein domains in dengue and Zika virus undergo structural reconfiguration at temperatures above 33 °C that strongly decreased the trafficking of viral particles.³¹ Unfortunately, information about the virus-host cell interaction as a function of temperature has not been reported for SARS-CoV-2. As the SARS-CoV-2 spike glycoprotein (S protein) is one of the primary targets for ongoing vaccine and therapeutic development, with antibodies against the S protein receptor-binding domain (RBD) being a prominent target,³² temperature-dependent studies of SARS-CoV-2 protein structural stability could be valuable for molecular design.^33,34

SARS-CoV-2 infects host epithelial cells in the nasal cavity, airway, and lungs upon binding to the angiotensin-converting enzyme 2 (ACE-2) surface receptor via its S protein.³⁵ The S protein's receptor-binding domain (S-RBD) on SARS-CoV-2 mediates the critical processes of binding to host ACE-2 receptors for viral uptake.³⁶ Given that SARS-CoV-2 is highly infectious and dangerous, we pursued a strategy using small polymer beads coated with the S-RBD as a viral mimicking particle. Viral protein-coated nanoparticles have been shown to mimic “infection” by initiating cellular uptake previously,^37,38 reducing experimental risk, cost, and time requirements. This system also offers a facile method to emulate virus parameters as new information arises about the virus structure.³⁹ Importantly, information from similar studies employing viral mimics has aided therapeutic development for other viral infectious diseases.^37,40,41

II. MATERIALS AND METHODS

A. Preparation of S-RBD-coated viral mimicking nanospheres

Surface coating of the S protein on nanospheres was performed using physical adsorption.⁴² In brief, the SARS-CoV-2 spike (S1) receptor-binding domain (S-RBD) protein (1 mg/ml, RayBiotech, 230-01101) and fluorescence nanospheres (beads) (10¹¹/ml) (Yellow-Green 100 nm carboxylated fluorescence beads; FluoSpheres, Thermofisher) were mixed in an Eppendorf (protein LoBind tube to minimize adhesion to tube wall surfaces) with the PBS buffer, and the solution was placed onto a thermo-block (Fisherbrand, ThermoFisher Scientific) at 300 rpm for 24 h at 4 °C. The protein-coated beads were then spun down by centrifugation (10 000 rpm, 15 min), and the supernatant was collected to determine the residual amount of proteins in a solution using a Bradford assay (Sigma) in accordance with the manufacturer's protocol.⁴³ Surface coating of the S-RBD protein on nanospheres was examined using transmission electron microscopy (TEM) (JEOL 2010F). The amount of protein coating on the surface of the nanospheres was estimated by subtracting the residual amount of proteins found in the supernatant solution from the initial amount used for the incubation. After initial protein adsorption, the S-RBD-coated beads were blocked with a 1% BSA (Sigma) solution for 24 h at 4 °C to ensure complete surface coverage, after which the beads were collected. For control experiments, BSA-coated beads were prepared in accordance with the aforementioned protocol without any S protein coating. The obtained protein-coated beads were diluted, aliquoted, and stored in the PBS buffer (10⁸ beads/ml) at −20 °C.

B. Cell culture and interaction with viral mimic nanospheres

Lung adenocarcinoma cells (Calu-3, ATCC HTB-55™) were obtained from American Type Culture Collection (ATCC), and NIH-3T3 fibroblasts (ATCC CRL-1658) were a gift from the Zoldan laboratory (UT Austin). Calu-3 cells were cultured in EMEM (with EBSS and l-glutamine, BioWhittaker, Lonza) containing 20% FBS (heat-inactivated, Gibco), 0.1 mM MEM nonessential amino acids (Gibco), and 10 mM HEPES (Gibco) at a pH of 7.4. The passage number was kept below ten for all experiments. NIH-3T3 cells were cultured in DMEM + GlutaMax™-I medium (Gibco) supplemented with 10% FBS. Cells were incubated at 37 °C under 5% CO₂ atmosphere, and culture media were regularly replaced every two days until cells reached 70%–80% confluency. For cell passaging and subculture, cells were removed using 0.25% Trypsin-EDTA (VWR).

To investigate the effect of S-RBD mediated specific binding, Calu-3 and control fibroblasts were seeded independently into 96 wells at a concentration of 125 000 cells/cm² and incubated at 37 °C. After 48 h of culture, the culture media were removed and replaced with a fresh aliquot of media containing suspended S-RBD-coated nanospheres, control BSA-coated and negative control neat nanospheres (10⁶ beads/well), and allowed to interact with cells at either 33 or 37 °C for 48 h. After 48 h of culture, the media containing the beads were removed, and the cells were washed two times with PBS, stained with 5 μM DRAQ5 (Thermo Scientific, nuclear dye, 5 min at 37 °C), and fixed with a 4% paraformaldehyde solution for 20 min and washed again with PBS. All the experiments were carried out with three technical repeats (N = 3), and each technical repeat had three subset biological repeats (n = 3).

Furthermore, in order to understand the role of the ACE-2 mediated interaction with S-RBD-coated nanospheres, a competitive binding experiment was conducted by blocking the ACE-2 receptor on the Calu-3 cell surface prior to bead exposure. To block the ACE-2 receptor, Calu-3 cells were cultured at 37 °C for 48 h and incubated with fresh culture media having a 1 μM full-length extracellular recombinant SARS CoV-2 S1+S2 protein (Abclonal, RP01260LQ) for 2 h before adding S-RBD-coated nanospheres (10⁶ beads/well) and cultured for further 48 h. Concurrently, unblocked Calu-3 cells were used as a control. After 48 h of culture, cells were fixed and stained as described above for confocal imaging.

C. Confocal imaging and quantification of viral mimicking nanosphere interactions with cells

Interactions of the viral mimicking nanospheres with Calu-3 and control NIH-3T3 fibroblasts cells were evaluated using confocal microscopy (Olympus FV3000). Z-stacks were acquired with a 20×, 0.75 NA air objective (Olympus). Nanospheres were excited with the 488 nm laser line, and their fluorescence was detected between 500 and 600 nm. DRAQ5 stained nuclei (and the leaky cytosolic signal) were excited with the 640 nm laser line, and fluorescence was detected at 700–800 nm. For quantification of the relative binding of BSA and S-RBD-coated nanospheres, the bead area in each field-of-view was segmented and normalized to an outline region of the cell area using a custom imagej script. Briefly, beads were identified from a z-projection using the average intensity projection, a gaussian blur with sigma 0.5 was applied, and then manual thresholding was performed to determine the area covered by beads. Regions of interest (ROIs) were manually selected from confocal fluorescence images to analyze only areas covered by cells and omit any particles that bound the bare well plate surface. The metric reported here is the relative area of nanospheres in an ROI divided by the whole ROI area (or the entire cell area) averaged from around 10–15 images per well and then further averaged for all subset biological repeats (n = 3).

D. ACE-2 receptor expression and immunofluorescence imaging

ACE-2 receptor expression was evaluated by culturing Calu-3 and NIH-3T3 cells at 33 and 37 °C for 48 h, followed by antibody labeling. After 48 h culture, cells were fixed with a 4% paraformaldehyde solution for 20 min and blocked with a 5% BSA solution for 1 h at room temperature followed by several PBS rinses. Then, cells were incubated with the ACE-2 antibody (4 μg/ml) conjugated with either Alexa Fluor 647 or 546 (Santa Cruz Biotechnology, SC-390851) overnight at 4 °C. Thereafter, the ACE2 antibody solution was removed; the sample was washed three times with PBS, covered, and imaged using confocal fluorescence microscopy using the same objective and settings as described above. The fluorescence intensity of the cell surface ACE-2 receptor was measured using imagej in accordance with the below-mentioned formula described in the previously reported literature.⁴⁴

Corrected total fluorescence intensity = Integrated Density − (Area of selected region × Mean fluorescence of background reading).

E. Simulating environmental changes for an S-RBD-coated nanosphere interaction with cells

To evaluate the effect of environmental temperature on virus-cell interaction, S-RBD-coated nanospheres suspended in PBS were exposed to different temperatures (25 °C, 33 °C, 37 °C, and 40 °C) for different time points (1 h, 4 h, 12 h, and 24 h) and then incubated with Calu-3 cells at 37 °C for 48 h to evaluate cellular binding efficiency. As the number of conditions for these experiments was much larger, S-RBD-coated beads bound to Calu-3 cells were imaged using a multimode plate reader (Cytation 5, BioTek) using a 10× objective with CFP/Texas Red filter cubes. The same relative area metric as described for confocal images was calculated here to quantify nanosphere binding to cells using an imagej script similar to that described for the confocal images above, though without any z-projection.

F. Thermal stability of the spike receptor binding protein

The thermal stability of recombinant S-RBD was monitored using nanodifferential scanning fluorimetry (nDSF) [Prometheus NT.48; (UV-LED 285 nm), NanoTemper Technologies], which measures the intrinsic fluorescence of proteins to quantify protein stability with changes in the temperature. The S-RBD protein (1 mg/ml) was measured in triplicates in microcuvette arrays, and fluorescence spectra were collected from 20 to 40 °C with a temperature ramp of 1 °C/min. The obtained intrinsic fluorescence spectra were further processed to create graphs of the fluorescence intensity ratio of 350 /330 nm and the first derivative of (F350/330) versus temperature.

G. Statistical analysis

Significant differences between the groups were analyzed using one-way ANOVA (analysis of variance) with Tukey’s test for multiple comparisons used for statistical analysis. Differences were considered statistically significant for p < 0.05 and indicated by symbols in the figure. All the analysis was performed with Origin Pro.

III. RESULTS AND DISCUSSION

A. SARS-CoV-2 mimicking spike RBD-coated nanospheres

Coronaviruses are naturally occurring viruses that are ∼100 nm in diameter and have a rich protein corona on the surface of a lipid envelope that encases the RNA genetic material of the virus. Over the years, researchers have conjugated native viral surface proteins on high surface area polymer beads to prepare coronavirus-mimicking nanospheres.^45,46 Chen et al. used a facile approach to formulate synthetic, viral mimicking nanospheres by physisorption of a recombinant Avian coronavirus spike glycoprotein on 100 nm gold particles. These synthetic, viruslike particles display morphological and antigenic properties similar to native Avian coronavirus particles.⁴⁵ In a similar way, we made SARS-CoV-2 mimicking particles by coating 100 nm carboxylated fluorescence nanospheres (beads) with recombinant S protein receptor-binding protein domain (S-RBD) as shown in Fig. 1(a) as a safer and simpler alternative to working with live SARS-CoV-2 or pseudotyped viruses, respectively, for the binding studies reported here. TEM images of S-RBD-coated nanospheres showed a protein shell compared uncoated nanospheres [Fig. 1(b)], suggesting the presence of S-RBD on the surface of nanospheres, as a viruslike particle.

FIG. 1.

Synthesis of SARS-CoV-2 mimicking nanospheres. (a) Schematics showing the SARS-CoV-2 mimicking fluorescent nanosphere next to the native virus. (b) TEM images of neat and S-RBD-coated nanospheres and (c) adsorbed protein quantification by the Bradford assay.

We quantified the adsorbed protein on S-RBD nanospheres as well as on control BSA-coated nanospheres. The average density of the adsorbed protein was 1.06 and 0.66 mg/m² on the S-RBD and BSA-coated nanospheres, respectively. Increased adsorption of the S-RBD on carboxylate nanospheres can be attributed to strong electrostatic interactions, with positively charged S-RBD, having a pI around 8.5 (the isoelectric point of the S-RBD protein is calculated from amino acid sequences—UniProtKB-P0DTC2; 319-541 amino acids using ExPASy ProtParam). On the other hand, the BSA adsorption onto carboxylated beads was likely mediated by hydrophobic interactions, as BSA (pI = 5.4) and the carboxylated bead (pI ∼ 4.5) surface are both negatively charged at physiological pH. The obtained surface adsorbed S-RBD-coated nanospheres mimic the native SARS-CoV-2 virus from a size and protein coating perspective.

B. S-RBD-coated nanospheres bind significantly more to ACE-2 expressing cells

We observed that cells incubated with S-RBD-coated beads retained their viability as they continued to grow in the presence of suspended nanospheres in culture media with negligible leakage of cytosolic lactate dehydrogenase (LDH) (Fig. S1).⁶⁵ As S-RBD only contains the ACE-2 receptor binding domain, it mainly facilitates cell-particle interaction on cell membranes when the ACE-2 receptor is present. The cellular interaction of S-RBD-coated nanospheres was significantly higher for Calu-3 cells, which express the ACE-2 receptor, in comparison with control, ACE-2 negative NIH-3T3 fibroblasts (Fig. S2).⁶⁵ S-RBD-coated nanospheres firmly adhered to the Calu-3 cell membrane forming clusters or aggregates on the membrane surface. In Calu-3 cells, very few nanospheres were found inside the cell cytoplasm and most remained aggregated on the membrane surface (Movie S1).⁶⁵ In contrast, control fibroblasts (NIH-3T3) lacking ACE-2 receptor expression (Fig. S2)⁶⁵ displayed a minimal amount of aggregated fluorescent S-RBD-coated nanospheres [Fig. 2(a)].

FIG. 2.

Interaction of S-RBD-coated nanospheres with cells: (a) confocal micrographs of Calu-3 cells and NIH-3T3 cells (DRAQ5, red) showing differential binding ability for S-RBD-coated nanospheres (green) (scale bar: 20 μm). (b) Quantification of bound S-RBD coated nanospheres on Calu-3 cells and NIH-3T3 cells via the relative projected area from microscopy data. (c) Co-staining of S-RBD-coated nanospheres (green), ACE-2 receptor expression (primary conjugate to Alexa546, blue), and nucleus (Draq5, red) of Calu-3 cells at 37 °C. Inset: zoomed image of a Calu-3 cell co-stained of S-RBD-coated nanospheres (green), ACE-2 receptor expression (blue), and nucleus (red). Statistically significant differences (p < 0.05) compared to Calu-3 are indicated by *.

By measuring the bead area with respect to the Calu-3 and NIH-3T3 cell area in confocal images, we quantified the binding of S-RBD coated nanospheres [Fig. 2(b)]. On average, Calu-3 cells showed 5.4% of the cellular area covered by beads compared to 0.2% in fibroblasts, resulting in nearly 27-fold higher retention of viral mimicking S-RBD-coated nanospheres by Calu-3 in comparison with fibroblasts. These results suggest the superior binding and retention capacity of viral mimicking S-RBD-coated nanospheres by lung cells (Calu-3)—bearing the ACE-2 receptor—in comparison with control fibroblasts. Further evidence is shown in Fig. 2(c), cyan, where a co-stain of ACE-2 and fluorescent S-RBD-coated nanospheres shows a significant overlap. The high magnification inset [Fig. 2(c)] further shows the presence of S-RBD-coated nanospheres (green) on the Calu-3 membrane surface expressing ACE-2 receptor (blue). Given that the SARS-CoV-2 virus is known to interact with lung epithelial cells with decisive specificity via the S-RBD against the host ACE-2 receptor,⁴⁷ our results are consistent with findings using pseudoviral particles.⁴⁸

To assess the effect of temperature on cellular binding of SARS-CoV-2 mimic particles, Calu-3 and control fibroblast cells were incubated with S-RBD-coated nanospheres at 33 and 37 °C in order to mimic the salient change during winter and summer seasonal temperature fluctuations in the host respiratory tract.²² We found no effect of the incubation temperature on the cellular interaction with S-RBD-coated nanospheres. Both cell lines (Calu-3 and NIT3T3) showed no difference in the S-RBD-coated nanosphere interaction at 33 and 37 °C, respectively, as can be seen from Figs. 3(a) and 3(b).

FIG. 3.

Temperature effect on binding of S-RBD-coated nanospheres by cells: (a) confocal images of Calu-3 cells and NIH-3T3 (DRAQ5, red) showing binding ability for S-RBD-coated nanospheres (green) at 33 and 37 °C (scale bar: 20 μm). (b) Quantification of bound S-RBD-coated nanospheres on Calu-3 cells and NIH-3T3 cells at 33 and 37 °C. Statistically significant differences (p < 0.05) compared to Calu-3 are indicated by *.

In order to distinguish between the effect of the temperature on receptor-mediated binding and nonspecific interactions of nanospheres with Calu-3 and NIH-3T3 cells, we conducted a control experiment using BSA-coated nanospheres and neat, uncoated nanospheres (Fig. S3).⁶⁵ Calu-3 and NIH-3T3 cells showed similar binding for nonspecific interactions with BSA-coated nanospheres and uncoated nanospheres (as a true negative control). Interestingly, both Calu-3 and NIH-3T3 fibroblasts showed a 65% and 47% increase (both statistically significant), respectively, in interaction with BSA-coated and uncoated nanospheres with an increase in the incubation temperature from 33 to 37 °C (Fig. 4 and Fig. S3).⁶⁵ Considering the similar binding of the uncoated and BSA-coated beads at both temperatures, these findings indicate: (1) a temperature-dependent, non ACE-2-mediated mechanism for interactions with BSA-coated and uncoated beads, which is fundamentally different than what we observed for S-RBD-coated nanospheres (2) that our bead assay is sensitive to small changes in bead binding, and (3) we can focus on the BSA-coated beads going forward. Calu-3 cells showed a significantly more interaction with S-RBD-coated nanospheres over BSA-coated nanospheres at both temperatures, while the opposite was true for control fibroblasts. These results support a ligand-receptor-mediated interaction, in this case between the S-RBD domain and the ACE-2 receptor, for the Calu-3 cells.⁴⁷ Since the control fibroblasts lack the ACE-2 receptor on their surface, there is no receptor-mediated binding of S-RBD-coated nanospheres, and alternative binding processes occur at a significantly lower rate. Nanoparticles coated with serum proteins participate in cellular uptake mediated through energy-driven, temperature-dependent phagocytic and endocytic pathways, as observed previously.^49–51 As an additional control experiment, for the specificity of the S-RBD-coated nanospheres to ACE-2, we performed a competitive binding experiment of the coated beads against Calu-3 cells preincubated with soluble, full-length extracellular S1+S2 subunits of the S protein. The S1+S2 strongly interacts with ACE-2 and should block the receptor before the cells encounter S-RBD-coated beads. Blocking the viral binding receptor on host mammalian cells by antibody or receptor inhibitor peptides has shown to strongly limit the interaction, binding, and uptake of virus and virus mimicking particles.^52–54 Calu-3 cells with a blocked ACE-2 receptor showed an approximately fourfold less interaction and retention of S-RBD-coated nanospheres compared to unblocked control Calu-3 cells on par with that of BSA-coated nanospheres [Figs. 4(a) and 5] on control Calu-3 cells. Taken together, our results demonstrate that S-RBD-coated viral mimicking nanospheres interact specifically with Calu-3 cells via ACE-2 receptor-mediated interactions, which does not show differences at 33 or 37 °C.

FIG. 4.

Effect of temperature on receptor-mediated and receptor-independent binding of nanospheres. (a) S-RBD- and BSA-coated nanosphere binding to Calu-3 at 33 and 37 °C. (b) S-RBD- and BSA-coated nanosphere binding to fibroblasts at 33 and 37 °C. Statistically significant differences (p < 0.05) compared to S-RBD nanospheres are indicated by “*” and to BSA by “♦,” respectively.

FIG. 5.

Effect of ACE-2 receptor blocking on specific interaction with S-RBD-coated nanospheres. (a) Confocal micrographs showing binding ability for S-RBD-coated nanospheres (green) with unblocked and blocked ACE-2 receptors on Calu-3 cells (DRAQ5, red) at 37 °C (scale bar: 20 μm). (b) Quantification of S-RBD coated nanosphere binding to unblocked and blocked ACE-2 receptors on Calu-3 cells at 37 °C. Statistically significant differences (p < 0.05) compared to unblocked ACE-2 receptor Calu-3 are indicated by *.

C. ACE-2 receptor expression and distribution as a function of temperature

Since ACE-2 specifically binds S-RBD coated nanospheres on the Calu-3 cells, we further quantified the expression and distribution of the ACE-2 receptor on Calu-3 cells at 33 and 37 °C. Temperature fluctuations can cause modulation in the cell surface receptor number and ligand-receptor binding efficiency.^55–57 Immunofluorescence images of Calu-3 cells presented a high expression of the ACE-2 receptor; NIH-3T3 cells were already shown to express undetectable ACE-2 (Fig. S2).⁶⁵ As noted above, ACE-2 receptors localized largely at the periphery of Calu-3 cell clusters [Fig. 6(a)], which is characteristic to lung epithelial cells.⁵⁸ High expression of the ACE-2 receptor at the cell periphery also resulted in significant enhancement of S-RBD-coated nanospheres at the periphery [Fig. 6(b)].

FIG. 6.

Expression and distribution of the ACE-2 receptor on the Calu-3 cell surface at different temperatures. (a) Confocal micrographs of ACE-2 antibody (primary conjugate to Alexa647, blue) stained Calu-3 cells (scale bar: 100 μm). (b) Confocal micrographs of S-RBD-coated nanospheres (green) bound to Calu-3 (DRAQ5, red) (scale bar: 20 μm). (c) Mean fluorescence intensity measurement of the ACE-2 receptor on the Calu-3 surface at 33 and 37 °C, respectively.

The mean fluorescence intensity of the ACE-2 receptor on Calu-3 cells showed no statistical difference at 33 °C compared to 37 °C [Fig. 6(c)], corroborating our measurements showing no significant change in the interaction of viral mimic nanospheres with Calu-3 cells at these temperatures (Fig. 3). We note that NIH-3T3 cells showed no noticeable ACE-2 expression at any experimental temperature [Fig. S3(a)].⁶⁵ Consequently, our results using viral mimic nanospheres suggest that the season-induced physiological temperature fluctuations in the respiratory tract are not sufficient to strongly influence the interaction of SARS-CoV-2 with respiratory tract lung cells. This result indicates that the SARS-CoV-2 pandemic will not follow a seasonal cycle such as the common flu or other coronaviruses.

D. Thermal stability of viral mimicking beads and the spike protein receptor binding domain

SARS-CoV-2 spreads through respiratory droplets and aerosols. Several epidemiological and mathematical models have suggested that the virus's stability at different ambient temperatures may affect its transmission in the community.^12,13,59,60 Hirneisen et al. reported that high thermal exposure (>60 °C) could destabilize and inactivate viruses by the disruption of the protein-lipid envelope.⁶¹ However, the relationship between the viral stability and transmission of SARS-CoV-2 and the 4–40 °C ambient temperature range remains unclear. An increase in the ambient temperature can potentially lead to irreversible structural changes of viral surface proteins that may influence receptor binding and trafficking ability.³¹ We mimicked the change in the temperature of airborne viral particle exposure to different temperatures for different amounts of time before coming into contact with the human respiratory tract by exposing S-RBD-coated nanospheres to different temperatures and then adding them to Calu-3 cells at 37 °C. As suggested by the limited effect of temperature found in Fig. 3, we observed no significant effect of the preincubation temperature or temporal period used here [Fig. 7(a)]. Therefore, an increase in the ambient temperature from 4 to 40 °C did not affect the S-RBD-coated nanosphere interaction with the Calu-3 ACE-2 receptor.

FIG. 7.

Cellular binding and thermal stability of viral mimic S-RBD-coated nanospheres and the S-RBD soluble protein. (a) S-RBD-coated nanospheres exposed to different temperatures (4–40 °C) and time periods (1–24 h) show no significant changes in binding to Calu-3 cells after incubation at 37 °C for 48 h. (b) nDSF thermal stability profile of S-RBD from 15 to 40 °C.

We further measured the S-RBD’s thermal stability using nDSF, which provides ultra-high-resolution detection of protein unfolding based on intrinsic amino acid fluorescence as a function of increasing temperature. We found that the thermal stability of S-RBD in PBS was unaffected by changing the ambient temperature between 15 and 40 °C [Fig. 7(b)]. S-RBD showed a nearly constant fluorescence ratio of 350/330 nm in the temperate range of 15–40 °C, with only slight changes in the derivative trace near 40 °C.⁶² The thermal stability of S protein RBD on SARS-CoV-2 over this (15–40 °C) range may help it retain the robust binding capacity for the ACE-2 receptor despite environmental and respiratory tract ambient conditions. Further heating of the S-RBD protein above 40 °C resulted in an increase in the intrinsic amino acid fluorescence intensity first derivative reaching a maximum at 51 °C (Fig. S4),⁶⁵ indicating 3D structural alterations or denaturation of the S-RBD protein. This is consistent with results demonstrating high temperature (above 50 °C) inactivation of SARS-CoV-1 and 2 and other coronaviruses.^63,64 However, the physiological and environmental ambient temperature range (4–40 °C) appears to have a minimal effect on SARS-CoV-2 spike RBD protein stability and interaction with the ACE-2 receptor in cell experiments.

IV. CONCLUSION

We incubated ACE-2 receptor-positive lung epithelial cells at 33 and 37 °C with SARS-CoV-2 mimicking nanospheres to simulate the salient features of winter and summer seasonal temperature fluctuations in the host respiratory tract. Calu-3 cells interacted significantly more with S-RBD-coated nanospheres over BSA-coated or uncoated nanospheres, mediated through the ACE-2 surface receptor that supports the ligand-receptor-mediated interaction. Although there was an increase in BSA-coated nanospheres from ACE-2-independent binding, no difference was found between Calu-3 cells and S-RBD viral mimicking particle interactions at 33 or 37 °C. Furthermore, we mimicked the change in airborne viral particle exposure to different temperatures for different amounts of time before coming into contact with the human respiratory tract. Exposure of S-RBD-coated nanospheres to a range of temperatures (4–40 °C) and times (1–24 h) had no effect on ligand-receptor binding efficiency and nearly no change in the protein structure. Overall, our experimental results suggest that the ambient temperature fluctuation has little effect on SARS-CoV-2-ACE-2 interactions in agreement with the global epidemiological data reported so far on COVID-19.

AUTHORS’ CONTRIBUTIONS

S.K., A.P., S.C., and S.P. performed the experiments. S.K., A.P., N.N., D.S.W., A.N., and C.M.J. performed data analysis. S.K., A.P., and S.H.P. conceptualized, designed the study, and wrote the manuscript. All authors contributed edits to the manuscript and approved its final form.

S.K. and A.P. contributed equally to this work.

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding authors upon reasonable request.