Old Paper (Pathogenesis): Inhibition of SARS Pseudovirus Cell Entry by Lactoferrin Binding to Heparan Sulfate Proteoglycans

Summary

In this study, Jianshe Lang et al. analyze two aspects of SARS-CoV. On the one hand, they analyze Lactoferrin (LF) as a potential SARS-CoV inhibitor. On the other hand, they introduce a novel binding mechanism, at the time, in that SARS-CoV cell binding is enhanced by Heparan Sulfate Proteoglycans (HSPG) receptor binding before functional cell entry receptors are employed (such as ACE2).

Hypotheses

Based on the previous knowledge that LF might play an important physiological role in the immune response to SARS-CoV disease due to its 150-fold elevated gene expression in SARS-CoV patients, they hypothesize that LF's well known antiviral effects could also be therapeutic with SARS-CoV.
They further hypothesize that HSPG receptors play an essential role in SARS-CoV binding to host cell surfaces as it has been shown for other viruses [10.1002/med.1026].

Background

At the time, exogenous heparin and LF were already known as HSPG binders [PMID2246253] [10.1161/01.atv.14.12.2025]. This study further verified the colocalization of LF with HSPG receptors. ACE2 was considered as the primary functional entry receptor.

Methods

A S protein-encoded SARS pseudovirus infection array was used to evaluate the infection ability in the absence or presence of LF or heparin. Transfected HEK293T cells were applied on HEK293E/ACE2-Myc, Vero E6, or Caco-2 target cells. A pseudovirus mock without spike protein was used as a negative control. BSA was used as base control. Different LF concentrations of 0.43 (or 0.34), 1.3, 4, and 12 µM were used to verify concentration-dependency. Viral replication was traced by the encoded GFP protein signaling. Fluorescence microscopy and flow cytometry were used for measurement.
A western blotting infection assay was used with HEK293E/ACE2-Myc target cells in the absence or presence of LF in the same concentrations as mentioned above.
An S1190-Fc binding assay should verify if LF prevents virus-cell interaction directly or indirectly. S1190-Fc is a recombinant SARS S protein (i.e., S1190) with an IgG Fc fused to its C terminus. LF-preincubated HEK293E/ACE2-Myc cells were mixed with S1190-Fc protein. Then, Fc antibodies were used to measure MFI via flow cytometry. Unbound FC was used as a negative control. The experiment was repeated with different concentrations of heparin (i.e., 1, 3, and 10 µM) to show the competitive effect of heparin and LF.
ELISA was used to verify if the cell fusion was mediated by ACE2-Myc directly. Therefore, ACE2-Myc was harvested from the cells. Direct protein binding of ACE2-Myc with S1190-Fc was measured with or without preincubation of 20 µM LF (i.e. twice the maximal concentration of the MFI experiment). Fc was used as a negative control.
Confocal microscopy was used to verify the colocalization of LF and heparin with HSPGs.
To further verify the key role of HSPGs during virus-cell binding, the pseudovirus and S1190 assay experiments were extended by heparinase I treatment to remove the heparan sulfate from the cell surface. The same was repeated by Chondroitinase ABC treatment to remove chondroitin sulfate
LF, heparin, and Heparinase treatment experiments were repeated with pseudovirus (i.e, SARS-CoV, and VSV-G) assays of Vero E6, Caco-2, and HEK293E/ACE2-Myc cell lines.

Accuracy

Mean values of triplicate samples were used. Standard deviations and p values are given.

Results

The pseudovirus infection assays and WB assays showed a clear LF-mediated concentration-dependent inhibition on S protein-mediated viral infection ability. IC50 was ca. 0.7 µM. The mock virus control verified that the inhibition targeted the S protein. The S1190 assays showed that LF inhibited S protein-cell fusion, but the inhibition was not mediated directly by S protein-ACE2-Myc binding. It further showed that heparin neutralizes the inhibitory effect of LF. Heparin showed weaker infection inhibition despite its stronger competitive binding affinity when compared to LF. Microscopy showed the colocalization of LF and heparin on the cell membrane surface. The authors depict a figure (Fig. 3B) that supports the hypothesis that heparin competes with LF for HSPG receptor binding. Heparin sulfate degradation via Heparinase I could inhibit S1190-Fc binding and pseudovirus infection. The single dosage they used led to ca. halved infection ability. That was consistent with the curve of heparin inhibition on infection ability, indicating a lower limit can be reached in the 0.4 to 0.6 range. Chondroitin sulfate degradation did not show a significant effect. A comparison of LF, heparin, and Heparinase on SARS pseudovirus infection ability showed different results in different cell lines (i.e., Vero E6, and Caco-2). Overall, LF showed the best concentration effectivity. VSV-G infection could also be inhibited by these three agents. However, LF did show less effect than heparin and relatively less effect than heparinase when compared to the SARS pseudovirus results.

Comments

In 2011, the authors still assumed that ACE2 is the only cell surface entry receptor. There are inconsistencies in the methods and results section. It’s unclear if they used 0.43 or 0.34 µM LF for their pseudovirus assays. Not all experiments are mentioned in the methods section. Some are also introduced in the figure descriptions. The flow cytometry figure is incomplete (Fig. 1G). The confocal microscopy only verified the colocalization of LF and heparin on cell membrane surface but no ACE2-deficient cell line or pure HSPG binding assay was used as a control to prove that the interaction is purely mediated HSPGs. There is no clear explanation of how the inhibition of heparin on LF-HSPG binding was measured (Fig. 3B). Enzymatic experiments lacked different concentrations in the design. It’s unclear if a knockout effect could be reached, which seems unlikely, though, since transmembrane entry proteins aren’t blocked. Still, it would be helpful to quantify the maximum impact and the relative expression of free HSPG necessary for effective virus adsorption. They used heparin, however, to find the limits of HSPGs involvement in cell infection rate that indicate at least a two-fold virus infection due to HSPG adsorption, consistent with their enzymatic results. The finding that the three tested treatments (i.e., LF, heparin, and heparinase) showed inverse concentration correlations for SARS-CoV and VSV-G indicates that either LF or heparin could have a secondary inhibitory mechanism in either SARS-CoV or VSV-G. These comparative experiments lack different concentrations in their design.

If SARS-CoV-2 utilizes the same S protein adsorption mechanism to support cell binding, this will be important for many other studies. It seems unlikely that HSPG adsorption only supports ACE2-mediated cell entry. The authors of this study speculate that the high expression of HSPGs but low binding affinity first lets SARS-CoV anchor to HSPG and then transfer to cell entry proteins that express a higher binding affinity than HSPG, consistent with the knowledge from other viruses [10.1002/med.1026]. If SARS-CoV-2 encodes a similar “viral surfing” mechanism, this finding would also be significant for in vitro study designs since the HSPG expression of cell lines can have a substantial impact on the measurement results.

All in all, HSPG plays an essential role in the pathogenesis of SARS-CoV. Protein interaction simulation could verify if SARS-CoV-2 has a similar binding affinity to heparan sulfate for a start.

greenelab / covid19-review