Blockage of the cellular waste disposal system can prevent cell death
in SARS-CoV-2 infected cells
We are currently dealing with a viral pandemic called COVID-19, caused by the SARS-CoV-2 virus.
Despite global efforts to deal with this virus through mitigation, physical distancing, and common sense measures,
many countries are still suffering from an increasing number of cases, hospitalizations and unfortunately, deaths.
At NCATS, we are utilizing our strengths in assay development and translational science research to rapidly establish a toolkit
for screening small molecules and biologics that may prevent viral infection. This toolkit will also allow us to understand the
biology of this virus, and how it interacts with human cells.
In the course of our research, we have reported that SARS-CoV-2 may be particularly susceptible to small molecules
that interrupt the host cell’s waste disposal system, a process known as autophagy (1). Healthy cells experience some wear and tear on
their components over time, and recycle waste materials in lysosomes. This process involves many different proteins that work together
to capture and compartmentalize cellular waste - as well as invading pathogens like bacteria and viruses - and degrade them before they
can become a problem. Persistent blockage of cellular autophagy causes real problems for cellular health and function and, in some cases,
can lead to cell death.
Coronaviruses such as SARS-CoV-2 have been found to hijack the autophagy machinery of host cells, namely the compartments called autophagosomes,
endosomes, and lysosomes. As a result, the host cell cannot dispose of the invader, while the virus uses these compartments as a platform for
its own replication and survival. Figure 1, from our recently published preprint, illustrates how a virus can use the cell’s own autophagy machinery
to enter and replicate.
Figure 1. Illustration of autophagy inhibitors and their blockade of viral infection.
(A) Healthy cells have normal autophagic flux
and the endocytic pathway is functional.
(B) Autophagy inhibitor treatment in healthy cells causes a blockade of normal fusion processes
and a buildup of endosomes and autophagosomes.
(C) In healthy cells, viral infection through endocytosis leads to the release
of viral RNA after endosome lysosome fusion. Similarly, autophagy of viral particles may result in formation of viral autophagosomes
but lysosome fusion would be blocked by the virus (orange X). Dotted arrow indicates a possible, but unverified event of viral RNA release
(D) Autophagy inhibitors can block steps (red Xs) within the viral life cycle including at the early steps of endocytosis,
the fusion of endosomes with the lysosome, to prevent the release of viral RNA and subsequent cell death.
Paradoxically, we found that treatment with small molecule drugs that inhibit the cell’s normal autophagic process may actually
prevent the virus from replicating within cells, leading to reduced host cell death. In conjunction with our collaborators at Southern Research Institute (SRI),
who have worked tirelessly in their biosafety level 3 facility, we tested a panel of autophagy inhibitors in the context of live virus infection.
We identified several compounds that were able to block cell death induced by SARS-CoV-2 infection (known as the cytopathic effect)
in the Vero-E6 African Green monkey kidney cell line. These particular cells express high levels of the ACE2 receptor that the virus uses for cell entry.
The assay we used measures host cell (Vero-E6) viability after 72 hours of SARS-CoV-2 infection, so small molecules capable of reducing the amount
of cell death may be adversely affecting the virus in some way.
We observed that Vero-E6 cells treated with autophagy inhibitors (including chloroquine and hydroxychloroquine) led to
a considerable reduction in host cell death. Our hypothesis is that by blocking cell autophagy, we were able to prevent the virus from replicating
within the host cells, leading to reduced cell death. We confirmed these compounds’ inhibitory effects on autophagy by screening three other human cell lines
for markers of autophagy, including the protein LC3B, and by using a dye that accumulates in acidic cellular compartments.
We showed that the potencies of these compounds as inhibitors of autophagy matched up closely with the potencies observed in the prevention
of SARS-CoV-2-mediated cell death (Figures 2 & 3).
Figure 2. Autophagy inhibition assay using LC3B immunostaining in Vero-E6 cells. Image montage of DMSO, CQ, HCQ, clomipramine,
mefloquine, ROC-325, and hycanthone (L to R) stained with Hoechst 33342 (cyan) and LC3B (magenta). Scale bar, 25 µm.
Figure 3. Autophagy inhibition assay using LysoTracker Deep Red staining in Vero-E6 cells.
Image montage of DMSO, CQ, HCQ, clomipramine, mefloquine, ROC-325, and hycanthone (L to R) stained with Hoechst 33342 (cyan), HCS Cell Mask Green (yellow),
and LysoTracker Deep Red (magenta). Scale bar, 25 µm.
Encouragingly, some of these drugs are already FDA-approved, providing the potential for repurposing to fight COVID-19.
Autophagy inhibitors have been shown to demonstrate antiviral activity, including Zika and Ebola, and while much work remains to better characterize autophagy
in the context of viral infection, this work has highlighted some promising leads and target pathways for potential use in the fight against SARS-CoV-2.
Drug repurposing to block SARS-CoV-2 infection?
Here at NCATS we’ve been actively developing assays to address many scientific questions about COVID-19.
One such question is whether we can block or impair the ability of SARS-CoV-2 to infect cells, either through drug repurposing or
developing new therapeutics. First, let’s discuss how this virus infects host cells.
We briefly covered this in a previous post, but we’ll recap the first step in SARS-CoV-2 infection here.
The outer surface of the SARS-CoV-2 viral capsid is coated with Spike proteins, which contain receptor binding domains (RBD)
that bind to the ACE2 receptor protein on the surface of human cells (Figure 1). This is the first step in viral entry.
Blocking the interaction between the SARS-CoV-2 Spike protein RBD and human ACE2 is a potential therapeutic approach for treating COVID-19,
because if the virus can’t bind to the surface of human cells, it can’t enter and be replicated.
Figure 1: Schematic showing stages of SARS-CoV-2 viral entry
In order to test Spike/ACE2 disruption as a potential therapeutic approach for COVID-19,
we first needed to develop an assay capable of measuring interactions between the Spike RBD and human ACE2. We used a technique called
which was developed by PerkinElmer, to measure RBD interactions with ACE2. AlphaLISA is a proximity assay that uses a pair of donor and acceptor
beads to measure interaction (or disruption) of two tagged proteins/targets of interest. When the donor bead is excited by light at 680 nm it
converts ambient oxygen to singlet oxygen. If an acceptor bead is in close proximity to the singlet oxygen, it emits a luminescent signal at 615 nm
(and when the two proteins are apart, there is no signal). More information about the AlphaLISA technology and related assays can be found in this
For this assay we used Protein A-labeled Acceptor beads, which recognize Fc tags, and Streptavidin-labeled Donor beads,
which recognize biotin. We used a recombinant SARS-CoV-2 Spike protein RBD fused to a Fc tag (RBD-Fc, Sino Biological, Catalog # 40592-V02H)
and recombinant soluble human ACE2 fused to a biotinylated Avi tag (ACE2-Avi, Acro Biosystems, Catalog # AC2-H82E6) to model a simplified Spike
RBD-ACE2 interaction system. The assay we designed can be simplified into two steps:
- 1) Mix RBD-Fc and ACE2-Avi together to allow them to interact
- 2) Add Protein A Acceptor beads (recognizes RBD-Fc) and Streptavidin Donor beads (recognizes ACE2-Avi)
- 3) Excite at 680 nm and measure luminescence at 615 nm
By measuring the luminescent signal, we can infer the degree to which the recombinant RBD and ACE2 are interacting.
A stronger luminescent signal indicates a higher degree of interaction between SARS-CoV-2 RBD and human ACE2. A weaker luminescent signal
indicates the RBD-ACE2 interaction is being disrupted. From this point we can add test agents to our RBD-ACE2 mix and determine if any disrupt
the RBD-ACE2 interaction.
It is important to note that every assay has limitations, and in the case of AlphaLISA, compounds that interfere with the
AlphaLISA chemistry can cause a decrease in luminescent signal, mimicking the effect seen with disruption of the proteins; these false positives
are known as “artifacts”. To filter these out, we use a control AlphaLISA assay called TruHits, which uses a set of acceptor and donor beads that
directly interact, without any SARS-CoV-2 RBD or ACE2 added. Lower luminescence in a TruHits assay indicates an agent is non-specifically interfering
with the AlphaLISA technology, rather than the RBD:ACE2 interaction we’re interested in. Figure 2 demonstrates how false positives like biotin
appear in this assay pair, compared to a potential true positive like GSK744. More information about types of assay artifacts and their impact on
high-throughput screening can be found in this
NCATS Assay Guidance Manual chapter
Figure 2. Top: Dose-response curve for SARS-CoV-2 RBD binding to ACE2 AlphaLISA assay (magenta) and TruHits counter assay
(blue) in the presence of biotin, a compound known to disrupt the AlphaLISA technology
Bottom: Dose response curve for SARS-CoV-2 RBD binding to ACE2 AlphaLISA assay (magenta) and TruHits counter assay (blue)
in the presence of GSK744 (cabotegravir)
So far, we’ve used this assay to screen all FDA-approved drugs (the
NCATS Pharmaceutical Collection
as well as a collection of anti-infectious agents for potency in disrupting RBD-ACE2 interactions (3,384 compounds in total);
these entire datasets were made publicly available here (
) and here
) immediately following data processing.
Most of the hits we identified also showed equivalent activity in the TruHits assay, suggesting a large portion were artifacts.
Though the remaining candidates had weaker activity, they still warrant further investigation. Importantly, we are also using this assay to
validate antibodies and proteins that can prevent Spike RBD from binding to ACE2 to select the best candidates for further testing.
Along with screening data, we aim to share interesting data and observations we make
around COVID19 science – some large and some small. Many of us have individual scientific questions– a single experiment to answer a
question that doesnʼt fit into a larger research narrative. Data portals and open lab books are one way to share interesting data.
For example, weʼve been working with a range of SARS-CoV-2 spike protein constructs in a number
of assay development projects. This is because we know that the SARS-CoV-2 spike protein has a receptor binding domain (RBD)
that binds to human/host ACE2 on the surface of cells, and is the first step for viral entry. ACE2 (angiotensin-converting enzyme),
as the name suggests, is an enzyme that is a component of the renin-angiotensin system. Itʼs job is to convert angtiotensin II
(inflammatory mediator) into angiotensin 1-7 (considered to be anti inflammatory),
and under normal circumstances ACE2 plays a major anti-inflammatory role.
Following binding by SARS-CoV-2, ACE2 internalizes, carrying the virus with it. In doing so,
ACE2 enzymatic activity is lost at the cell surface. But we wondered whether spike protein binding to ACE2 directly
inhibited activity, or whether in fact it might activate ACE2.
To test this, we sent a recombinant spike protein (SARS-CoV-2 Spike S1-His, Sino Biological, Cat 40591-V08H) to Reaction Biology
(Malvern, PA) and asked them to test its effect on ACE2
biochemical activity. RB scientists use a standard protease assay protocol, where a fluorogenic
peptide substrate (MCA-YVADAPK(Dnp)-OH, MCA: 7-Methoxycoumarin-4-yl)acetyl) is added with enzyme, and as peptide substrate is cleaved,
a fluorophore is released, and fluorescence measured every 5 min for 2 h. As such, inhibitors would reduce kinetic activity of enzyme,
resulting in lower fluorescent signal. The positive control small molecule ACE2 inhibitor is MLN-4760 (developed by Millennium Pharmaceuticals,
and you can read more about it here https://www.ncbi.nlm.nih.gov/pubmed/14754895). As expected, MLN-4760 potently inhibited ACE2 biochemical
activity, confirming the enzyme if functioning and the assay is 'inhibitable'.
Left: Dose-response curve for ACE2 biochemical activity in the presence of the positive control
When we looked at the effect of SARS-CoV-2 Spike protein, we found that it did NOT inhibit ACE2 biochemical activity.
In fact, the SARS-CoV-2 Spike S1-His protein appears to activate ACE2 activity at higher concentrations (high nM and 1 uM).
We were certainly surprised. Are these concentrations of spike protein physiologically relevant? Possibly in extracellular
space in the lung epithelium. Could this be relevant given that ACE2 is internalized, or is there a consequence of activating
ACE2 in the intracellular space? We donʼt know. But we hope this piece of data stimulates discussion and thought about the
impact of SARS-CoV-2 spike protein and its interaction with ACE2 - a critical host enzyme for SARS-CoV-2.
Left: Raw activity-over-time fluorescence data for the ACE2 biochemical assay.
Concentrations of SARS-CoV-2 Spike S1-His added are shown in legend. Right: Dose-response curve for ACE2 biochemical activity in the presence of SARS-CoV-2 Spike S1-His protein.
Credit to Ying Fu and Alex Renn at NCATS,
and Haiching Ma and Kayleigh McGovern-Gooch at Reaction Biology for design and execution of this work.