ACE-2, TMPRSS2, and Neuropilin-1 Receptor Expression on Human Brain Astrocytes and Pericytes and SARS-CoV-2 Infection Kinetics

Angiotensin Converting Enzyme 2 (ACE-2), Transmembrane Serine Protease 2 (TMPRSS-2) and Neuropilin-1 cellular receptors support the entry of SARS-CoV-2 into susceptible human target cells and are characterized at the molecular level. Some evidence on the expression of entry receptors at mRNA and protein levels in brain cells is available, but co-expression of these receptors and confirmatory evidence on brain cells is lacking. SARS-CoV-2 infects some brain cell types, but infection susceptibility, multiple entry receptor density, and infection kinetics are rarely reported in specific brain cell types. Highly sensitive Taqman ddPCR, flow-cytometry and immunocytochemistry assays were used to quantitate the expression of ACE-2, TMPRSS-2 and Neuropilin-1 at mRNA and protein levels on human brain-extracted pericytes and astrocytes, which are an integral part of the Blood-Brain-Barrier (BBB). Astrocytes showed moderate ACE-2 (15.9 ± 1.3%, Mean ± SD, n = 2) and TMPRSS-2 (17.6%) positive cells, and in contrast show high Neuropilin-1 (56.4 ± 39.8%, n = 4) protein expression. Whereas pericytes showed variable ACE-2 (23.1 ± 20.7%, n = 2), Neuropilin-1 (30.3 ± 7.5%, n = 4) protein expression and higher TMPRSS-2 mRNA (667.2 ± 232.3, n = 3) expression. Co-expression of multiple entry receptors on astrocytes and pericytes allows entry of SARS-CoV-2 and progression of infection. Astrocytes showed roughly four-fold more virus in culture supernatants than pericytes. SARS-CoV-2 cellular entry receptor expression and “in vitro” viral kinetics in astrocytes and pericytes may improve our understanding of viral infection “in vivo”. In addition, this study may facilitate the development of novel strategies to counter the effects of SARS-CoV-2 and inhibit viral infection in brain tissues to prevent the spread and interference in neuronal functions.


Introduction
Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has emerged as a major pandemic impacting lives worldwide. Johns Hopkins University reported 676.6 million COVID-19 cases and over 6.88 million total deaths as of 10 March 2023, the last date of data collection [1]. Recent evidence 2 of 11 suggests that SARS-CoV-2 infects epithelial cells localized to lung and gut tissues and additionally some brain cells [2][3][4]. Neurological abnormalities with symptoms such as confusion and disorientation are observed in some SARS-CoV-2 infected individuals [5,6]. The severity of neurological diseases, including meningitis and encephalitis, increased in COVID-19 patients with convulsions [7]. A mechanistic understanding of neurological abnormalities during SARS-CoV-2 infection in COVID-19 patients was obtained from a meta-analysis of 153 patients who exhibited neuro-physiological disorders and later advanced to psychosis [8]. FDG-Positron-Emission tomography/computer tomography of frontal lobe and cerebellum regions following SARS-CoV-2 infection in COVID-19 patients showed acceleration of encephalitis due to lowering of metabolism [9].
Inflammation is a major driver of aberrant lung function and progressive SARS-CoV-2 infection and is frequently associated with abnormal brain function in COVID-19 patients [10]. More aggressive immune responses may cause damage to brain cells in COVID-19 patients and may cause Guillen-Barré syndrome (GBS), a disorder of the peripheral nerves, and even Miller-Fisher syndrome, a rare nerve disease [11][12][13][14][15]. Higher levels of secretory cytokines that regulate inflammation and autoantibodies are associated with central nervous system (CNS) complications in COVID-19 patients [16][17][18].
Effects of SARS-CoV-2 are not only associated with the respiratory system but also commonly associated with the gastrointestinal system [19], cardiovascular system [20], reproductive system [21,22], and neurological system [23,24]. Brain infection with SARS-CoV-2 has been associated with neurological disorders and cerebral artery stroke in COVID-19 patients [18,24,25]. SARS-CoV-2 was detected in the cerebrospinal fluid (CSF) of a 34-month-old child with encephalitis and in the brain tissue of an adult patient [23,26,27]. However, the effects of SARS-CoV-2 infection on the blood-brain barrier (BBB) and cell types located in this layer were rarely investigated.
Entry of coronaviruses into the brain compartment occurs through BBB transport passing the cerebral microvascular endothelial monolayer barrier [28][29][30]. Invasive entry of SARS-CoV-2 into brain tissue requires the contribution of molecular receptors on brain cells such as ACE-2 (Angiotensin Converting Enzyme 2) and Neuropilin-1 [31]. Transmembrane Protease 2 (TMPRSS-2) is a cellular protease that primes viral spike protein to attach to ACE-2 and facilitates SARS-CoV-2 entry into target cells [32]. Eminence capillaries and tanycytes of the hypothalamus express TMPRSS-2 and ACE-2 and facilitate SARS-CoV-2 infection of brain tissue [33]. The role of host cofactors or tight junctional proteins involved in SARS CoV-2 transport through the BBB remains unidentified. Human brain microvascular endothelial cells (hBMVEC) express low levels of ACE-2 and the recombinant SARS-CoV-2 spike protein enhances ACE-2 expression [34]. In contrast, human cortical astrocytes express ACE-2 and Neuropilin-1 [35,36]. However, Neuropilin-1 association with neuronal pericytes, which are an integral part of BBB, remains unknown.
In this study, we analyzed the expression of ACE-2, Neuropilin-1, and TMPRSS-2 both on the surface of astrocytes and pericytes and at the mRNA level, using highly specific Taqman ddPCR, flow-cytometry, and immunoassays. Our data provide evidence on ACE-2, TMPRSS-2 and Neuropilin-1 expression on primary pericytes and astrocytes and the potential involvement of these proteins in SARS-CoV-2 infection of these cell types.

Materials and Methods
Cells and Culture system-Primary human brain extracted astrocytes (CAT# 1800), pericytes (CAT# 1200) and human brain microvascular endothelial cells (hBMVECs, CAT# 1000) as well as human lymph node extracted endothelial cells (hLNECs, CAT# 2500) were purchased from ScienCell Research Laboratories (SCRL), Carlsbad, CA, USA. Cell culture media and the growth supplements required for culturing brain cells including astrocyte media (AMCAT, CAT# 1801) with growth supplement (AGS, CAT# 1852), pericyte media (PMCAT# 1201) with growth supplement (PGS, CAT# 1252) were purchased from the manufacturer, SCRL, Carlsbad, CA, USA. hBMVECs and hLNECs were cultured in an endothelial cell medium (ECM, CAT# 1001) with a growth supplement (ECGS, CAT# 1052) that was purchased from SCRL, Carlsbad, CA, USA. Fetal bovine serum (FBS, CAT# 0010) and penicillin/streptomycin solution (P/S) (CAT# 0503) media supplements were purchased from SCRL, Carlsbad, CA, USA. The cell thawing procedure was followed as described by the manufacturer. Per the experimental requirement, cells were grown in 25/75/150 cm 2 tissue culture flasks (TPP # 90076). Tissue culture flasks and plates used for culturing brain cells and hLNECs were pre-coated with bovine fibronectin (2 µg/mL) (SCRL, CAT# 8248). Confluent cell monolayers were passaged according to the manufacturer's instructions. In this, cells were washed with DPBS (Dulbeccos, CAT# 1960454) and detached from the surface via treatment with 0.25% trypsin (CAT# CC-5012, Lonza, USA) for 1-2 min. Following trypsin treatment, cells were mixed with FBS followed by adding a culture medium. Cells were pelleted by centrifugation at 1000 rpm for 5 min at room temperature (RT or 25 • C). Cell count was performed by mixing the cell suspension with trypan blue at a 1:1 ratio and 10 µL of suspension was loaded onto the slide provided by the manufacturer (Invitrogen, Eugene, OR, USA) and counted in Countess cell counter (Invitrogen, Chicago, IL, USA). Cells below 5 passages were used for the experiments to maintain the integrity of the phenotype of cells.
Immunostaining-Rabbit monoclonal ACE-2 primary antibody (Clone: SN0754, CAT# MA5-32307) and goat anti-rabbit Alexa Fluor-488 secondary antibody (CAT# A-11034) were purchased from Invitrogen. All staining steps were performed at RT. Astrocytes and pericytes were fixed with 4% paraformaldehyde for 10 min. and then incubated in a blocking buffer (10% goat serum in PBS) for 40 min. Cells were incubated with ACE-2 primary antibody (5 µg/mL) in 0.1% TWEEN-PBS or PBST (Phosphate Buffered Saline with Tween 20 (0.1%)) overnight at 4 • C. Following three PBST washes, cells were incubated with a secondary antibody at 1:50 dilution for 2-hrs at 37 • C. Then, the cells were washed with PBST three times and were air dried before mounting with DAPI solution (Prolong TM diamond antifade mount CAT# P36962, Thermofisher Scientific, Waltham, MA, USA). Sample images were obtained using a Zeiss confocal microscope at the University of Nebraska Medical Center Advanced Microscope Core Facility, University of Nebraska, Omaha, NE, USA.
RNA extraction and cDNA synthesis-Total RNA was extracted from 8-10 × 10 6 cells stored in RLT buffer (600 µL) according to the manufacturer's instructions (Qiagen, Germantown, MD, USA). The quantity of RNA was assessed with a nanodrop (NanoDrop ONE C , Thermofisher Scientific, Waltham, MA, USA). RNA (2 µg) was used as a source to synthesize cDNA using Superscript IV VILO Master Mix (Invitrogen, Chicago, IL, USA # 11754-050) according to the manufacturer's instructions.
ACE-2, TMPRSS-2, and Neuropilin-1 quantification by ddPCR-TaqMan ddPCR assays were designed to amplify ACE-2, TMPRSS-2, and Neuropilin-1 targets using primerprobe pairs listed in Table 1 and were purchased from IDT. In the ddPCR assay, 50 ng cDNA was mixed with a primer-probe mixture, and 12.5 µL of 2× ddPCR super mix (Biorad, Hercules, CA, USA) was added in a total of 25 µL reaction mixture according to the manufacturer's instruction. Samples were tested in four replicates (n=4) in the ddPCR assay in a 96 well plate (Biorad, Hercules, CA, USA # 12001925) and were sealed with Biorad Sealer PX1, and droplets were generated in AutoDG instrument (Biorad, Hercules, CA, USA). ddPCR assays were performed as described previously in our laboratory [37]. Table 1. List of Primer-Probes used to amplify ACE-2, TMPRSS2, and Neuropilin target genes.

SARS-CoV-2 infection of cells-Pericytes
, astrocytes, hBMVECs, and hLNECs were seeded at 0.25 million cells/well in a 12-well plate and following 24 hrs in an incubator (37 • C, 5% CO 2 ). Cells were then infected with SARS-CoV-2 isolate USA-WI1/2020 (BEI; cat# NR-52384) at 1 MOI (multiplicity of infection). One-hour post viral infection, cells were washed with PBS to remove the cell-free virus and supplied with fresh medium. Cells and cell culture supernatants were collected at 24-, 48-and 72-h post-infection.

Discussion
The contribution of ACE-2, TMPRSS-2, and Neuropilin-1 receptors present on astrocytes and pericytes for SARS-CoV-2 infection is unknown. In this present study, we analyzed the expression of ACE-2, TEMPRSS-2, and Neuropilin-1 on the surface of human brain-derived pericytes and astrocytes at the mRNA and protein level by employing gene expression, flow-cytometry, and imaging methodologies. The low amount of variable ACE-2 expression (23.1 ± 20.7%, Mean ± SD) observed on pericytes, is in agreement with an earlier report on mouse pericytes [38] including the variable expression observed in some healthy donors. This was further confirmed by Immunostaining ( Figure 1C). mRNA level ACE-2 expression using specific Taqman probes precisely quantitated the ACE-2 mRNA copies in ddPCR assays. ACE-2 mRNA copies were 2.3-fold (p value ≤ 0.004) higher in astrocytes compared to pericytes. Our multi-faceted approach provided substantial evidence of the expression of ACE-2 on brain astrocytes and pericytes suggesting possible involvement in SARS-CoV-2 infection.
Pericytes expressed a 3.4-fold higher number of TMPRSS-2 mRNA copies ( Figure 2B) compared to astrocytes. TMPRSS-2 mRNA expression in human brain cells (hBMVECs) mirrors the finding by Torices et al. [39]. Thus, our results not only affirm the earlier study findings but provide additional data on the mRNA level expression of TEMPRSS-2 in brain cells. Next, Davies et al. showed Neuropilin-1 expression in brain regions, including its presence in astrocytes. However, Neuropilin-1 mRNA and protein expression in pericytes at a cellular level remains unknown [36] and therefore, our results suggested the expression of Neuropilin-1 is present in pericytes. We observed high Neuropilin-1 mRNA copies in pericytes as compared with astrocytes. In contrast, lower Neuropilin-1 was expressed at the protein level. The reasons for these differences remain unclear, and future studies may focus on investigating these differences. Despite the differences in ddPCR

Discussion
The contribution of ACE-2, TMPRSS-2, and Neuropilin-1 receptors present on astrocytes and pericytes for SARS-CoV-2 infection is unknown. In this present study, we analyzed the expression of ACE-2, TEMPRSS-2, and Neuropilin-1 on the surface of human brain-derived pericytes and astrocytes at the mRNA and protein level by employing gene expression, flow-cytometry, and imaging methodologies. The low amount of variable ACE-2 expression (23.1 ± 20.7%, Mean ± SD) observed on pericytes, is in agreement with an earlier report on mouse pericytes [38] including the variable expression observed in some healthy donors. This was further confirmed by Immunostaining ( Figure 1C). mRNA level ACE-2 expression using specific Taqman probes precisely quantitated the ACE-2 mRNA copies in ddPCR assays. ACE-2 mRNA copies were 2.3-fold (p value ≤ 0.004) higher in astrocytes compared to pericytes. Our multi-faceted approach provided substantial evidence of the expression of ACE-2 on brain astrocytes and pericytes suggesting possible involvement in SARS-CoV-2 infection.
Pericytes expressed a 3.4-fold higher number of TMPRSS-2 mRNA copies ( Figure 2B) compared to astrocytes. TMPRSS-2 mRNA expression in human brain cells (hBMVECs) mirrors the finding by Torices et al. [39]. Thus, our results not only affirm the earlier study findings but provide additional data on the mRNA level expression of TEMPRSS-2 in brain cells. Next, Davies et al. showed Neuropilin-1 expression in brain regions, including its presence in astrocytes. However, Neuropilin-1 mRNA and protein expression in pericytes at a cellular level remains unknown [36] and therefore, our results suggested the expression of Neuropilin-1 is present in pericytes. We observed high Neuropilin-1 mRNA copies in pericytes as compared with astrocytes. In contrast, lower Neuropilin-1 was expressed at the protein level. The reasons for these differences remain unclear, and future studies may focus on investigating these differences. Despite the differences in ddPCR and flow cytometry results, the data confirms the Neuropilin-1 expression in astrocytes and pericytes, suggesting the availability of a potential candidate receptor for SARS-CoV-2 on brain astrocytes and pericytes.
To understand the susceptibility and SARS-CoV-2 infection kinetics in primary astrocytes and pericytes, both cell types were infected with 1 MOI of SARS-CoV-2. Results indicated that pericytes were infected with SARS-CoV-2 but were still variable among different donors ( Figure 5). All four-donor derived pericytes had similar infection kinetics reaching maximum viral RNA copies at 24 h post SARS-CoV-2 infection ( Figure 6). The trends of infection in these cells were different from standard cell lines such as Vero, where infection usually peaks at 72 h time point then declines later time points. However, based on SARS-CoV-2 viral RNA copies data, astrocytes were infected, and viral RNA copies are higher than pericytes following infection with the same amount of input virus ( Figure 6).
In conclusion, this report provides evidence of the presence of ACE-2, TMPRSS-2, and Neuropilin-1 proteins on primary pericytes and astrocytes and has demonstrated infection of SARS-CoV-2 in these cells. However, detailed time course studies are still needed to fully understand whether SARS-CoV-2 fully replicates in these cells. One limitation of our study is the lack of complementation or knockout of ACE-2, TMPRSS2, and Neuropilin-1 expression either at the mRNA level or blocking their expression on their surface. These studies are critical and may provide direct involvement of each of these receptors in SARS-CoV-2 infection in astrocytes and pericytes. Our data show that ACE-2, TMPRSS-2, and Neuropilin-1 expression on brain cells and potential contribution to SARS-CoV-2 infection. Considering the neurological symptoms and long COVID-19related CNS complications observed following some COVID-19 infections, identifying the cells susceptible to SARS-CoV-2 infection and the contributing cellular receptors or factors that increase their susceptibility to SARS-CoV-2 infection has become highly important. Our study findings provide insight into the neurotropism of SARS-CoV-2. "In vivo" brain cell infection kinetics is still under investigation, and our future studies will focus on investigating these details and developing strategies to counter the effects of SARS-CoV-2 infection entry, blocking productive infection of pericytes to prevent passage through the BBB. However, the findings of the present study are helpful to provide a primary platform for future studies targeting novel strategies to inhibit SARS-CoV-2 entry into brain tissues.