key: cord-0751200-4f90ihpv authors: Hirunpattarasilp, Chanawee; James, Gregory; Freitas, Felipe; Sethi, Huma; Kittler, Josef T.; Huo, Jiandong; Owens, Raymond J.; Attwell, David title: SARS-CoV-2 binding to ACE2 triggers pericyte-mediated angiotensin-evoked cerebral capillary constriction date: 2021-04-01 journal: bioRxiv DOI: 10.1101/2021.04.01.438122 sha: 9b997a6b531fdf98be7570e145347baf5dda3010 doc_id: 751200 cord_uid: 4f90ihpv The SARS-CoV-2 receptor, ACE2, is found on pericytes, contractile cells enwrapping capillaries that regulate brain, heart and kidney blood flow. ACE2 converts vasoconstricting angiotensin II into vasodilating angiotensin-(1-7). In brain slices from hamster, which has an ACE2 sequence similar to human ACE2, angiotensin II alone evoked only a small capillary constriction, but evoked a large pericyte-mediated capillary constriction generated by AT1 receptors in the presence of the SARS-CoV-2 receptor binding domain (RBD). The effect of the RBD was mimicked by blocking ACE2. A mutated non-binding RBD did not potentiate constriction. A similar RBD-potentiated capillary constriction occurred in human cortical slices. This constriction reflects an RBD-induced decrease in the conversion of angiotensin II to angiotensin-(1-7). The clinically-used drug losartan inhibited the RBD-potentiated constriction. Thus AT1 receptor blockers could be protective in SARS-CoV-2 infection by reducing pericyte-mediated blood flow reductions in the brain, and perhaps the heart and kidney. Despite the primary site of infection by SARS-CoV-2 being the respiratory tract, the virus evokes dysfunction of many other organs, including the brain, heart and kidney: 36% of hospitalised patients show neurological symptoms 1 , 20% develop myocardial injury 2 and 41% experience acute kidney injury 3 . This could reflect either a spread of virus via the blood 4 , or the effects of inflammatory mediators released from the lungs. These effects may contribute to "long Covid", in which clouding of thought and physical exhaustion extend for months after the initial infection. The receptor 5, 6 for SARS-CoV-2 is the enzyme ACE2 (part of the renin-angiotensin system that regulates blood pressure), which converts 7 vasoconstricting angiotensin II into vasodilating angiotensin- (1) (2) (3) (4) (5) (6) (7) . The Spike protein of SARS-CoV-2 binds to ACE2 to trigger its endocytosis 6 . For the closely-related SARS virus, binding of only the receptor binding domain (RBD) is sufficient 8 to evoke internalisation of ACE2. In the heart 9 and brain 10 the main cells expressing ACE2 are pericytes enwrapping capillaries (and some endothelial cells), and pancreas and lung pericytes also express ACE2 11, 12 . Pericytes express contractile proteins and in pathological conditions have been shown to decrease blood flow in the brain 13, 14 , heart 15 and kidney 16 . Interestingly, a decrease of blood flow has been reported for SARS-CoV-2 infection in the brain 17 and kidney 18 . This could be due to pericyte dysfunction caused by SARS-CoV-2 reducing the activity of ACE2, either by occluding its binding site for angiotensin II (although this is thought not to occur for the related SARS virus 19 ) or by promoting internalisation of the enzyme 6, 8 . In the presence of angiotensin II (either renally-derived and reaching the brain parenchyma via a compromised blood-brain barrier, or generated by the brain's own renin-angiotensin system 20 To study the effect of the SARS-CoV-2 RBD on cerebral capillary pericyte function, we employed live imaging 21 of brain slices from Syrian golden hamsters. Hamsters have an ACE2 Spike sequence which is more similar to that in humans than is the rat and mouse ACE2 Spike sequence 22 . In particular amino acid 353 in hamsters and humans is a lysine (K) rather than a histidine (H), and this is a key determinant 23 of how well coronaviruses bind to ACE2, making hamsters a good model for studying SARS-CoV-2 effects 22 . We assessed the location of ACE2 and contractile properties of pericytes in the cerebral microvasculature of the hamster, which have not been studied previously (Fig. 1 ). Immunohistochemistry (IHC) revealed ACE2 to be predominantly expressed in capillary pericytes expressing NG2 and PDGFR (Fig. 1a) . Quantification of overlap with the pericyte marker PDGFR (Fig. 1b) revealed ~75% co-localisation (Fig. 1c) , and comparison of expression in capillaries and arterioles showed that capillaries exhibited ~75% of the ACE2 expression (Fig. 1d) . These results are consistent with transcriptome and IHC data from mouse and human brain 10 and human heart 9 . As for brain pericytes in rats 24 , the thromboxane A 2 analogue U46619 (200 nM) evoked a pericyte-mediated capillary constriction and superimposed glutamate evoked a dilation (Fig. 1e ). Applying angiotensin II (150 nM) evoked a transient constriction, which was inhibited by the AT1 receptor blocker losartan (20 M, Fig. 1f ). Similar angiotensin-evoked pericytemediated capillary constriction has been reported in the kidney 25 and retina 26 . The transience of the constriction might reflect receptor desensitisation 27 at this relatively high angiotensin II concentration, or a delayed activation of Mas receptors after the angiotensin II is converted to angiotensin (1-7). Blocking either AT2 receptors (with 1 M PD123319) or Mas receptors (with 10 M A779) increased the angiotensin II evoked constriction (approximately 4.5-fold for MasR block, p<10 -4 , Fig. 1g-h) , consistent with the AT1R-mediated constriction being opposed by angiotensin II activating AT2 receptors, by angiotensin-(1-7) activating Mas receptors, or by activation of AT2/Mas heteromeric 28 receptors. Acute application of the RBD of Covid 19 (at 0.7 mg/l, or ~22.5 nM, which is approximately 5 times the EC 50 for binding 29 ) for up to 40 mins evoked a small and statistically insignificant reduction of capillary diameter at pericytes (Fig. 2a) . On applying a very high level of angiotensin II (2 M) in the absence of RBD, a transient constriction of capillaries at pericytes was observed (6.3±3.6% in 6 capillaries, not significantly different from the 7.5±1.6% observed using 150 nM angiotensin II in 9 capillaries in Fig. 1f, p=0 .73). However, if brain slices were exposed for 30 min to RBD (0.7 mg/l) before the same concentration of angiotensin II was applied together with the RBD, then the angiotensin II evoked a 5-fold larger constriction of 31.5±9.3% in 4 capillaries (significantly different to that seen in the absence of RBD, p=0.019, Fig. 2b ). The 30 min pre-exposure period was used in order to allow time for the large RBD molecule to diffuse into the slice, and was mimicked for the experiments without the RBD. This large constriction-potentiating effect of the RBD was not a non-specific effect on the pericytes' contractile apparatus, because the contractile response to U46619 (200 nM) was unaffected by the RBD (Fig. 2c) . The high concentration of angiotensin II used in Fig. 2b is probably unphysiological and evokes a transient response for reasons that are discussed above. We therefore switched to a lower angiotensin II concentration (50 nM, Fig. 2d ), which is more similar to levels found physiologically within the kidney 30,31 and heart 32 . In the presence of the RBD, the constricting response to angiotensin II was increased from an insignificant dilation of -4.5±3.0% to 7.8±3.6% (9 capillaries each, p=0.02), i.e. effectively a constriction of ~12% (from 100*{1 -(92.2%/104.5%)}). Using surface plasmon resonance to assess binding of RBD mutants to immobilised ACE2, we identified the Y489R mutation as reducing binding by ~94% (Fig. 2e) . Applying this mutated RBD (for which glycosylation of the protein is expected to be the same as for the normal RBD) had essentially no effect on the response to angiotensin II (Fig. 2d, f) . Thus, the potentiation of the angiotensin II response by the RBD is a result of it binding to ACE2. We hypothesised that the potentiating effect of the RBD on the response to angiotensin II reflects a decrease in the conversion by ACE2 of vasoconstricting angiotensin II into vasodilating angiotensin-(1-7). Such a decrease is expected if RBD binding promotes ACE2 internalisation 6, 8 or if it occludes the angiotensin II binding site. We therefore tested the effect of an ACE2 inhibitor (MLN-4760, 1 M 33 ) on the response to 50 nM angiotensin II. This closely mimicked the potentiating effect of the RBD, confirming that the RBD reduces effective ACE2 activity (Fig. 3a, b) . With a view to reducing SARS-CoV-2 evoked capillary constriction and any associated reduction of microvascular blood flow, we tested whether the AT1 receptor blocker losartan prevented the constricting effect of the RBD. Losartan completely blocked the angiotensin II evoked constriction seen in the presence of the RBD (Fig. 3c, d) . In human SARS-CoV-2 infection it has been suggested that one pathological mechanism is a loss of pericytes caused by viral infection reducing their viability or their interactions with endothelial cells 10 . In a transgenic model of pericyte loss (decreasing PDGFR signalling) it was found that endothelial cells upregulated von Willebrand Factor (vWF) production, and thus produced a pro-thrombotic state, which could explain the coagulopathy seen in SARS-CoV-2 patients 10 . However, exposing hamster brain slices to RBD (0.7 mg/l) for 3 hours, in the absence or presence of 50 nM angiotensin II, had no significant effect on pericyte death as assessed by propidium labelling (Fig. 4a) . Nevertheless, infection with the actual virus might have more profound effects on pericyte function or viability than does exposure to the RBD. To assess whether the potentiation of capillary constriction, characterised above in hamsters, also occurs in human capillaries, we employed brain slices made from live human brain tissue that was removed in the course of tumour removal surgery 14 . Consistent with the similar binding 22 of the SARS-CoV-2 RBD to human and hamster ACE2, we found that the RBD greatly potentiated the pericyte-mediated capillary constriction evoked by 50 nM angiotensin II (Fig. 4b-c ). SARS-CoV-2 binding would therefore be expected to decrease human cerebral blood flow assuming that, as in rodents, the largest resistance to flow within the brain parenchyma is provided by capillaries 34 . The data presented above are consistent with the scheme shown in Fig. 4d -e. Binding of the SARS-CoV-2 RBD to ACE2 in pericytes leads to a decrease in ACE2 activity, either as a result of ACE2 internalisation 6, 8 or due to occlusion of the angiotensin II binding site. This leads to an increase in the local concentration of vasoconstricting angiotensin II and a decrease in the concentration of vasodilating angiotensin-(1-7). The resulting activation of contraction via AT 1 receptors in capillary pericytes reduces capillary diameter locally by ~12% when 50 nM angiotensin II is present. As most of the vascular resistance within the brain is located in capillaries 34 , this could significantly reduce cerebral blood flow (as occurs following pericyte-mediated constriction after stroke and in Alzheimer's disease 13, 14 ) . Presumably the same mechanism could evoke a similar reduction of blood flow in other organs where pericytes express ACE2 and AT 1 receptors. From our data it is complicated to predict the magnitude of blood flow reduction expected. If a 12% diameter reduction occurred uniformly in the cerebral vasculature then, becoming stuck at narrow parts of the vessel, for example near constricted pericytes [36] [37] [38] , and this could also transform a small constriction into a much larger reduction of blood flow. In order for SARS-CoV-2 to evoke pericyte-mediated capillary constriction (or to cause pericyte dysfunction that upregulates vWF production 10 ) the virus would need to bind to the ACE2 that is located in pericytes located on the opposite side of the endothelial cell barrier from the blood. Access to pericytes in the brain parenchyma might occur via initial infection of the nasal mucosa and movement from there up the olfactory nerve into the brain 39, 40 . Alternatively, movement of the S1 part of the Spike protein across the blood-brain barrier by transcytosis has been reported 41 , and crossing the endothelial cell layer may also occur via infection of monocytes (which express ACE2 highly 42 and can cross endothelial cells), or via breakdown of the blood-brain barrier as a result of cytokines released as a result of lung inflammation 43 . The reduction of blood flow produced by pericyte-mediated capillary constriction, together with any upregulation of vWF that may occur 10 , will tend to promote clotting in the microvasculature. SARS-CoV-2 infection is associated with thrombus formation 44 in large vessels that can be imaged, but it seems possible that thrombi of microvascular origin 45 may add to this, and could perhaps even seed these larger clots. Together, capillary constriction and thrombus formation will reduce the energy supply to the brain and other organs, initiating deleterious changes that probably contribute to the long duration symptoms 46 Animals Brain slices (200-300 m thick) were made from the brains of Syrian golden hamsters (age 5-24 weeks) of both sexes, which were humanely killed (in accordance with UK and EU law) by cervical dislocation after being anaesthetised with isoflurane. In each slice only one pericyte was studied. The constriction evoked by angiotensin II in the presence of the RBD showed overlapping ranges of value for 2 female vessels and 9 male vessels (p=0.67). Imaging of pericyte mediated constriction Pericytes on cortical capillaries were identified visually as previously described (Fig. S1 of ref. 14) and imaged with a CCD camera as described 21 . Diameter was measured in ImageJ by drawing a line across the vessel between the inner walls of the endothelial cells. and RBD-Y489R_R, as well as RBD-Y489R_F and TTGneo_RBD_R; followed by joining the two resulted fragments with TTGneo_RBD_F and TTGneo_RBD_R. The gene carrying the Y489R mutation was then inserted into the vector pOPINTTGneo incorporating a C-terminal His6 tag by Infusion® cloning. The plasmid was sequenced to confirm that the mutation had been introduced successfully. Recombinant protein was transiently expressed in Expi293™ (ThermoFisher Scientific, UK) and purified from culture supernatants by an immobilised metal affinity using an automated protocol implemented on an ÄKTAxpress (GE Healthcare, UK ) followed by a Superdex 200 10/300GL column, using phosphate-buffered saline (PBS) pH 7.4 buffer. Recombinant RBD-WT and ACE2-Fc were produced as described 48 . Neurologic Manifestations of Hospitalized Patients With Coronavirus Disease Association of Cardiac Injury With Mortality in Hospitalized Patients With SARS-CoV-2 in Wuhan Risk Factors, Clinical Characteristics, and Prognosis of Acute Kidney Injury in Hospitalized SARS-CoV-2 Patients: A Retrospective Cohort Study Clinical and virological data of the first cases of SARS-CoV-2 in Europe: a case series SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor Characterization of spike glycoprotein of SARS-CoV-2 on virus entry and its immune cross-reactivity with SARS-CoV The ACE2/Angiotensin Endocytosis of the receptor-binding domain of SARS-CoV spike protein together with virus receptor ACE2 The ACE2 expression in human heart indicates new potential mechanism of heart injury among patients infected with SARS-CoV-2 Pericyte-specific vascular expression of SARS-CoV-2 receptor ACE2 -implications for microvascular inflammation and hypercoagulopathy in SARS-CoV-2 SARS-CoV-2 Receptor Angiotensin I-Converting Enzyme Type 2 (ACE2) Is Expressed in Human Pancreatic β-Cells and in the Data Descriptor: Single cell RNAseq of mouse brain and lung vascular and vessel-associated cell types Capillary pericytes regulate cerebral blood flow in health and disease Amyloid β oligomers constrict human capillaries in Alzheimer's disease via signaling to pericytes Capillary pericytes mediate coronary no-reflow after myocardial ischaemia Descending vasa recta pericytes mediate medullary no-reflow after renal ischaemia Neurovascular and perfusion imaging findings in coronavirus disease 2019: Case report and literature review Critically ILL SARS-CoV-2 Patients With Acute Kidney Injury Have Reduced Renal Blood Flow and Perfusion Despite Preserved Cardiac Function; A Case-Control Study Using Contrast Enhanced Ultrasound Structure of SARS coronavirus spike receptorbinding domain complexed with receptor Within the Brain: The Renin Angiotensin System Imaging pericytes and capillary diameter in brain slices and isolated retinae Simulation of the Clinical and Pathological Manifestations of Coronavirus Disease 2019 (SARS-CoV-2) in a Golden Syrian Hamster Model: Implications for Disease Pathogenesis and Transmissibility Molecular Determinants of Severe Acute Respiratory Syndrome Coronavirus Pathogenesis and Virulence in Young and Aged Mouse Models of Human Disease Astrocytes mediate neurovascular signaling to capillary pericytes but not to arterioles Pericyte regulation of renal medullary blood flow In situ observation of living pericytes in rat retinal capillaries Mechanisms and functions of AT 1 receptor internalization Dimerization of AT 2 and Mas Receptors in Control of Blood Pressure Characterization of the receptor-binding domain (RBD) of 2019 novel coronavirus: implication for development of RBD protein as a viral attachment inhibitor and vaccine Endogenous angiotensin concentrations in specific intrarenal compartmentsof the rat Renal interstitial fluid angiotensin. Modulation by anesthesia, epinephrine, sodium depletion, and renin inhibition Compartmentalization of angiotensin II generation in the dog heart. Evidence for independent mechanisms in intravascular and interstitial spaces Murine recombinant angiotensin-converting enzyme 2: effect on angiotensin II-dependent hypertension and distinctive angiotensin-converting enzyme 2 inhibitor characteristics on rodent and human angiotensin-converting enzyme 2 The capillary bed offers the largest hemodynamic resistance to the cortical blood supply Blood viscosity in tube flow: dependence on diameter and hematocrit Neutrophil adhesion in brain capillaries reduces cortical blood flow and impairs memory function in Alzheimer's disease mouse models Susceptibility to capillary plugging can predict brain region specific vessel loss with aging Cerebral blood flow decrease as an early pathological mechanism in Alzheimer's disease Spread of MHV-JHM from nasal cavity to white matter of spinal cord. Transneuronal movement and involvement of astrocytes Olfactory transmucosal SARS-CoV-2 invasion as a port of central nervous system entry in individuals with SARS-CoV-2 The S1 protein of SARS-CoV-2 crosses the blood-brain barrier in mice COVID-19 infection induces readily detectable morphologic and inflammation-related phenotypic changes in peripheral blood monocytes Possible routes of SARS-CoV-2 invasion in brain: In context of neurological symptoms in SARS-CoV-2 patients Incidence of thrombotic complications in critically ill ICU patients with SARS-CoV-2 Brain microvascular occlusive disorder in SARS-CoV-2: a case report SARS-CoV-2: what do we know about "long Covid Long-term micro-structure and cerebral blood flow changes in patients recovered from COVID-19 without neurological manifestations Neutralizing nanobodies bind SARS-CoV-2 spike RBD and block interaction with ACE2