SARS-CoV-2 triggers pericyte-mediated cerebral capillary constriction

Abstract

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 evoked a small pericyte-mediated capillary constriction via AT1 receptors, but evoked a large constriction when the SARS-CoV-2 receptor binding domain (RBD, original Wuhan variant) was present. A mutated non-binding RBD did not potentiate constriction. A similar RBD-potentiated capillary constriction occurred in human cortical slices, and was evoked in hamster brain slices by pseudotyped virions expressing SARS-CoV-2 spike protein. This constriction reflects an RBD-induced decrease in the conversion of angiotensin II to angiotensin-(1-7) mediated by removal of ACE2 from the cell surface membrane and was mimicked by blocking ACE2. The clinically used drug losartan inhibited the RBD-potentiated constriction. Thus, AT1 receptor blockers could be protective in COVID-19 by preventing pericyte-mediated blood flow reductions in the brain, and perhaps the heart and kidney.

Keywords: SARS-CoV-2; capillary; pericyte.

Figure 1

Cerebral pericytes express ACE2 and constrict capillaries in response to Ang II. (A) Labelling of hamster cortical slice with antibodies to the SARS-CoV-2 receptor ACE2, the pericyte markers NG2 and PDGFRβ and with DAPI to label nuclei. (B) Lower magnification (note different scale bar) maximum intensity projection of ACE2 and PDGFRβ labelling, showing capillaries and penetrating arteriole. (C) Integrated ACE2 labelling overlapping with a binarized mask of PDGFRβ labelling and with the inverse of this mask. (D) Integrated ACE2 labelling over capillaries versus penetrating arterioles (PA). In (C) and (D) the number of image stacks is on the bars; data in (C) and (D) were each from two animals. (E) Average normalized diameter changes (mean ± SEM) at seven pericytes (in different brain slices from four animals) exposed to the thromboxane A₂ analogue U46619 (200 nM) and then with the neurotransmitter glutamate (500 μM) superimposed. (F) Average normalized diameter changes at nine pericytes (in different slices from seven animals) exposed to 150 nM angiotensin II alone (i.e. in artificial CSF, aCSF) and 10 pericytes (from three animals) exposed to angiotensin II in the presence of the AT1R blocker losartan (20 μM). (G) As in F (aCSF plot is the same) but showing angiotensin II response in the presence of the AT2R blocker PD123319 (1 μM, nine pericytes from four animals) or the Mas receptor blocker A779 (10 μM, five pericytes from two animals). (H) Peak constriction evoked by angiotensin II in different conditions (number of pericytes studied shown on bars). Points superimposed on bar graphs here and in subsequent figures are individual data-points (pericytes or image stacks) contributing to the mean.

Figure 2

The SARS-CoV-2 RBD potentiates angiotensin-evoked capillary constriction. (A) Perfusion of brain slices with RBD (0.7 mg/l) has no significant effect on capillary diameter at pericytes (mean ± SEM; n = 8 pericytes each for artificial CSF (aCSF) and RBD, from three and four animals, respectively). (B) After pre-incubation in aCSF for 30 min, applying 2 μM angiotensin II evokes a small transient capillary constriction at pericytes (n = 6, from three animals), while including RBD (0.7 mg/l) in the solutions results in an ∼5-fold larger response to angiotensin II (n = 4 from two animals; peak constriction plotted is slightly larger than the mean value quoted in the text because the latter was averaged over five frames and here only every fifth frame is plotted). (C) RBD has no effect on constriction evoked by 200 nM U46619 (n = 6 pericytes for aCSF and 5 for RBD from two animals each). (D) Response to 50 nM angiotensin II after pre-incubation and subsequent perfusion with aCSF or aCSF containing RBD or Y489R mutant RBD (n = 9 for each, from three, four and three animals, respectively). (E) Surface plasmon resonance responses for RBD and mutant (Y489R) RBD binding to immobilized human ACE2. (F) Mean constriction between t = 29.67 and 30.00 min in D.

Figure 3

The effect of RBD is mimicked by blocking ACE2 and reduced by losartan. (A) Capillary constriction at pericytes in response to 50 nM angiotensin II in the absence (n = 9) and presence (n = 9) of the RBD (mean ± SEM, replotted from Fig. 2D) or the presence of the ACE2 inhibitor MLN4760 (1 μM, nine pericytes from three animals, with no RBD). (B) Constriction in A between t = 29.67 and 30.00 min. (C) Response to 50 nM angiotensin II after 30-min incubation in (and continued perfusion with) artificial CSF (aCSF) containing RBD (0.7 mg/l, replotted from Fig. 2D) or additionally losartan (20 μM, 10 pericytes from three animals). (D) Constriction in C between t = 29.67 and 30.00 min.

Figure 4

SARS-CoV-2 potentiates constriction in human and hamster capillaries. (A) Percentage of pericytes dead (assessed with propidium iodide) in hamster brain slices (numbers on bars, from two animals) after 3 h incubation in artificial CSF (aCSF) or aCSF containing 50 nM angiotensin II and/or RBD (0.7 mg/l). (B) Effect of 50 nM angiotensin on capillary diameter (mean ± SEM) at pericytes in human brain slices in the presence (five pericytes from two humans) and absence (four pericytes from two humans) of the RBD (30-min pre-incubation). (C) Mean constriction at 30 min from data in B. (D and E) SARS-CoV-2 pseudotyped virus (see ‘Materials and methods’ section) evokes pericyte-mediated capillary constriction. (D) Capillary diameter at hamster cerebral cortex pericyte somata after incubation with angiotensin II alone (305 pericytes from two animals) or with pseudotyped virus and angiotensin II (289 pericytes from two animals). (E) Capillary diameter as a function of distance from pericyte somata in the presence (289–255 pericytes per point from two animals) and absence (305–277 pericytes per point from two animals) of pseudotyped virus, in both cases with angiotensin II. The pseudotyped virus induces constriction specifically at the somata. (F and G) Likely mode of operation of RBD binding to ACE2. (F) Normally, angiotensin II [e.g. generated by the brain renin–angiotensin system (RAS)] can act on vasoconstricting AT1 receptors or vasodilating AT2 receptors, and is converted (pink arrow) by pericyte ACE2 to vasodilating angiotensin-(1-7) that acts via vasodilating Mas receptors. (G) In the presence of SARS-CoV-2, binding of the spike protein RBD to ACE2 leads to its internalization or cleavage and secretion (see main text), reducing the conversion of angiotensin II to angiotensin-(1-7). Angiotensin II (derived from the brain RAS or from the peripheral RAS) will then evoke a different balance of responses via the receptors shown, generating a larger constriction because of less activation of Mas receptors.