Because ACE2 is used by SARS-NCoV-2 to enter the cell, I am curious what factors determine its expression. Interestingly, myocardial infarction increases ACE2 expression in the heart in an animal model ( https://www.ncbi.nlm.nih.gov/pubmed/15671045 ). The paper found no significant effect from ramipril, an ACE inhibitor, but I don't think I can assume ARBs would work the same way. (I am suspicious they might increase it based on the increase in the product Ang(1-7) level - https://www.nature.com/articles/hr200974 ) Note that soluble sACE2 can bind cells that did not produce it, by an RGD independent association with integrin beta 1.

Anything that influences the amount of ACE2 on cell surfaces is a useful answer.


3 Answers 3


I should follow up with some things I've found. As a starting point, the normal expression pattern is available at https://onlinelibrary.wiley.com/doi/pdf/10.1002/path.1570 - note the staining of simple squamous epithelium in the alveoli and the endothelia of blood vessels. Another paper details a sharp decrease in expression with age in rats, but I don't know if that extends to humans.

For regulation of expression, I was surprised to see that Google Scholar actually provides a very good assortment of search results, easier to work with than PubMed, which were topped by the MI regulation I described in the question. Additionally https://onlinelibrary.wiley.com/doi/abs/10.1002/path.1670 (yes, that IS a different link) describes ACE2 expression in renal disease but not in normal kidney. ACE2 expression correlated with proteinuria, but negatively with GFR. https://link.springer.com/article/10.1007/s00125-008-0988-x There are discordant notes in some other situations: for example, subtotal nephrectomy decreased ACE2 expression in a way that was partially prevented by ramipril. https://portlandpress.com/clinsci/article/118/4/269/68827/Reduction-in-renal-ACE2-expression-in-subtotal

The mRNA and protein are both described in atherosclerotic lesions: https://onlinelibrary.wiley.com/doi/abs/10.1002/path.2357

In the lungs https://jvi.asm.org/content/79/23/14614.short describes a positive correlation with differentiation state in airway epithelium. The regulation by HIF-1 alpha in hypoxia seems of special importance: https://journals.physiology.org/doi/full/10.1152/ajplung.90415.2008 but it is complicated (both positive and negative).

ACE2 is also expressed at high levels in placenta during pregnancy - at least in rats. https://journals.physiology.org/doi/full/10.1152/ajpregu.90592.2008

Last but not least, pharmacologic data is reviewed at https://link.springer.com/content/pdf/10.1007/s11906-008-0076-0.pdf which describes up-regulation of ACE2 by angiotensin II receptor blockers, ACE blockers, and mineralocorticoid receptor blockers. This includes the common drugs lisinopril and losartan. (see also https://onlinelibrary.wiley.com/doi/full/10.1111/jcmm.12573 ) Unfortunately the paper did not report substances inhibiting ACE2 expression. But https://journals.physiology.org/doi/full/10.1152/ajpheart.00239.2008 reported that high glucose (!) or PKC inhibitors could reduce ACE2 expression. (but the high glucose was causing kidney injury, which as described above could mean more ACE2...) The anti-diabetes drug liraglutide increased ACE2: https://academic.oup.com/endo/article/156/10/3559/2422879 To top it all off? SARS itself decreases ACE2 expression, and this might be part of the process injuring the lungs: https://www.nature.com/articles/nm1267

This is by no means a complete survey, and there is much I don't understand. The relationship between ACE2 biology and the emerging COVID-19 risk groups seems apparent - as is the urgent need for research to determine which manipulations to ACE2 expression have positive versus negative effects on the prognosis of that disease.

Update: a NEJM podcast today described this biology as 'complicated' link. It is clearly of interest, but still under investigation. For example, losartan is presently the subject of two clinical studies news, presumably these ClinicalTrials


Perhaps take a look at the GeneCards record for ACE2 (https://www.genecards.org/cgi-bin/carddisp.pl?gene=ACE2), which contains annotations for regulatory regions (e.g., enhancers and promoters) which contain transcription factor binding sites. TFs could potentially bind to these regions to regulate the expression of this target. ChIP-seq data can help lend experimental support.

  • 1
    $\begingroup$ Thanks for the suggestion - I found a link there to europepmc.org/article/MED/15151696 which shows an increase in HUMAN heart conditions. The transcription factor sites, well... it's hard to be sure what is meaningful. It seems very easy to start chasing daydreams. MITF is in monocyte-derived lineages, and macrophages produce ACE2, so ...? But it seems like that site is everywhere. $\endgroup$ Mar 10, 2020 at 21:24
  • 1
    $\begingroup$ A lot of pathways may be redundant or constitutive. I'm not sure if a deeper dive into papers related to knock-out or knock-down mutants related to ACE2 regulation (or upstream) may help? regulatorynetworks.com may be of interest. $\endgroup$ Mar 10, 2020 at 22:11
  • $\begingroup$ It might be useful to know about the physiological role of ACE2 and its product Angiotensin (1-/). If 1 to 7 is good for you, why not over-express. Also, I wonder about some general theory on "over-expression of hormone receptors". $\endgroup$ Nov 12, 2020 at 13:21

Over-expression of ACE2 in case of ACE inhibition has been found: Carlos M Ferrario et al., Effect of angiotensin-converting enzyme inhibition and angiotensin II receptor blockers on cardiac angiotensin-converting enzyme 2, https://pubmed.ncbi.nlm.nih.gov/15897343/

These findings seem to be confirmed by findings that vice versa, in the opposite situation of high levels of AngiotensinII downregulation of ACE2 occurs:

Deshotels MR, Xia H, Sriramula S, Lazartigues E, Filipeanu CM: Angiotensin-II mediates angiotensin converting enzyme type 2 internalization and degradation through an Angiotensin-II type I receptor-dependent mechanism. Hypertension 2014; 64: 1368–75 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4231883/

Two studies, however, apparently did not find any down- or upregulation, which is in line with the reference you cited:

  1. Ramchand J, Patel SK, et al., Elevated plasma angiotensin converting enzyme 2 activity is an independent predictor of major adverse cardiac events in patients with obstructive coronary artery disease. Plos One 2018; 13: e0198144
  2. Walters TE, Kalman JM, et al., Angiotensin converting enzyme 2 activity and human atrial fibrillation (...) Europace 2017; 19 (8): 1280–7

As your question seems to be unsettled here is my individual opinion:

The feedback circle you suggest makes sense as a mechanism of upregulation (more product leads to more production). Why not have more input if output can be made and has been generated in larger and larger amounts? However, there also is "demand or no demand" down the cascade. A large amount of product that accumulates thus may signal a lack of its being needed - I'd cast my vote for down-regulation of the expression of input receptors, as I'd vote for "we have enough". So contrary to your answering your question I'd suggest a high amount of product Angotensin (1-7) might cause a down-, not upregulation of ACE2.

Over-expression of receptors, to my knowledge, explains drug addiction and getting used to medication.

If in surplus of ligand hormone (in that case Angiotensin II) all receptors are "filled", it makes sense to broaden the basis and up regulate expression of ACE2. Thus, a block of ACE (not ACE2!) by ace-inhibitors would not lead to overexpression which seems coherent with the results of the study on blood pressure medication you cited.

Interestingly, there is an alternative pathway of ligand transformation, and an alternative product of ACE2, it's Angiotensin (1-9). It think it is a possibility that those ACE2 cells do over-express that are engaged in the modification of AT 1-9, not AT II, as ACE-inhibitors ("beta blockers" in German) do not address or inhibit renin production which should remain elevated. Renin becomes Angiotensin I, the alternative ligand of ACE2. However, downward that alternativ pathway, it's ACE again that transforms, and its being inhibited dampens "demand".

  • 1
    $\begingroup$ Your post starts out by seeming to answer the OPs question including a relevant reference. I was prepared to say good job and upvote at that point. However, you then continue with text that appears to be speculative if not entirely fictional and stop including references. Are you trying to rescue your initial speculation by tacking on something like a real answer at the beginning? No matter how much you love your ideas this isn't the place for them. (Also note that "beta blockers" is not German.) Please take the tour and read through the help center starting with How to Answer and edit accordingly. $\endgroup$
    – tyersome
    Nov 14, 2020 at 2:36
  • $\begingroup$ By " your post starts out by seeming..." you refer to my edited newer version (yes, I see, 5 hours ago). So, yes indeed, no misleading as introduction. It's intension to be polite, I remember deleting "contrary" to what you suggest and refere in your answer. So without saying I think reputabel references that come to diverging conclusions that are just the opposite of each other should give the right here and now to speak up with an opinion that is frankly declared as such. Feel free to down vote. You clearly say that this is not the place for my "ideas"? $\endgroup$ Nov 14, 2020 at 8:09
  • $\begingroup$ To lyersome: Also note that "Betablocker" is German, yes, it is. Are you German, too? (By the way, I can tell you you wouldn't want to say "betablockers" with an s if you want the plural. $\endgroup$ Nov 14, 2020 at 8:11
  • $\begingroup$ To lyersome, again: are you offended by my clumsy use of English when I said "Here is my own personal answering"? Sorry for that, it sure aint etzetera. What I find annoying that I feel peer reviewed without peer or editor taking up on verifying if indeed there are divergent studies (which I - it's up to you as administrator of this place I happily admit to you - think is a legitimate basis to offer one's own thought. Are you serious: this is not the place for my own ideas? $\endgroup$ Nov 14, 2020 at 8:15
  • $\begingroup$ To lyersome: I have revised according to your criticism as far as I could get me. Even this I "Write to the best of your ability" :-). Citation meant to be some proof of me having read your recommended help page on how to write a good answer. There, I could not find anything in favour of your point that answers should not contain any own ideads (what's more, if they are clearely declared as personal knowing, it seems just the contrary, it surely is and should be encouraged - the more if there really is no or diveregent authority!) $\endgroup$ Nov 14, 2020 at 8:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .