Stroke paradox: Recovery with less blood flow.

This blog is based on the recent publication, in June 2026, in Cell Death and Disease, by Kamal Narayana, Isabel C. Lambert, Sam Burford, Emilie Gosselin, Jakob Körbelin & Craig E. Brown, titled

"Knockdown of endothelial Serpine1 improves stroke recovery by attenuating peri-infarct blood flow and blood-brain barrier disruption".

When we think of stroke treatment, one principle dominates: restore blood flow, QUICK. What if the story doesn't end once the blocked artery is reopened?

Even the authors followed this conventional knowledge as they noted: "this is what motivated us to inhibit Serpine1/PAI-1 signalling".

But the 'opposite' happened! This is what inspired the authors (mostly) from the University of Victoria to investigate the role of Serpine1 (which encodes plasminogen activator inhibitor-1, or PAI-1) after stroke. Previous studies had shown that PAI-1 increases in ischemic tissue relative to non-ischemic tissue (Yanev et al., 2025); it also as a result of aging and hypertension (Khoddam et al.,2025), and that systemic inhibition PAI-1 could reduce the size of the initial brain injury (Chan et al., 2018), but it remained unclear whether targeting endothelial cells within blood vessels could influence recovery. Rather than asking whether they could prevent stroke, researchers asked, could modifying the brain's blood vessels create a better environment for recovery?

The effect of stroke on Serpine1:

Before testing whether Serpine1 influenced stroke recovery, researchers needed to establish a simple yet essential fact: does stroke change Serpine1 expression in the brain?

The answer was yes; as shown in their Figure 1. D. (Reproduced with permission.)

They induced stroke in a highly localized and reproducible manner in mice and noted that after 3 days following stroke, the protein levels encoded by Serpine1 reach their highest levels. Protein levels are normalized to the baseline PAI-1 in the sham stroke (induced stroke without the actual blockage within vessels).

Much of this increase was observed around blood vessels in the tissue surrounding the infarct, as shown in the confocal fluorescence microscopy images below, suggesting that endothelial cells may play an important role in the brain's response to stroke. DAPI (binding DNA) is blue, PAI-1 in green, and collagen in magenta, as a marker of brain blood vessels.

There comes the question: if endothelial cells produce significant levels of Serpine1 after stroke, what role is it playing? Is it helping the brain recover, or could it be contributing to further damage?

To answer this, the researchers selectively reduced Serpine1 expression in endothelial cells using a viral vector carrying Cre recombinase (Sternberg and Hamilton, 1981). Unlike previous studies that eliminated Serpine1 throughout the body, this approach allowed them to investigate the specific contribution of the brain's vascular endothelium while leaving other cell types unaffected. Their experiments confirmed that the knockdown successfully reduced endothelial Serpine1 expression, providing the foundation for testing its role in stroke recovery.

The next findings challenged the authors' original expectations.

Because Serpine1 is best known for regulating blood clot breakdown, the researchers initially predicted that reducing its activity might improve blood flow around the injured tissue. Instead, they observed the opposite. Mice with endothelial Serpine1 knockdown showed reduced capillary blood flow in the peri-infarct region during recovery, as shown in their Fig 2, F, reproduced with permission:

Note red blood cell, or blood flow, velocity on the Y-axis and days post-stroke on the X-axis, with arteriole blood (blood coming from the heart, oxygenated blood, in red and venule blood blood going to the heart, in purple).

At first glance, this seems counterintuitive. Conventional thinking suggests that increasing blood flow should help damaged tissue recover. Yet despite this reduction in capillary flow, the mice demonstrated better functional recovery.

This study challenges the consensus that increased blood flow is beneficial after the brain has suffered a stroke. Both the control group and Serpine1 mutant mice developed strokes of similar size, and blood flow during the first few hours after stroke was comparable. However, the specific endothelial Serpine1 knockdown appeared to improve the brain regions in the vicinity of the induced stroke.

One clue came from the blood-brain barrier (BBB), a protective membrane that regulates what enters the brain from the bloodstream. After stroke, this barrier becomes disrupted, allowing inflammatory molecules and immune cells to enter.

Mice lacking endothelial Serpine1 showed less BBB leakage three days after stroke. This was accompanied by reduced expression of several inflammatory genes and lower levels of inflammatory cytokines, including IL-6. Together, these findings suggest that endothelial Serpine1 influences recovery by preserving the integrity of the vasculature and the blood-brain barrier.

My Thoughts

It has been well-established that one main role of PAI-1 (Serpine1) is regulating blood clot breakdown (fibrinolysis). Since PAI-1 inhibits tissue plasminogen activator (tPA), reducing PAI-1 would be expected to promote fibrinolysis, the body's natural process of dissolving clots.

The authors focus primarily on how endothelial Serpine1 knockdown appears to preserve the blood-brain barrier and reduce inflammation, both of which likely contribute to improved recovery. But I couldn't help wondering whether enhanced fibrinolysis might also play a supporting role.

Could the improved outcomes arise not from a single mechanism, but from several processes working together? In other words, could stabilizing the blood-brain barrier while simultaneously improving the clearance of small residual clots create a more favourable environment for recovery?

The study wasn't designed to answer that question, and it would require additional experiments to test. Nevertheless, it's an intriguing possibility and a reminder that biology does not work on one isolated pathway. Recovery after stroke is complex; it's the interplay between many mechanisms. No single "magic bullet"!

References:

Chan, S. L., Bishop, N., Li, Z., & Cipolla, M. J. (2018). Inhibition of PAI (Plasminogen Activator Inhibitor)-1 Improves Brain Collateral Perfusion and Injury After Acute Ischemic Stroke in Aged Hypertensive Rats. Stroke, 49(8), 1969-1976. https://doi.org/10.1161/STROKEAHA.118.022056

Khoddam, A., Vaughan, D., & Wilsbacher, L. (2025). Role of plasminogen activator inhibitor-1 (PAI-1) in age-related cardiovascular pathophysiology. J Cardiovasc Aging, 5(2). https://doi.org/10.20517/jca.2024.38

Narayana, K., Lambert, I. C., Burford, S., Gosselin, E., Korbelin, J., & Brown, C. E. (2026). Knockdown of endothelial Serpine1 improves stroke recovery by attenuating peri-infarct blood flow and blood-brain barrier disruption. Cell Death Dis. https://doi.org/10.1038/s41419-026-09062-9

Sternberg N Fau - Hamilton, D., & Hamilton, D. (1981). Bacteriophage P1 site-specific recombination. I. Recombination between loxP sites. J. Mol. Biol., 150(0022-2836 (Print)), 467-486. https://doi.org/10.1016/0022-2836(81)90375-2

Yanev, P., Martin-Jimenez, C., Vesga-Jimenez, D. J., Zvinys, L., Weinrich, N., Cree, M. A., Preuss, T. M., Zhang, X., & Yepes, M. (2025). Plasminogen activator inhibitor-1 mediates cerebral ischemia-induced astrocytic reactivity. J Cereb Blood Flow Metab, 45(1), 102-114. https://doi.org/10.1177/0271678X241270445