Figuring out the pathophysiology of VITT
Our paper, recently published shows that platelet factor 4 activates platelets through the thrombopoietin receptor, c-Mpl.
We’ve just published a paper describing a new mechanism of platelet activation in VITT. Read the paper here.
In 2021, a small fraction of people who received the the Oxford-AstraZeneca COVID-19 vaccine developed a dangerous, clotting syndrome - vaccine-induced immune thrombocytopaenia and thrombosis (VITT). The temporal association between vaccination and unusual blood clots was quickly noticed and published independently and simultaneously by three groups from the UK, Germany, and Norway in the same issue of the New England Journal of Medicine.
My boss, Pip Nicolson, spent a frantic couple of months gathering samples from 12 patients with VITT in Birmingham, and figuring out the mechanism by which platelets are activated.
In VITT, antibodies form against a platelet protein - platelet factor 4 (PF4). This is a really interesting molecule. It’s a ball-shaped protein with a strong band of positive charge running around its equator. This means it can bind strongly to negative charge and it in all of us it performs functions in immunity, stopping the growth of new blood vessels, and acting as a break on platelet production.
When platelets get activated, they release plenty of PF4 and it’s been known for quite a while that PF4 can help to activate platelets but we didn’t know how.
We first noticed that platelets were being activated by PF4 (on aggregation traces) and my colleague Luis, who is now working in Bristol, suggested that we send some PF4-activated platelets for phosphoproteomic analysis by mass spectrometry. On our part, this was dead easy. We added some PF4 to platelets, lysed them, ran the protein on an electrophoresis gel, cut out some interesting bands at 95 kilodaltons and sent (well, hand delivered) the samples to the Biosciences department at the University of Birmingham.
After a couple of weeks, the data came back. In our sample, there was phosphorylated STAT5, a signal transducer protein. This was a big clue. We know that the STAT proteins are activated by another group of proteins called the JAKs. We looked for phosphorylated JAK in our PF4 samples and it was there. Then, the next step was to find the receptor. The most famous JAK-associated receptor in platelets is the thrombopoietin receptor, c-Mpl. We looked for it being activated in response to PF4 and hey presto, it was!
c-Mpl is a well known receptor for the really important hormone, thrombopoietin. Thrombopoietin is produced by the liver and drives the differentiation and proliferation of megakaryocytes, increasing the number of platelets that they produce. Mature platelets keep the c-Mpl receptor so that they can provide a check on platelet production. If you have lots of platelets already, all the c-Mpl receptors soak up thrombopoietin so less of it gets to the bone marrow to drive platelet production - a neat feedback mechanism.
My colleague, Elena Slater, confirmed that PF4 interacts with c-Mpl using a technique called surface plasmon resonance, showing that it does interact and that this interaction is about 100 times weaker than thrombopoietin - unsurprising.
We next wondered if this mechanism of platelet activation was important in VITT. Using the JAK2 inhibitor, ruxolitinib, we were able to block platelet activation to serum and antibodies taken from patients with VITT. Finally, we used a c-Mpl blocking antibody to inhibit platelet activation to VITT antibodies and found that this also worked.
Like all science, we have used a reductionist methodology. We have simplified complex biological processes by isolating platelets and combining them with isolated PF4 and antibodies, when in reality, the pathophysiological mechanisms in VITT and thrombosis in general are happening in vessels with endothelium, moving blood, and a whole host of other molecules, activators and inhibitors. Having said this, we still think that it is likely that this mechanism plays a role in humans with VITT and another closely related condition - heparin-induced thrombocytopaenia (HIT). Work on HIT is on-going.
So, the question is, could we block this process in humans with VITT (and HIT). The answer is “Yes, we could” but we think that it would be far more fruitful to block the main way that platelets are activated - through the immunoglobulin receptor FcγRIIA. We could use drugs like Bruton tyrosine kinase inhibitors to completely block the activation of platelets through this receptor and we think that this could be a promising treatment for both VITT and HIT. We definitely need new treatments for these diseases. HIT is common in all countries where heparin is used and VITT continues to be a problem in parts of the world that are still using the Oxford-AZ (and other adenoviral vector COVID-19 vaccines). Furthermore, if a new, high mortality pandemic broke out tomorrow, we would again need adenoviral vector technology. Our knowledge of the diagnosis and treatment of VITT would hugely improve outcomes in the few people that develop it.
Our findings also have some implications for our understanding of platelet production. As PF4 is known to inhibit platelet production, an obvious question is, can it act by blocking access to the c-Mpl receptor for thrombopoietin. This is certainly plausible and we did some preliminary experiments to show that this may well be the case. However, lots more work on this is needed. If this is the case, it’ll be a curious mechanism as PF4 is also an agonist for c-Mpl; as far as we can tell, it induces the same downstream signalling as thrombopoietin when we look in platelets. Of course, this could be different in megakaryocytes and recent work published in Cell has shown how mutations of thrombopoietin can lead to different functional outcomes in megakaryocytes despite still activating the receptor.
If you’re up for it, you can read the whole paper here. It’s a Brief Report so won’t take you to much time. If you want to ask anything, feel free to comment this post.