New data reported in a presentation at the American College of Rheumatology (ACR) annual meeting showed that tumor necrosis factor (TNF) receptor 1 (TNFR1) rather than TNFR2 is responsible for TNF-mediated pulmonary endothelial and stromal pathology that causes pulmonary arterial hypertension (PAH), in a mouse model of connective tissue disease-associated PAH.
In this exclusive MedPage Today video, lead author Benjamin Korman, MD, of the University of Rochester Medical Center in New York, discusses the study results and their potential implications for human disease.
Following is a transcript of his remarks:
Our work focuses on a pulmonary hypertension phenotype that we found in TNF-transgenic mice. So we had previously described that a particular version of mice that overexpress a single copy of human TNF develop a progressive and obliterative pulmonary vascular phenotype, where particularly the female mice mostly die by five and a half months of age. And this phenotype was shown to very closely phenocopy at a genetic level, human connective tissue disease associated pulmonary arterial hypertension. These mice had very high pulmonary pressures and genomically, they looked very similar to what we see in scleroderma and other connective tissue disease types of pulmonary hypertension. So one of our interests was to try and understand the biology, the mechanism, by which these mice were developing this disease.
And so to go about doing this, we took two different approaches. One was to try and identify which of the TNF receptors may be driving the phenotype. And this is largely because there are two TNF receptors, receptor 1 and receptor 2, and there’s kind of equipoise as to which one of those might be involved. So while receptor 1 has been largely implicated in a lot of our rheumatic inflammatory diseases, there is a significant vascular component that TNF receptor 2 plays. And certainly thinking about therapeutics, if receptor 2 was the major driver, some of our drugs that we use might not necessarily be the right way to target this phenotype.
The second thing that we wanted to do was to better understand the cellular heterogeneity that was happening in the lungs of the mice. And so we actually performed single-cell RNA sequencing of mice, both wild type and TNF at 2 months, 4 months, and five and a half months. And particularly focused on their endothelial and their stromal mesenchymal cells because we had previously shown that it was non-hematopoietic cells that were driving the phenotype.
So what we ended up finding was, very interestingly, that mice that were deficient in TNF receptor 1 were completely protected from the pulmonary arterial hypertension as well as the arthritis phenotype that these mice have. And so this goes along with one of the other TNF receptor models that have been described previously, but in contrast to another. And so, sort of pointing [out] that at least the pulmonary hypertension part is basically completely driven by receptor 1, which while maybe not completely unexpected certainly is important in understanding the pathobiology of the disease and certainly making us think that targeting with anti-TNF drugs that we have would be a rational strategy. And in fact, we’ve previously shown that that works in these mice. But thinking about safety, that hopefully we’re not gonna have the MS [multiple sclerosis] situation. And so hopefully this is a good and a safe therapeutic option.
Then in terms of the single-cell RNA sequencing, we actually had a lot of interesting things, but I’ll highlight a couple. So, consistent with a pulmonary hypertension phenotype, we did end up seeing an increase in vascular smooth muscle cells. We also saw a loss in endothelial cells, particularly the general capillary endothelial cells, not the aerocyte capillary endothelial cells, as well as major changes in gene transcription in all of these cells. And we sort of go through pathways, but I’m not gonna talk about that here.
We also saw changes in other mesenchymal populations. So pericytes went away, very interestingly. And we saw a shift in the phenotype of the fibroblasts. So whereas normally you have both what we call lipofibroblast, which are usually collagen 13 expressing, and a different kind of matrix fibroblast that expresses collagen 14, what we saw was basically almost a complete loss of the lipofibroblast phenotype and a significant overexpression of those collagen 14 matrix lipofibroblasts in pulmonary hypertension. Suggesting that TNF and specifically TNF receptor 1 is driving this pathobiology, whereas you’re getting both a simultaneous upregulation of a number of important stromal cells, particularly col-14 fibroblasts and smooth muscle cells. But that same biology is also driving the loss of endothelial cells. And since we know that it’s not the immune cells that are actually causing these mice to get pathology, TNF receptor 1 by altering those two different populations of cells is making it so the mice are dying of pulmonary hypertension.
So this gives us a lot of interesting approaches that we can tackle in human disease. So thinking as we’re going to be trying to get samples from patients with scleroderma and other connective tissue disease-associated pulmonary hypertension to actually be able to go after these specific cell populations. And then maybe we can figure out if there are more tractable targets than what we have right now.