Kidney-Derived Thrombopoietin and Platelet Generation—A Proinflammatory Loop in CKD

Platelets play essential roles in hemostasis where they act to rapidly seal off damaged blood vessels and limit blood loss after vascular injury. In addition to this critical function, a growing body of evidence indicates that platelets can also promote inflammatory responses in numerous settings.¹ The term “thromboinflammation” was coined to describe circumstances in which interdependent platelet accumulation and inflammation combine to exacerbate tissue damage.² This phenomenon is facilitated by the numerous inflammatory molecules expressed on or released by activated platelets.¹ While originally identified in stroke, thromboinflammation is now known to promote pathology in deep vein thrombosis, sepsis, and, most recently, organ-specific inflammation in COVID-19.^2,3 This contribution to inflammatory injury is well-recognized in the pulmonary microvasculature in acute lung injury and COVID-19 and in larger vessels in atherosclerosis and deep vein thrombosis. But what about the complex and inflammation-prone vasculature of the kidney?

In severe forms of renal inflammation, including hemolytic uremic syndrome and lupus nephritis, thrombosis-associated platelet accumulation can cause renal dysfunction by occluding the glomerular microvasculature. However, in progressive forms of GN, platelet accumulation is much less extensive and often can be difficult to detect. Despite this, it has long been recognized that platelets contribute to renal dysfunction and pathology in GN. The evidence for this includes platelet consumption and reduced platelet life span in some forms of GN and detection of platelets and their products in glomeruli and platelet secretory products in the bloodstream of patients with GN.⁴ These clinical observations are supported by numerous findings in experimental GN models demonstrating pathogenic roles for platelets.^5,6 As reducing circulating platelet number attenuates glomerular pathology in the models of acute glomerular injury, understanding what controls the circulating platelet mass in patients with progressive forms of GN could provide therapeutically useful information.

Thrombopoietin (TPO) is the key growth factor that promotes the development of megakaryocytes, the hemopoietic cell type responsible for platelet generation.⁷ TPO is generated primarily in the liver and kidney and, to a lesser extent, in the bone marrow and spleen.⁸ Plasma TPO levels are regulated by the circulating platelet load such that when platelet numbers are low, TPO is elevated and available to stimulate megakaryocyte generation and platelet production.⁷ In progressive GN, it is conceivable that this tightly controlled system could become dysregulated due to possible effects of the disease on platelet counts and lifespan and on TPO generation in the kidney. However, little is known about the effects of GN on TPO production and the potential contribution of this pathway to GN pathology.

To address this point, in this issue, Douté et al. examined a mouse model of antibody-mediated chronic kidney disease induced using an antiserum against the glomerular basement membrane.⁹ In this model, the number of circulating platelets increased 5 days after disease initiation along with the activation of circulating platelets and platelet accumulation in glomeruli. Consistent with previous observations, ongoing platelet depletion protected the kidney from pathologic and functional injury and inhibition of TGFβ, one the inflammatory proteins released by platelets, reduced disease-associated interstitial inflammation and glomerulosclerosis. Notably, renal injury increased the expression of TPO selectively in the kidney, leading to increased TPO in the circulation. In parallel, hemopoietic stem cells in the bone marrow, including those responsible for megakaryocyte generation, underwent expansion and TPO inhibition prevented disease-associated thrombopoiesis indicating a role for TPO in this response. Similar to the experimental findings, ANCA vasculitis patients displayed glomerular platelet deposition and elevated serum TPO. In these patients, therapies that reduced renal injury also reduced serum TPO demonstrating that a reduction in serum TPO is predictive of a positive response to therapeutic intervention. This idea was supported by further experimentation in the animal model in which TPO immunoneutralization normalized platelet counts, reduced platelet activation, and decreased glomerular and interstitial inflammation. Together, these findings indicate that one of the sequelae of immune attack on the kidney in antibody-mediated GN is loss of the tight control of TPO-driven platelet generation. This results in increased circulating platelets with the capacity to accumulate in the glomerulus and promote renal pathology (Figure 1).

Figure 1.

The study by Douté et al. reveals in a model of antibody-mediated chronic kidney disease that a proinflammatory cycle exists linking platelet-mediated kidney injury, thrombopoietin (TPO) generation, and increased platelet generation. In this model, platelets accumulate in inflamed glomeruli where they promote leukocyte-mediated renal pathology. In response to this injury, the damaged kidney increases TPO generation. This TPO feeds back to the bone marrow to promote megakaryocyte development and maturation, resulting in an increase in the circulating platelet mass. The elevated platelet count affords the opportunity for these cells to further contribute to renal injury.

So do these findings identify TPO/platelet generation as a potential therapeutic target in antibody-mediated glomerular disease? Many of the current therapies for immune-mediated GN are nonspecific and have their own toxicities, and therefore, new treatments are urgently needed. Here, the authors used an antibody against TPO to inhibit the function of this growth factor, achieving a therapeutic effect on kidney disease without the undesirable side effects of weight loss and elevated cholesterol seen in mice treated with the corticosteroid, dexamethasone. Notably, anti-TPO prevented the thrombocytosis that commenced 5 days after initiation of disease but did not reduce platelet counts to below normal until two weeks into the model, suggesting that the risk of hemorrhage from this treatment would be limited. Perhaps TPO inhibition would be suitable for short-term treatment of disease flares associated with increased platelet counts. However, whether this is a strategy that could be used in an ongoing fashion is unclear. Lupus patients can develop both anti-TPO antibodies and thrombocytopenia, effects that elevate the risk of life-threatening hemorrhage.¹⁰ Clearly targeting this pathway would need to be undertaken with an abundance of caution.

Nevertheless, the authors compared the effects of anti-TPO with other similarly specific interventions targeting IL-6 and TGFβ. IL-6 inhibition failed to completely prevent the increase in TPO and did not modulate systemic platelet activation or glomerulosclerosis. By contrast, TGFβ inhibition not only reduced aspects of renal inflammation and renal injury but also resulted in exacerbated weight loss indicating that it was not suitable as a therapeutic approach. As such, by comparison, specific targeting of TPO demonstrated efficacy against disease but less evidence of adverse effects compared with both broad spectrum (dexamethasone) and targeted (anti-IL-6 and anti-TGFβ) interventions. These novel and intriguing findings indicate that the TPO-platelet-renal pathology axis is worthy of more detailed investigation in immune-mediated glomerular disease.

Disclosures

M.J. Hickey reports Ownership Interest: Cochlear, CSL, ResMed, and Vertex; Research Funding: National Health and Medical Research Council (Australia); Honoraria: Wiley (Deputy Editor of Microcirculation); and Advisory or Leadership Role: Director of Monash University Centre for Inflammatory Diseases.

Funding

None.

Author Contributions

Writing – original draft: Michael J. Hickey.

Writing – review & editing: Michael J. Hickey.