It really IS the red cell

In this issue of Blood, Patel and colleagues demonstrate that placental growth factor derived from hemoglobin S erythroid cells upregulates the expression of both ET-1 and ET-BR via HIF-1α in the absence of hypoxia.

It has become increasingly clear that the process leading to vaso-occlusion in sickle cell disease (SCD) is quite complex and likely brings into play not only red-cell adhesion and red-cell sickling, but also leukocyte adhesion and activation, cytokine production, activation of coagulation, and induction of endothelial-cell activation. Together, these processes lead to further exacerbation of the occlusive process, hypoxia reperfusion injury, and extension of tissue damage. Thus, many investigators have focused recently on the role played by the nonerythroid cells and factors involved in these processes. Attention has been paid to the role of leukocytes, the elevated levels of proinflammatory cytokines, the activation of thrombogenesis, and platelet activation. And yet, one must ask: do all these cells and processes become involved just because sickled red cells become stuck in small vessels?

Patel and colleagues in this issue of Blood now redirect our attention to the primary cause of sickle cell disease and vaso-occlusion—the red blood cell. Sickle cell disease is, after all, a disease of hemoglobin (Hb), whose expression is restricted to red blood cells. While the propensity of cells with predominantly Hb S to sickle at low oxygen tension and to adhere to endothelial cells was described decades ago, the degree to which abnormal red cells play a truly dynamic role in sickle cell disease is only now being appreciated.

Connecting the dots, Patel and colleagues have now discovered an intriguing pathway that may explain a great deal, especially about the process of lung and pulmonary vascular injury in SCD. Tordjman et al showed in 2001 that erythroid cells were the only bone marrow hematopoietic cells that coexpressed 2 angiogenic factors, VEGF-A and PlGF; moreover, they showed that expression of these factors increased during erythroid maturation, and that erythroblasts secreted these factors and thus were capable of inducing migration of both monocytes and endothelial cells.¹  PlGF is a member of the vascular endothelial growth factor (VEGF) family of proteins. It is typically secreted and interacts with several receptor tyrosine kinases in the VEGFR family. In addition to being a proangiogenic factor, it is also proinflammatory and may play an important role in the instability of atherosclerotic plaques, as well as in tumor neovascularization.

In 2003, companion papers by Perelman et al²  and Selvaraj et al³  showed that levels of PlGF were increased in SCD at least roughly proportionately to the frequency of vaso-occlusive episodes. They also showed that PlGF directly activated monocyte chemotaxis and mRNA levels of interleukin-1, interleukin-8, monocyte chemoattractant protein-1, and VEGF. Furthermore, Hb SS erythroid cells appear to contain more PlGF per cell than do normal cells, and this is hypothesized to account for the increased PlGF levels in SCD.⁴  Investigation of the mechanism whereby PlGF stimulated monocytes revealed that PlGF activates the monocyte Flt-1, which then leads to activation of PI2 kinase/AKT and ERK-1/2 signaling.⁴ 

In their paper in this issue of Blood, Patel and colleagues have now shown that the pathways activated by PlGF are potentially involved in the development of SCD-associated pulmonary hypertension, a grave complication affecting approximately one-third of adults with SCD. They showed not only that PlGF induces increased expression of endothelin-B receptor (ET-BR) by monocytes, but also that it induces expression of endothelin-1 (ET-1), an ET-BR ligand, by human microvascular endothelial cells. ET-1 is known to be increased in SCD and to become further elevated during vaso-occlusive episodes and acute chest syndrome.⁴  Interestingly, the effects of PlGF on endothelial cells and monocytes occurred via activation of PI-3 kinase and also involved hypoxia-inducible factor-1α (HIF-1α) in the absence of hypoxia. These effects potentially constitute a double whammy that may lead to a vicious cycle of both vasoconstriction and inflammation in the pulmonary circulation.

At this point, definite links between PlGF-induced processes and pulmonary hypertension have not been established. For example, serum PlGF levels during pregnancy peak at up to 1000 pg/mL,⁵  far higher than the levels of less than 20 pg/mL reported in SCD patients, and women with SCD nevertheless routinely survive pregnancy. Nonetheless, these findings are intriguing. ET-1 overexpression in SCD has previously been attributed to the effects of chronic anemia on oxygen delivery. The current study, however, suggests that hypoxia, which may be an aggravating factor, is not necessary for the effects of PlGF despite the involvement of HIF-1α. The participation of HIF-1α in a process not involving hypoxia remains novel, and further work is needed to confirm the physiologic importance of this observation. Meanwhile, as the authors show, we have yet to fully appreciate the panoply of effects exerted by sickled red cells on human tissues (see table).