Hydroxyurea and a cGMP-amplifying agent have immediate benefits on acute vaso-occlusive events in sickle cell disease mice

Abstract

Inhibition of leukocyte adhesion to the vascular endothelium represents a novel and important approach for decreasing sickle cell disease (SCD) vaso-occlusion. Using a humanized SCD-mouse-model of tumor necrosis factor-α-induced acute vaso-occlusion, we herein present data demonstrating that short-term administration of either hydroxyurea or the phosphodiesterase 9 (PDE9) inhibitor, BAY73-6691, significantly altered leukocyte recruitment to the microvasculature. Notably, the administration of both agents led to marked improvements in leukocyte rolling and adhesion and decreased heterotypic red blood cell-leukocyte interactions, coupled with prolonged animal survival. Mechanistically, these rheologic benefits were associated with decreased endothelial adhesion molecule expression, as well as diminished leukocyte Mac-1-integrin activation and cyclic guanosine monophosphate (cGMP)-signaling, leading to reduced leukocyte recruitment. Our findings indicate that hydroxyurea has immediate beneficial effects on the microvasculature in acute sickle-cell crises that are independent of the drug's fetal hemoglobin-elevating properties and probably involve the formation of intravascular nitric oxide. In addition, inhibition of PDE9, an enzyme highly expressed in hematopoietic cells, amplified the cGMP-elevating effects of hydroxyurea and may represent a promising and more tissue-specific adjuvant therapy for this disease.

Figure 1

The NO-cGMP pathway. HU acts as a NO donor in vivo and/or directly activates intracellular sGC. NO stimulates intracellular sGC to produce cGMP from guanosine-5′-triphosphate. Stimulation of cGMP-dependent protein kinase (PKG) by cGMP in erythroid lineage cells elevates HBG and fetal hemoglobin production and decreases leukocyte (WBC) adhesive mechanisms. PDE9 degrades intracellular cGMP to GMP and is highly expressed in hematopoietic cells. BAY73-6691 is a selective inhibitor of PDE9, and therefore elevates intracellular cGMP in cells that express the enzyme. PTIO is a radical scavenger for nitric oxide. L-NAME is an arginine analog that inhibits nitric oxide synthase activity. ODQ (1H-[1,2,4]oxadiazolo-[4,3-a]quinoxalin-1-one) is a selective inhibitor of sGC, whereas KT5823 selectively inhibits PKG.

Figure 2

Effects of coadministration of HU and BAY73-6691 on leukocyte recruitment in TNF-α–treated SCD mice. (A) Leukocyte rolling; (B) leukocyte rolling velocity (velocity of each leukocyte observed; n = 64-82 venules from 3-4 mice per group); (C) leukocyte adhesion, and (D) leukocyte extravasation (n = 5-7 mice per group for A-C-D). (E) Plasma cGMP levels at the time of death/sacrifice as determined by ELISA (n = 5-6 mice per group; *P < .05, **P < .01, ***P < .001). (F) Representative images of SCD mice venules at 180 minutes after TNF-α stimulation. Black stars indicate adherent leukocytes; white arrowheads, RBCs; and white arrow, RBC-WBC interaction. The longer white arrow indicates the direction of blood flow. Scale bar: 10 μm.

Figure 3

Effects of oral administration of hydroxyurea on leukocyte recruitment in TNF-α–treated SCD mice. HU (250 mg/kg) was given orally to SCD mice at the time of TNF-α administration. (A) Leukocyte rolling; (B) adhesion; (C) extravasation; and (D) quantification of RBC-WBC interactions were determined at 180 minutes after TNF-α (*P < .05 compared with vehicle; n = 3-5 mice per group).

Figure 4

Effects of coadministration of HU and BAY73-6691 on leukocyte recruitment in TNF-α–treated native SCD mice. HU (100 mg/Kg, intravenously) and BAY73-6691 (3 mg/Kg, intravenously) were administered to native SCD mice at the time of TNF-α administration. (A) leukocyte rolling; (B) adhesion; and (C) extravasation were determined at 180 minutes after TNF-α (**P < .01, ***P < .001 compared with vehicle control, n = 3 mice per group).

Figure 5

Expression of endothelial surface adhesion molecules in exteriorized cremaster muscle tissues of TNF-α–treated SCD mice. Endothelial cell expression of (A) ICAM-1, (B) P-selectin, and (C) E-selectin in cremaster muscle tissues isolated from SCD mice in steady state or after TNF-α administration in the presence or absence of HU and BAY 73-6691 or drug vehicle, n = 37-69 venules PECAM-1⁺ vascular areas quantified from 4-6 mice per group (*P < .05, ***P < .0001 compared with SCD mice in steady state; ###P < .0001 compared with TNF-α–stimulated SCD mice without HU and BAY-73-6691). (D) Representative images of ICAM-1, P-selectin, and E-selectin expression in cremaster muscle isolated from TNF-α–stimulated SCD mice at the end of the in vivo observation period. Scale bar: 20 μm.

Figure 6

Effect of HU and BAY73-6691 treatment on leukocyte activation and survival of TNF-α–treated SCD mice. (A) Quantification of RBC-WBC interactions, n = 5-7 mice per group (*P < .05 compared with vehicle). (B) Mac-1 integrin activity on adherent leukocytes as assessed by capture of fluorescent microspheres, n = 3 to 5 mice per group (**P < .01 compared with vehicle, Mann-Whitney test). Representative images of fluorescent spheres bound to adherent leukocytes after treatment with vehicle (C) or HU and BAY-73-6691 (D). The white arrow indicates the direction of blood flow. (E) Kaplan-Meier survival curve after treatment with HU and BAY73-6691 or vehicle. Log-rank (Mantel-Cox) test (*P < .05 compared with vehicle, n = 5-7 mice per group). Scale bar: 10 μm.

Figure 7

HU reduces leukocyte adhesion and extravasation via NO production and a cGMP-dependent pathway. (A) Leukocyte rolling, (B) adhesion and (C) extravasation in TNF-α–treated C57BL6 WT mice after administration of vehicle or HU in the presence or absence of L-NAME, PTIO, or ODQ, n = 4-10 mice per group (*P < .05, **P < .01, ***P < .001, compared with vehicle alone; ##P < .01, ###P < .001 compared with HU alone; ∞∞P < .01, compared with vehicle + L-NAME).

Figure 8

BAY73-6691 reduces leukocyte adhesion and extravasation via a cGMP-dependent pathway. (A) Leukocyte rolling, (B) adhesion, and (C) extravasation in TNF-α–treated C57BL6 WT mice after administration of vehicle or BAY73-6691, in the presence or absence of KT5823 (n = 4-10 mice per group; *P < .05, ***P < .001, compared with vehicle alone; #P < .05, ###P < .001, compared with BAY73-6691 alone).