Abstract 2120

In sickle cell disease (SCD), the mitogen-activated protein kinase (MAPK) ERK1/2 is constitutively active and can be inducible by agonist-stimulation only in sickle but not in normal human erythrocytes. ERK1/2 is involved in activation of ICAM-4-mediated sickle red blood cell (SSRBC) adhesion to the endothelium. However, other effects of the ERK1/2 activation in SSRBCs leading to the complex SCD pathophysiology, such as alteration of RBC hemorheology are still unknown. To further characterize global ERK1/2-induced changes in membrane protein phosphorylation within human RBCs, a label-free quantitative phosphoproteomic analysis was applied to sickle and normal RBC membrane ghosts pre-treated with U0126, a specific inhibitor of MEK1/2, the upstream kinase of ERK1/2 activation, in the presence or absence of recombinant active ERK2. Across eight unique treatment groups, 375 phosphopeptides from 155 phosphoproteins were quantified with an average technical coefficient of variation in peak intensity of 19.8%. Consistent with other RBC membrane phosphorylation studies, the phosphoproteins of SSRBC membrane ghosts with the highest number of uniquely phosphorylated peptides (>10), were ankyrin-1 of the ankyrin complex (n=33), spectrin β chain of the cytoskeleton network (n=15), and proteins of the junctional complex involved in binding integral membrane proteins to cytoskeletal proteins, including α- and β-adducins (n=22 and n=18, respectively), dematin (n=16) and protein 4.1 (n=17). In addition, several other phosphoproteins with >5 unique phosphorylated peptides, affecting RBC shape, flexibility, anion transport and protein trafficking, and adhesion, all of which contribute to the pathophysiology of SCD, were also observed. However, the MEK1/2 inhibitor U0126 was able to significantly down-regulate 37 unique RBC membrane phosphopeptides (from 21 unique phosphoproteins) in SSRBCs. We found that MEK1/2-dependent ERK1/2 activation in SSRBCs affected membrane-bound proteomes of both the junctional and ankyrin complexes, including dematin, α-adducin, β-adducin with phosphorylation of residues within the ERK1/2 consensus motif, and glycophorin A. MEK1/2/ERK1/2 signaling in SSRBCs induced changes within the actins/spectrins network as well by affecting phosphorylation of β-spectrins. Furthermore, the peptide metabotropic glutamate receptor 7 (mGlu7) also underwent serine phosphorylation at the ERK consensus motif. This could explain the rate of active glutamate transport in these cells. Significant changes only in membrane ghosts prepared from SSRBCs treated with U0126 or after addition of exogenous active ERK2 to these membrane ghosts, were also observed in the status of leucine-rich repeats and immunoglobulin-like domains protein 2, leucine-zipper-like transcriptional regulator 1, glucose transporter 1, and adenylyl cyclase-associated protein 1 (CAP1), which may potentially disturb degradation of misfolded glycoproteins and receptor ubiquitination, protein transcription, glucose transport and cAMP production, respectively. These data also suggest that a negative regulatory mechanism might exist in normal cells to prevent activation of ERK1/2-dependent phosphorylation of these membrane proteins. Among all these phosphorylated proteomes, glycophorin A was the most affected protein in SSRBCs by this ERK1/2 pathway, which contained 12 unique phosphorylated peptides, suggesting that in addition to its effect on sickle RBC adhesion, increased glycophorin A phosphorylation via the ERK1/2 pathway may also affect glycophorin A interactions with band 3, which could result in decreased in both anion transport by band 3 and band 3 trafficking. The abundance of thirteen of the thirty-seven phosphopeptides was subsequently increased in normal RBCs co-incubated with recombinant ERK2, and therefore represent specific MEK1/2 phospho-inhibitory targets mediated via ERK2. These findings expand upon the current model for the involvement of ERK1/2 signaling in RBCs. These findings also identify additional protein targets of this pathway other than the RBC adhesion molecule ICAM-4 and enhance the understanding of the mechanism of small molecule inhibitors of MEK/1/2/ERK1/2, which could be effective in ameliorating RBC hemorheology and adhesion, the hallmarks of SCD.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.