Role of channels in the O2 permeability of murine red blood cells II. Morphological and proteomic studies

Abstract

In this second of three papers, we examine red blood cell (RBC) morphometry and RBC-membrane proteomics from our laboratory mouse strain (C57BL/6Case). In paper #1, using stopped-flow absorbance spectroscopy to ascertain the rate constant for oxyhemoglobin (HbO2) deoxygenation (kHbO2), we find substantial kHbO2 reductions with (1) membrane-protein inhibitors p-chloromercuribenzenesulfonate (pCMBS) or 4,4-diisothiocyanatostilbene-2,2-disulfonate (DIDS); (2) knockouts of aquaporin-1 (AQP1-KO), or Rhesus blood-group-associated A-glycoprotein (RhAG-KO), or double knockouts (dKO); or (3) inhibitor+dKO. In paper #3, reaction-diffusion mathematical modeling/simulations reveal that kHbO2 could fall secondary to slowed intracellular O2/HbO2/Hb diffusion. Here in paper #2, blood smears as well as still/video images and imaging flow cytometry (IFC) of living RBCs show that [~]97.5% to [~]98.6% of control (not drug-treated) cells are biconcave disks (BCDs) across all genotypes. Pretreatment with pCMBS raises non-BCD abundance to [~]8.7% for WT and [~]5.7% for dKO; for DIDS pretreatment, the figures are [~]41% and [~]21%, respectively. Modeling (paper #3) accommodates for these shape changes. Light-scattering flow cytometry shows no significant difference in RBC size or shape among genotypes. IFC reveals minor differences among genotypes in RBC major diameter (OMajor), which (along with mean corpuscular volume, paper #1) yields RBC thickness for simulations in paper #3. Label-free liquid chromatography/tandem mass spectrometry (LC/MS/MS) proteomic analyses of RBC plasma-membrane ghosts confirm the deletion of proteins targeted by our knockouts, and rule out changes in the 100 proteins of greatest inferred abundance. Thus, genetically induced changes in kHbO2 must reflect altered abundance of AQP1 and /or the Rh complex. Key PointsO_LIO2-offloading from red blood cells (RBCs) depends not only on membrane O2 permeability and oxyhemoglobin dissociation, but also on RBC size and shape. In this second of three papers, we use blood smears, still/video images of living RBCs, and imaging flow cytometry to examine morphometry of RBCs from paper #1. C_LIO_LIWe find that mouse RBCs of all genotypes--wild-type, aquaporin-1 knockout (AQP1-KO), Rhesus blood group-associated A-glycoprotein knockout (RhAG-KO), and double knockout--are dominantly biconcave discs, with [~]1.4% to [~]2.5% poikilocytosis (shape change, SC). Drug pre-treatment increases %SC. C_LIO_LIUsing label-free liquid chromatography/tandem mass spectrometry to assess apparent abundance of RBC-ghost proteins, we find no significant differences among genotypes for any of the [~]100 most abundant protein species except, as appropriate, AQP1, RhAG, or Rhesus blood group D antigen. C_LIO_LIThus, the substantial effects observed in paper #1 cannot be attributed to differences in morphometry or protein content. C_LI