Professional Documents
Culture Documents
Red Cell Membrane Remodeling in Sickle Cell Anemia: Sequestration of Membrane Lipids and Proteins in Heinz Bodies
Red Cell Membrane Remodeling in Sickle Cell Anemia: Sequestration of Membrane Lipids and Proteins in Heinz Bodies
30 Liu et al.
Figure 1. Phospholipid la-
beling of Heinz bodies in
(A) sickle red cells, (B)
sickle red cell ghosts, and
(C) sickle red cells
quenched using albumin.
Sickle red cells were la-
beled with NBD-PC or
NBD-PS (final concentra-
tion, 1.25 mM) at 378C for
30 min. The labeled cells
were lysed or quenched us-
ing 4% delipidated albumin
and examined by fluores-
cence (Fluo) and phase-
contrast microscopy. Note
that NBD-lipids labeled
Heinz bodies in both intact
cells and ghosts with
higher intensity than the
plasma membrane. The la-
beling of Heinz bodies was
not quenched by albumin
treatment of intact cells.
questered at sites of Heinz bodies. First, we asked whether ies attached to the plasma membrane, NBD-PS or NBD-PC la-
NBD-PS and NBD-PC were capable of interaction with dena- beled the whole Heinz body but not the membrane site at
tured, cell-free hemoglobin S. Acetylphenylhydrazine was which the Heinz body was attached (Fig. 4 B). Confocal sec-
used to denature the ghost-free sickle cell hemolysate and to tioning through the mass of the Heinz body (arrow, Fig. 4 A)
produce a hemoglobin S precipitate which was readily visible showed that the fluorescence intensity was greater in the cen-
by phase-contrast microscopy. Exposure of the hemoglobin S ter than at the edge of the particle, suggesting that the particles
precipitate to either NBD-PS or NBD-PC resulted in distribu- represented Heinz bodies and not endocytic lipid vesicles and
tion of the fluorescent label over the entire surface of the pre- that the NBD-lipid was incorporated within the Heinz bodies,
cipitate (Fig. 2). This finding raised the possibility that, in in- probably on the hydrophobic surface of denatured hemoglobin.
tact cells, a fraction of bilayer lipids was sequestered within Finally, two different membrane lipid analogues were used
sickle red cells and deposited on the surface of intracellular ag- to study the molecular mechanism by which exogenously
gregates of denatured hemoglobin S (Heinz bodies). added NBD-lipids colocalized with Heinz bodies in sickle red
Second, we took advantage of the observation that some of cells. The first lipid probe was fluorescein-PE, a phospholipid
the Heinz bodies in sickle red cell ghosts were detached from analogue with a polar fluorescein moiety attached to the
the membrane and exhibited distinct Brownian motion. Suc- amine function of the PE head group. Fluorescein-PE has a
cessive fluorescence and phase-contrast images of the same much slower rate of exchange between membranes than do
sickle red cell ghost labeled with NBD-PS (Fig. 3) or NBD-PC acyl chain–labeled lipid analogues, such as NBD-PS and NBD-
(data not shown) showed that the fluorescent phospholipid an- PC, possibly because fluorescein-PE has a slow rate of “flip-
alogue comigrated with the Heinz bodies. These data directly flop” between lipid bilayer leaflets (27–30). Fluorescein-PE la-
suggested that a fraction of bilayer phospholipids was internal- beled the sickle cell surface but not the Heinz bodies (Fig. 5),
ized and sequestered within the cells. Third, fluorescence im-
aging of optical sections of NBD-lipid labeled sickle red cells
was used to define further the association between NBD-lipids
and Heinz bodies. In these confocal microscopy images, fluo-
rescently labeled phospholipids colocalized with the entire
Heinz body in the cytoplasm of cells (Fig. 4 A). For Heinz bod-
suggesting that translocation of a lipid probe from the outer to red cells (Fig. 5), suggesting that the surface of Heinz bodies
the inner lipid bilayer leaflet was a prerequisite for subsequent provided a hydrophobic environment on which membrane lip-
lipid removal from the plasma membrane and sequestration in ids were internalized and sequestered.
Heinz bodies. Consistent with this interpretation, albumin Internalization and colocalization of band 3 protein, spec-
treatment of fluorescein-PE–labeled sickle red cells resulted in trin, ankyrin, protein 4.1, and Heinz bodies within sickle red
a . 95% decrease in plasma membrane fluorescence and did cells. We next examined whether the intracellular sequestra-
not reveal sites of punctate labeling (data not shown). The sec- tion of sickle red cell membrane phospholipids was accompa-
ond membrane lipid probe was DiI, an amphiphilic carbocya- nied by a similar redistribution of membrane proteins. Cells
nine with two long alkyl tails (28, 31) that translocates readily were labeled with NBD-PS and lysed, and the resulting ghosts
across plasma and internal membranes (28–30, 32). DiI- were double labeled with antibodies against band 3 protein,
labeled Heinz bodies as well as the plasma membrane of sickle spectrin, ankyrin, or protein 4.1. Band 3, spectrin, ankyrin, and
protein 4.1 were all observed to colocalize with the NBD-PS–
Figure 5. Fluorescein- labeled Heinz bodies (Fig. 6). No immunostaining was seen
PE and DiI labeling of when the ghosts were incubated with preimmune rabbit serum
sickle red cells. Sickle and then labeled with the second (RITC-conjugated) antibody
red cells were labeled
(data not shown).
with fluorescein-PE
(Fluo-PE) (final con-
Several approaches were used to verify that the bulk of the
centration, 63 mM) or morphologically punctate band 3 protein staining, like that of
DiI (final concentra- NBD-PS and NBD-PC, was internalized and sequestered in
tion, 0.14 mM) at 378C cells rather than clustered in plasma membrane regions overly-
for 30 min. The labeled ing membrane-associated Heinz bodies. First, sickle red cells
cells were lysed and ex- were labeled with eosin-5-maleimide, which binds to Lys-430
amined by fluorescence at an exoplasmic loop of the band 3 protein and which does
(Fluo) and phase-con- not permeate across the intact red cell membrane (33). Eosin-
trast microscopy. Note labeled cells showed a uniform distribution of band 3 protein
that Fluo-PE labeled
over the entire surface of the membrane and a notable absence
the plasma membrane
but not the Heinz bod-
of clustering over membrane areas adjacent to Heinz bodies
ies. In contrast, DiI la- (Fig. 7). Second, confocal fluorescence microscopy and poly-
beled both the plasma clonal anti–band 3 antibodies were used to visualize band 3
membrane and the clusters in permeabilized sickle red cell ghosts. Band 3 protein
Heinz bodies. in the plasma membrane of such cells was uniformly distrib-
32 Liu et al.
Figure 7. Eosin-male-
imide labeling of band 3
in intact sickle red cells.
Sickle red cells (RBC)
were incubated with
eosin-5-maleimide
(0.25 mg/ml) in KPBS.
Cells were then washed
three times in KPBS
with 1% BSA. Eosin-
labeled cells and mem-
brane ghosts, prepared
by hypotonic lysis, were
examined by fluores-
cence (Fluo) and phase-
contrast microscopy.
Note that the eosin la-
bel was distributed uni-
formly on the cell sur-
face; there was a
notable absence of clus-
tering of eosin label
over membrane areas
adjacent to Heinz
bodies.
pholipid analogues NBD-PS and NBD-PC, and to the mem- Heinz bodies in sickle red cells. Third, when visualized by con-
brane lipid probe DiI, produces a punctate labeling pattern. focal fluorescence microscopy, both the fluorescent phospho-
Neither NBD-PS nor NBD-PC can be extracted by albumin lipid analogues and the labeled band 3 protein are found in as-
from the sites of punctate labeling, suggesting that the phos- sociation with the entire mass of the Heinz body. The latter
pholipid analogues are internalized within the cells. Second, finding can also be replicated by exposing a denatured precipi-
NBD-PS, NBD-PC, and immunofluorescently labeled band 3 tate of cell-free HbS to NBD-PS or NBD-PC. Fourth, the fluo-
protein, spectrin, ankyrin, and protein 4.1 all colocalize with rescently labeled clusters of NBD-PS or NBD-PC comigrate
with Heinz bodies during their Brownian motion.
Figure 9. NBD-PS and anti–band 3 (Anti-B3) antibody labeling of Figure 10. NBD-PS and anti–band 3 antibody labeling of Heinz bod-
Heinz bodies in ghosts from APH-treated red cells. Normal red cells ies in ghosts from red cells containing unstable hemoglobin E. Red
were pretreated with APH (15 mM, 378C, 15 min) and subsequently cells from a homozygous hemoglobin E individual were labeled with
incubated at 378C for 48 h without APH. The cells were then labeled NBD-PS, washed, and hypotonically lysed. The resulting ghosts were
with NBD-PS and hypotonically lysed, and the ghosts were indirectly indirectly labeled with rabbit anti–band 3 (Anti-B3) antibody and
labeled with anti–band 3 antibodies as described in the legend to Fig. RITC-conjugated goat anti–rabbit IgG. Note that the Heinz body la-
6. Note that band 3 colocalized with NBD-PS in Heinz bodies. Fluo, beling by NBD-PS and anti–band 3 antibody was similar to that ob-
fluorescence. served in ghosts from sickle red cells. Fluo, fluorescence.
34 Liu et al.
Previous studies have shown that red cells containing phosphatidylcholine in sickled erythrocytes: a reversible process. J. Biol. Chem.
258:8435–8442.
Heinz bodies exhibit clustering of band 3 protein and of 6. Franck, P. F. H., E. M. Bevers, B. H. Lubin, P. Comfurius, D. T. Y. Chiu,
ankyrin at sites of attachment of the Heinz bodies to the mem- J. A. F. Op den Kamp, R. F. A. Zwaal, L. L. M. van Deenen, and B. Roelofsen.
brane. Such sites are postulated to create a senescent red cell 1985. Uncoupling of the membrane skeleton from the lipid bilayer: the cause of
accelerated phospholipid flip-flop leading to an enhanced procoagulant activity
antigen that binds autologous IgG which, in turn, targets the of sickled cells. J. Clin. Invest. 75:183–191.
red cells for premature destruction by cells of the reticuloen- 7. Platt, O. S., J. F. Falcone, and S. E. Lux. 1985. Molecular defect in the
dothelial system (10, 11, 15–18). Although our data are not in- sickle erythrocyte skeleton: abnormal spectrin binding to sickle inside-out vesi-
cles. J. Clin. Invest. 75:266–271.
consistent with the suggestions that microaggregates of band 3 8. Low, P. S., S. M. Waugh, K. Zinke, and D. Drenckhahn. 1985. The role of
are found at such sites, we show here that the band 3 redistri- hemoglobin denaturation and band 3 clustering in red blood cell aging. Science
bution in sickle red cells is considerably more profound than (Wash. DC). 227:531–533.
9. Waugh, S. M., and P. S. Low. 1985. Hemichrome binding to band 3: nu-
that associated with clustering within the plane of the mem- cleation of Heinz bodies on the erythrocyte membrane. Biochemistry. 24:34–39.
brane. It appears that a fraction of band 3 molecules is inter- 10. Waugh, S. M., B. M. Willardson, R. Kannan, R. J. Labotka, and P. S.
nalized and sequestered within the cells. Low. 1986. Heinz bodies induce clustering of band 3, glycophorin, and ankyrin
in sickle cell erythrocytes. J. Clin. Invest. 78:1155–1160.
Red cells containing Heinz bodies exhibit a reduced sur- 11. Schlüter, K., and D. Drenckhahn. 1986. Co-clustering of denatured he-
face area. The surface area deficiency has been attributed to moglobin with band 3: its role in binding of autoantibodies against band 3 to ab-
the removal of Heinz bodies, together with the overlying normal and aged erythrocytes. Proc. Natl. Acad. Sci. USA. 83:6137–6141.
12. Corbett, J. D., and D. E. Golan. 1993. Band 3 and glycophorin are pro-
plasma membrane, by cells of the reticuloendothelial system gressively aggregated in density-fractionated sickle and normal red blood cells:
including splenic macrophages (34). Our data are consistent evidence from rotational and lateral mobility studies. J. Clin. Invest. 91:208–217.
with the hypothesis that the loss of membrane surface area 13. Liu, S. C., L. H. Derick, S. Zhai, and J. Palek. 1991. Uncoupling of the
spectrin-based skeleton from the lipid bilayer in sickled red cells. Science
within the Heinz body–containing cells may also result from (Wash. DC). 252:574–576.
sequestration of a fraction of membrane lipids and integral 14. Hebbel, R. P. 1990. The sickle erythrocyte in double jeopardy: autoxida-
membrane proteins. Reticuloendothelial remodeling of such tion and iron decompartmentalization. Semin. Hematol. 27:51–69.
15. Lutz, H. U., R. Flepp, and G. Stringaro-Wipf. 1984. Naturally occurring
cells is likely to involve removal of the entire Heinz body–asso- autoantibodies to exoplasmic and cryptic regions of band 3 protein, the major
ciated lipid/protein complexes. integral membrane protein of human red blood cells. J. Immunol. 133:2610–
Surface area deficiency has also been demonstrated in irre- 2618.
16. Kannan, R., R. Labotka, and P. S. Low. 1988. Isolation and character-
versibly sickled red cells (35, 36). These permanently de- ization of the hemichrome-stabilized membrane protein aggregates from sickle
formed cells are dehydrated, dense, and rigid (1, 36, 37), and erythrocytes: major site of autologous antibody binding. J. Biol. Chem. 263:
the permanent deformation has been reproduced in irrevers- 13766–13773.
17. Turrini, F., P. Arese, J. Yuan, and P. S. Low. 1991. Clustering of integral
ibly sickled red cells ghosts and membrane skeletons (38). We membrane proteins of the human erythrocyte membrane stimulates autologous
propose that the surface area deficiency of irreversibly sickled IgG binding, complement deposition and phagocytosis. J. Biol. Chem. 266:
red cells is, at least in part, caused by membrane lipid and pro- 23611–23617.
18. Turrini, F., F. Mannu, P. Arese, J. Yuan, and P. S. Low. 1993. Character-
tein sequestration at sites of Heinz bodies. This process is ization of the autologous antibodies that opsonize erythrocytes with clustered
likely to be compounded by a loss of membrane surface area integral membrane proteins. Blood. 81:3146–3152.
mediated by the release of skeleton-free membrane spicules 19. Chiu, D., B. Lubin, and S. B. Shohet. 1979. Erythrocyte membrane lipid
reorganization during the sickling process. Br. J. Haematol. 41:223–234.
(13). 20. Schlegel, R. A., B. M. Phelps, A. Waggoner, L. Terada, and P. William-
son. 1980. Binding of merocyanine 540 to normal and leukemic erythroid cells.
Cell. 20:321–328.
Acknowledgments 21. Williamson, P., W. Massey, B. M. Phelps, and R. A. Schlegel. 1981.
Membrane phase state and the rearrangement of hematopoietic cell surface re-
ceptors. Mol. Cell Biol. 1:128–135.
We thank Dr. Wendy Reenstra at the Boston University School of 22. Stillwell, W., S. R. Wassall, A. C. Dumaual, W. D. Ehringer, C. W.
Medicine for assistance in using the confocal laser scanning micro- Browning, and L. J. Jenski. 1993. Use of merocyanine (MC540) in quantifying
scope. We also thank Diana Fraser and Donna Marie Mironchuk for lipid domains and packing in phospholipid vesicles and tumor cells. Biochim.
manuscript preparation. Biophys. Acta. 1146:136–144.
23. Choi, H. R., R. A. Schlegel, E. Rubin, P. Williamson, and M. P. Wester-
This work was supported by National Institutes of Health re-
man. 1986. Alteration of red cell membrane organization in sickle cell anemia.
search grants HL-15157 (to J. Palek, S. C. Liu, and D. E. Golan) and Br. J. Haematol. 63:761–773.
HL-32854 (to D. E. Golan). 24. Kremer, J. M., M. W. Esker, C. Pathmamanoharan, and P. H.
Wiersema. 1977. Vesicles of variable diameter prepared by modified injection
method. Biochemistry. 16:3932–3935.
25. Derick, L. H., S. C. Liu, A. H. Chishti, and J. Palek. 1992. Protein immu-
References nolocalization in the spread erythrocyte membrane skeleton. Eur. J. Cell Biol.
57:317–320.
1. Hebbel, R. P. 1991. Beyond hemoglobin polymorphism: the red cell 26. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer
membrane and sickle disease pathophysiology. Blood. 77:214–237. of protein from polyacrylamide gels to nitrocellulose sheets: procedure and
2. Lubin, B., D. Chiu, J. Bastacky, B. Roelofsen, and L. L. M. van Deenen. some applications. Proc. Natl. Acad. Sci. USA. 76:4350–4354.
1981. Abnormalities in membrane phospholipid organization in sickled eryth- 27. Struck, D. K., and R. E. Pagano. 1980. Insertion of fluorescent phospho-
rocytes. J. Clin. Invest. 67:1643–1649. lipids into the plasma membrane of mammalian cell. J. Biol. Chem. 255:5404–
3. Middelkoop, E., B. H. Lubin, E. M. Bevers, J. A. F. Op den Kamp, P. 5410.
Comfurius, D. T. Y. Chiu, R. F. A. Zwaal, L. L. M. van Deenen, and B. Roelof- 28. Haugland, R. P. 1992. Handbook of Fluorescent Probes and Research
sen. 1988. Studies on sickled erythrocytes provide evidence that the asymmetric Chemicals. Molecular Probes, Inc., Eugene, OR. 421 pp.
distribution of phosphatidylserine in the red cell membrane is maintained by 29. Golan, D. E., C. S. Brown, C. M. L. Cianci, S. T. Furlong, and J. P.
both ATP-dependent translocation and interaction with membrane skeletal Caulfield. 1986. Schistosomula of Schistosoma mansoni use lysophosphatidyl-
proteins. Biochim. Biophys. Acta. 937:281–288. choline to lyse adherent human red blood cells and immobilize red cell mem-
4. Mohandas, N., M. Rossi, S. Bernstein, S. Ballas, Y. Ravindranath, J. brane components. J. Cell Biol. 103:819–828.
Wyatt, and W. Mentzer. 1985. The structural organization of skeletal proteins 30. Caulfield, J. P., C. P. Chiang, P. W. Yacono, L. A. Smith, and D. E. Go-
influences lipid translocation across erythrocyte membrane. J. Biol. Chem. 260: lan. 1991. Low density lipoproteins bound to Schistosoma mansoni do not alter
14264–14268. the rapid lateral diffusion or shedding of lipids in the outer surface membrane.
5. Franck, P. F. H., D. T. Y. Chiu, J. A. F. Op den Kamp, B. Lubin, L. L. M. J. Cell Sci. 99:167–173.
van Deenen, and B. Roelofsen. 1983. Accelerated transbilayer movement of 31. Sims, P. J., A. S. Waggoner, C. H. Wang, and J. F. Hoffman. 1974. Stud-
36 Liu et al.