From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Oxidation-Induced Changes in Microrheologic Properties
of the Red Blood Cell Membrane
By Robert P. Hebbel, Andrew Leung, and Narla Mohandas
It has been hypothesized that some of the irreversible
microrheologic abnormalities of sickle red blood cell (RBC)
membranes could result from autoxidative perturbation.
To model this possibility, we used micromechanicalmanipulation t o examine the static extensional rigidity and
inelastic or plastic behavior of normal RBCs exposed to
phenazine methosulfate (PMS), an agent that generates
superoxide from within the cell. In response to this stress,
RBC membranes became stiff as evidenced by increasing
extensional rigidity. At 50 pmol/L PMS they were as stiff
as the membranes of most dense, dehydrated sickle RBCs;
and at 25 pmol/L PMS the membranes were similar t o
somewhat less dense sickle RBCs. When examined for
inelastic behavior, RBCs exposed to PMS even at 10
pmol/L showed hysteresis in loading and unloading phases
of the curve relating aspiration length t o suction pressure,
and they developed membrane bumps that persisted after
RBC release from the pipette. Examination of single cells in
both isotonic and hypotonic buffers showed that the effect
of PMS on RBC microrheology is not mediated by cellular
T
HE CLINICAL COURSE of sickle cell anemia is
punctuated by episodic vasocclusive events caused by
the abnormal microcirculatory behavior of sickle blood. This
pathophysiology results, a t least in part, from the presence of
poorly deformable erythrocytes (RBCs) characterized by
defects of cellular microrheology that are attributable to
changes in both cell cytoplasm and membrane organization.
The former, an increase in cytoplasmic viscosity due to
cellular dehydration, is thought to be a dominant factor
responsible for decreased deformability of sickle cells.’-3
However, recent microrheologic data also have implicated
unusual membrane rigidity stemming from a reversible
association of sickle hemoglobin with the membrane, as well
as from irreversible changes in membrane skeletal protein
~rganization.~.~
For dense, dehydrated sickle RBCs, altered membrane
rheology due to irreversible membrane changes is most
evident in the persistence of membrane deformation after the
removal of deforming forces (plastic flow) and the continued
manifestation of abnormally increased extensional static
rigidity despite normalization of cell hydration statu^.^ The
biochemical and structural basis for these phenomena has
not been delineated yet. Rank et a15 previously speculated
that the abnormal thiol oxidation of cytoskeletal proteins in
the sickle RBC membrane might account for this plastic
behavior. More recently, it has been shown that there is a
correlation between membrane thiol oxidation and membraneassociated heme in sickle R B C S . Importantly,
~
the extent of
both thiol oxidation’ and membrane-associated heme6 is
highest in most dense sickle RBCs, the population of cells
exhibiting the most marked abnormalities in membrane
rheologic properties.
To determine whether irreversible rheologic abnormalities
of sickle membranes could result from oxidation-induced
changes in membrane organization, we have examined the
Blood, Vol 76. No 5 (September 1). 1990: pp 1015-1020
dehydration. Independent confirmation of the membrane
stiffening effect of PMS was obtained by ektacytometric
analysis of resealed RBC ghosts, with sickle-like increases
in membrane rigidity observed between 50 and 100 pmol/L
PMS. The rigidity of these ghosts was partially ameliorated
by exposure to a thiol reductant. In terms of biochemical
abnormalities, treated RBCs became significantly different
from control RBCs at 25 pmol/L PMS, at which point they
just began t o enter the sickle range for amounts of
membrane thiol oxidation and membrane-associatedheme.
The sickle average was achieved at 50 pmol/L PMS (for
thiol oxidation) to 100 pmol/L PMS (for membrane heme).
Thus, micromolar concentrations of PMS induce abnormalities of membrane microrheology that closely mimic those
of unmanipulated sickle RBCs while reproducing similar
degrees of oxidative biochemicalchange. We conclude that
membrane protein oxidation could explain existence of an
irreversible component t o the abnormal rheology of the
sickle membrane.
0 1990by The American Society of Hematology.
microrheology of normal RBCs in which the excessive
autoxidation of sickle RBCs*-’’ is simulated by incubation
with phenazine methosulfate, an agent that stimulates intracellular generation of superoxide.”
MATERIALS AND METHODS
Cell preparation. Fresh citrated blood (citrate:bld = 1:9)
was obtained from volunteer normal donors. Without delay, RBCs
were washed four times with PBSG (1 part 0.12 mol/L sodium
phosphate buffer, 3 parts 0.154 mol/L NaCI, 1 g/L glucose, pH 7.4)
with removal of buffy coat. To stimulate intracellular generation of
superoxide,” we incubated normal RBCs with various concentrations of phenazine methosulfate (PMS) at hematocrit 5% in PBSG
for 60 minutes at 37OC, after which RBCs were thoroughly washed
with PBSG. They were then used for biochemical assays, or they
were suspended in PBS (10 mmol/L phosphate, pH 7.4, NaCl to 290
mOsmol/L) for micropipette studies.
Biochemical assays. Several parameters of oxidative damage
were evaluated to monitor the effect of incubating RBCs with PMS.
Amount of thiol oxidation in sodium dodecyl sulfate-solubilized
~~~~
From the Department of Medicine, University of Minnesota,
Minneapolis; and the Division of Cell and Molecular Biology,
Lawrence Berkeley Laboratory, Berkeley, CA.
Submitted November 27,1989; accepted May 1.1990.
Supported by grants from the National Institutes of Health (HL
30160 and HL 31579).
Address reprint requests to Robert P. Hebbel. MD, University of
Minnesota, Department of Medicine, Box 480 UMHC, Harvard St
at East River Rd, Minneapolis, MN 55455.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C.section 1734 solely to
indicate this fact.
0 1990 by The American Society of Hematology.
0006-4971/90/7605-0009$3.00/0
1015
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1016
HEBBEL, LEUNG, AND MOHANDAS
membrane proteins was measured both by titrating reduced thiols
using 2,2-dithiodipyridine and by thiol-disulfide exchange
chromatography.’ Amount of peroxidation was estimated by measurement of thiobarbituric acid reactive substances (TBARS), with
correction for any contaminating chromogens as previously
described.I2Amount of truly membrane-associated heme and heme
protein was measured spectrophotometrically on inside-out membranes (IOM) solubilized in formic acid.6
Microrheologic measurements. Micropipette aspiration technique was used to evaluate the contribution of oxidative changes in
the RBC membrane to its rheologic properties. In this study we focus
on the extensional properties of the membrane and on the recoverability of cell shape after deformation. Details of the micromechanical
system and analysis used to measure RBC mechanical properties
have been described previou~ly.~~~
Briefly, the membrane extensional rigidity ( p ) was derived from
observation of cell length (L) aspirated into a small micropipette
(with radius Rp) in response to the increase in suction pressure (P).
Analysis of this experiment has shown that p is proportional to the
derivative of the pressure with respect to length: p a Rp2 x dP/dL.
Inelastic or plastic behavior of the membrane was evaluated by
comparing RBC mechanical behavior during the progressive increase in aspiration pressure (loading phase) to the behavior observed when the aspiration pressure was slowly reduced (unloading
phase). For a truly elastic material, the curves describing the
force/deformation relationship for loading and unloading phases are
very similar; ie, there is no hysteresis. In contrast, for materials that
exhibit inelastic or plastic behavior, the loading and unloading
curves are different, and hysteresis is evident. This can be illustrated
dramatically by persistent, residual bumps on the cell surface after
release of applied forces.
To determine whether altered membrane properties were due to
changes in cell hydration state, we performed mechanical tests on
PMS-treated RBCs in both isotonic buffer (290 mOsmol/L) and
hypotonic buffer (210 mOsmol/L). For accurate comparison, the
tests were performed on the same cell, which was transferred by
micromanipulation from one chamber on the microscope stage
containing isotonic buffer to an adjacent chamber containing hypotonic buffer. Transfers were accomplished by insertion of each cell
into a large bore pipette that was drawn through the air-water
interfaces.
We also quantified dynamic membrane deformability of normal
and oxidized RBCs using the ektacytometer. For these measurements, resealed RBC ghost membranes were suspended in 3 mL of
Stractan I1 (St Regis Paper Co, Tacoma, WA) (290 mOsmol/L, 22
cp, pH 7.4) and exposed to increasing shear stress as previously
described.” Analysis of the relationship between deformability index
and shear stress provides a measure of membrane def0rmabi1ity.I~It
should be noted that unlike the micropipette method, which generates an accurate measure of elastic modulus, the membrane deformability determined by ektacytometry is a complex property influenced by a number of material properties including elastic and
bending moduli, as well as membrane viscosity and surface area.
RESULTS
Biochemical sequelae of oxidation. RBCs exposed to all
concentrations of P M S retained normal morphology with
smooth biconcave contour. However, the effect of PMS on
membrane protein was evident in a diminished number of
titratable reduced thiols in RBC ghosts as the P M S concentration increased (Fig 1). A comparable dose-response relationship was observed for thiol oxidation measured by a
different technique, thiol-disulfide exchange chromatogra-
=
2
100- 0-
c)
C
8
90-
r
0
ap
v
-ul0
IU
s
Q)
80 7060-
U
2
50-
t
2-
1 -
O
k
9
L I
0 2.5
I
I
1
I
5
10
25
50
-
1
-
100 Sickle
PMS (umol/l)
Fig 1. Biochemical alterations induced by PMS. After exposure
of normal RBCs to PMS at the indicated concentrations, the
number of titratable reduced thiols (0)and the amount of generic
were determined in 3 and 5 experiments,
heme on IOM
respectively. The asterisk indicates values that have become
significantly different from control RBCs. For comparison, thiol
(n = 8 ) and IOM heme (n = 311 values previously obtained’,’ for
unmanipulated sickle RBCs are shown at the far right. All data are
shown as mean ? SD.
phy (data not shown). In addition, the amount of membraneassociated heme increased progressively as PMS concentration was increased (Fig 1). Little change in these biochemical
parameters was observed after exposure of RBCs to PMS up
to 10 pmol/L, but both effects began to enter the sickle range
(and became significantly different from control) a t 25
pmol/L PMS. The average degree of biochemical alteration
observed in native, unmanipulated sickle cell membranes was
fully reproduced a t 50 pmol/L P M S for thiol oxidation and
a t 100 pmol/L PMS for membrane heme (Fig 1). The extent
of lipid peroxidation resulting from PMS exposure somewhat
exceeded physiologically relevant amounts. For example, 10
and 100 pmol/L PMS induced 12.1
1.7 and 38.7 2 13.0
nmol TBARS per ml RBC in 60 minutes, the former amount
slightly exceeding that spontaneously generated from sickle
RBCs in 20 hours.14
Micropipette study of normal and oxidized RBCs.
Measured values for static extensional rigidity (p) of normal
and PMS-treated RBCs are shown in Table 1. RBCs exposed
to P M S at concentrations up to 10 pmol/L remained similar
to normal control cells, but a t higher concentrations of PMS,
the shear modulus deteriorated ( p increased) as dose escalated. Indeed, the shear modulus identified for normal RBCs
exposed to 50 pmol/L P M S is comparable with that of dense,
dehydrated sickle RBCs (previously found to have p = 21 *
7.8 x
dyne/cm), and our result for RBCs exposed to 25
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1017
OXIDATION AND RED CELL MEMBRANE RHEOLOGY
Table 1. Static Extensional Rigidity of PMS-Treated RBCs
PMS Dose
Iwnol/L)
Control
2.5
10
25
50
100
Shear Modulus,p
(x
dyne/cm)
8.1
9.4
7.5
11.1
21.2
53.5
+
1.7(10)
f 0.5 (5)
+ 0.8 (5)
f 2.1 (5)
f 5.3 (12)
f 38.5 (12)
Values shown as mean +SD (number of cells examined). Results are
here combined for two separate experiments using different normal
donors.
pmol/L PMS is similar to that reported for somewhat less
dense sickle RBCS.~
During the micromechanical measurement of membrane
extensional rigidity, we observed that the curves relating
aspirated length to suction pressure described different
functional behavior for normal and oxidized RBCs. For
normal cells, the force/deformation relationship remained
constant during increasing aspiration pressure (loading phase)
and during decreasing aspiration pressure (unloading phase)
(Fig 2A). This absence of hysteresis indicates truly elastic
membrane behavior. In contrast, for RBCs treated with
PMS, the curve describing aspirated length versus suction
pressure was markedly different during loading and unloading phases (Fig 2, B and C), indicating that oxidized
membranes exhibit inelastic or plastic-like behavior.
This plastic-like behavior also was manifested by residual,
persistent bumps observed on the RBC membrane surface
after its release from the pipette following extensional
deformation (Fig 3). Indeed, these residual bumps developed
in every RBC that had been exposed to PMS at 10 pmol/L or
higher concentrations, while none of the normal RBCs
exhibited them. Interestingly, we thus found that oxidized
cells exhibited plastic-like behavior at PMS concentrations
that did not increase the shear modulus value. This is evident,
for example, in that the curve for the representative RBC
depicted in Fig 2B shows some hysteresis but preserves a
normal slope ( p ) .
To critically evaluate the potential contribution of any
cellular dehydration to the observed properties of oxidized
RBCs, we studied the effect of increased hydration on their
mechanical properties. The PMS-treated RBCs were first
tested in isotonic buffer (290 mOsmol/L) and then in
hypotonic buffer (210 mOsmol/L). Increased hydration in
the latter buffer did not alter the shear modulus of the
oxidized cells. For example, for one cell the shear modulus
was 19 x
dyne/cm in isotonic buffer and 20 x
at
210 mOsmol/L. Examination of four other oxidized RBCs
gave similar results. Likewise, swelling of oxidized RBCs did
not alter their plastic behavior, so that residual bumps were
seen on the membrane surface after RBC release from the
pipette following extensional deformation even at 210 mOsmol/L (data not shown). Thus, we found no evidence that
observed microrheologic properties of PMS-treated RBCs
could be explained by any dehydrating effect of oxidant
exposure.
Ektacytometric studies. Independent confirmation of
the apparent membrane-stiffening effect of PMS was obtained by ektacytometric examination of resealed RBC
ghosts. As summarized in Fig 4, membranes prepared from
oxidized RBCs required much higher levels of applied shear
stress to reach equivalent deformability index (DI), an effect
that was magnified a t progressively larger PMS concentrations. For example, membranes prepared from RBC exposed
to 500 bmol/L PMS required an 8.5-fold higher shear stress
A
4.0-
3.0-
2.0 -
1.o
1
I
B
0
4.0 -
I
I
I
I
I
3.0-
2.0-
1.o
J
1.o-I
I
0.00
0.01
0.02
DP x RPf2
I
0.03
Fig 2. Micropipette studies show the abnormal microrheology
of PMS-treated RBCs. This plot shows L/RP (aspiration length
normalized by pipette radius [RP]) as a function of membrane
tension (DP x RP/2 in dynes/cm where DP is the aspiration
show the loading phase of the
pressure). The open circles (0)
experiment, and the slope of the straight line fit shown here gives
the modulus of extensional rigidity ( p )for the cell membrane. The
closed circles (0) show the unloading phase of the aspiration
experiment. The data shown here (a representative experiment)
are for a control RBC (A), an RBC exposed to 10 pmol/L PMS with
abnormal hysteresis but normal p (B). and an RBC exposed to 50
pmol/L PMS with hysteresis and abnormal p (C).
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1018
HEEEEL. LEUNG. AND MOMANDAS
A
C
B
Fig 3. Rdduml bump &or nk.w of o r M W RECs from tho mkropipme lndiate lrwl.nk proportin. Th...vid.ankrogr.phs
were takon of a normal REC (A) and RBCs treated with PMS ot 26 L ~ / (B)
L or 100 r d / L IC). Thew por8lstent bumps on the RBC
membrane were soen eher releaw of every coll that hod been exposed to PMS at 10 pmol/L or grester concentrationr.
we also used ektacytometry to examine the influence of the
reducing agent dithiothreitol (DTT) on the membrane rheology of resealed ghosts. For these studies, ghosts were made
from PMS-treated RBCs. and the ghosts were subjected to
an additional incubation with or without 20 mmol/L DTT.
Rigidity of membrana from PMS-treated cells was improved significantly but only partially (an average 30%
decrease in relative rigidity) by this manipulation (Fig 5).
than normal membranes to reach equivalent deformation.
which implies an 8.5-fold increase in membrane dynamic
rigidity. Notably. exposure of normal RBCs to PMS in the
concentration range of 50 to I O 0 pmol/L thus induced the
same degree of membrane rigidity (1.6- to 3.3-fold increase)
previously observed for ghosts made from most dense sickle
RBCs (a 3.1-fold increase in rigidity compared with normal
membranes [unpublished data. March 1989)). It should be
noted that magnitude of increased membrane rigidity documented by ektacytometry cannot be directly extrapolated to
precise measurements of shear modulus (as documented by
micropipette) because behavior during ektacytometry is
dependent on additional mcmbrane properties.
Because we suspected that protein oxidation is the likely
cause of altered mcmbrane properties in PMS-treated RBCs.
DISCUSSION
These studies have documented the effect of PMS-induced
oxidation on RBC microrheologic properties. We chose PMS
for this work bccause it uses cytoplasmic hydrogen donors to
stimulate intracellular generation of superoxide.” thus enabling us to reproduce the directionality of pathologic
0.6
0.5
Fig 4. E l r a y c o m m y confhms tho membreno-stifhing
.fhct of PMS. The relationship
betwoen deformability index
(linear . a l e ) and shear stress
(log sale) provides a viswl representation of the membrano
rigidity shown by these studies
performed on resealed RBC
ghosts. Curves are shown for
ghosts derived from control
RBCs (01and for RBCs treated
withPMSat26lunol/L(r.).M)
#mol/L IC). 1 0 0 rma4lL (A).
250#md/L(.),andM)O#mol/L
(=I. Data are plotted for the
linear portlon of the curves: at
higher shear stress. the curves
flatten and converge.
x
0.4
3
.E
P
4 0.3
5
2
8
0.2
0.1
0.0
I
1
20
I
I
I
I
I
I
I
I
50
100
Shear stress (dyneslcmg
I
200
I
I
I
500
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1019
OXIDATION AND RED CELL MEMBRANE RHEOLOGY
No DTT
n20 mM DTT
PMS
1.0 mM
PMS
2.5 mM
PMS
Fig 5. DTT ameliorates the membrane-stiffening effect of
PMS. After exposure of RBCs to the indicated PMS concentrations, ghost membranes were examined in the ektacytometer, and
relative rigidity was assessed as for Fig 4. Subsequent rescue with
DTT partially corrected the membrane stiffness of PMS-treated
cells.
it is particularly notable that the micromolar P M S concentrations used here result in membrane changes that closely
mimic the abnormal microrheologic properties of dense,
dehydrated sickle
That these abnormalities of oxidized RBC are not ameliorated by hypotonic buffer argues
against an altered state of cell hydration as the cause of
observed changes in the membrane rheologic properties and
implies a direct role for structural alterations in the membrane.
It was previously suggested that the component of abnormal extensional rigidity and plastic-like behavior of the
sickle RBC membrane that is not reversed by cellular
rehydration might result from association of hemoglobin
with the membrane and/or from oxidative changes in membrane protein 0rganization.4.~The present findings enable us
to suggest that such effects might, indeed, account for the
irreversible component of microrheologic abnormalities observed during studies of sickle cells. We still cannot definitively distinguish between effects of heme on the membrane
and effects of oxidant per se; but, of course, in the context of
oxidative pathobiology the roles of membrane heme and
endogenous oxidant may be entwined i n e ~ t r i c a b l y . ' ~Never,'~
theless, it may be noted that we unequivocally influenced
microrheology using only 10 wmol/L PMS, at which concentration there was minimal, if any, increase in membrane
heme. Therefore, we favor the hypothesis that the observed
adverse microrheologic effect is due to protein oxidation
rather than heme per se. That DTT only partially ameliorates the PMS effect could simply reflect the participation of
oxidative damage to other, nonthiol amino acids in this
phenomenon. Because of the remarkably low effective concentration of PMS, its effect almost certainly is different from
that induced by high concentrations of exogenous peroxide
(formation of hemoglobin/spectrin adducts)." Thus, the
putative protein targets relevant to microrheologic alteration
remain to be defined, and it is hoped that future studies will
refine these observations further.
ACKNOWLEDGMENT
oxidant stress. Judging from the amount of thiol oxidation
induced at the PMS doses used here, it appears that we have
also reproduced a pathophysiologically relevant amount of
oxidant damage: albeit on a collapsed time scale. Therefore,
We thank Dr Evan Evans for allowing us to use his equipment and
for providing his expertise in analysis of the micromechanical
measurements. We gratefully acknowledge the technical assistance
of David Tukey, Wendy Foker, and Krishna Goyal.
REFERENCES
1. Clark MR, Mohandas N, Shohet SB: Deformability of oxygenated irreversibly sickled cells. J Clin Invest 65:189, 1980
2. Evans E, Mohandas N, Leung A: Static and dynamic rigidities
of normal and sickle erythrocytes: Major influence of cell hemoglobin concentration. J Clin Invest 73:477, 1984
3. Nash GB, Johnson CS, Meiselman HJ: Mechanical properties
of oxygenated red blood cells in sickle cell (HbSS) disease. Blood
63:73, 1984
4. Evans EA, Mohandas N: Membrane-associated sickle hemoglobin: A major determinant of sickle erythrocyte rigidity. Blood
70:1443, 1987
5. Rank BH, Carlsson J, Hebbel RP: Abnormal redox status of
membrane-protein thiols in sickle erythrocytes. J Clin Invest 75:
1531,1985
6. Kuross SA, Rank BH, Hebbel RP: Excess heme in sickle
erythrocyte inside-out membranes: Possible role in thiol oxidation.
Blood 71376, 1988
7. Rank BH, Moyer NL, Hebbel R P Vesiculation of sickle
erythrocytes during thermal stress. Blood 72: 1060, 1988
8. Hebbel RP, Eaton JW, Balasingam M, Steinberg MH: Spontaneous oxygen radical generation by sickle erythrocytes. J Clin Invest
70:1253, 1982
9. Hebbel RP, Morgan WT, Eaton JW, Hedlund BE: Accelerated autoxidation and heme loss due to instability of sickle hemoglobin. Proc Natl Acad Sci USA 85:237, 1988
10. Hebbel RP: The sickle erythrocyte in double jeopardy:
Autoxidation and iron decompartmentalization. Semin Hematol
2751. 1990
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1020
1 1. Maridonneau I, Braquet P, Garay R P Na' and K+ transport
damage induced by oxygen free radicals in human red cell membranes. J Biol Chem 258:3107, 1983
12. Gilbert HS, Stump DD, Roth EF Jr: A method to correct for
errors caused by generation of interfering compounds during erythrocyte lipid peroxidation. Anal Biochem 137:282,1984
13. Chasis JA, Mohandas N: Erythrocyte membrane deformability and stability: Two distinct membrane properties that are independently regulated by skeletal protein associations. J Cell Biol 103:343,
1986
14. Hebbel RP, Miller WJ: Phagocytosis of sickle erythrocytes:
HEBBEL, LEUNG, AND MOHANDAS
Immunologic and oxidative determinants of hemolytic anemia.
Blood 64:733,1984
15. Hebbel RP: Erythrocyte antioxidants and membrane vulnerability. J Lab Clin Med 107:401, 1986
16. Hebbel RP, Eaton JW: Pathobiology of heme interaction with
the erythrocyte membrane. Semin Hematol26: 136, 1989
17. Snyder LM, Fortier NL, Trainor J, Jacobs J, Leb L, Lubin B,
Chiu D, Shohet S, Mohandas N: Effect of hydrogen peroxide
exposure on normal human erythrocyte deformability, morphology,
surface characteristics, and spectrin-hemoglobin cross-linking. J
Clin Invest 76:1971, 1985
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1990 76: 1015-1020
Oxidation-induced changes in microrheologic properties of the red
blood cell membrane
RP Hebbel, A Leung and N Mohandas
Updated information and services can be found at:
http://www.bloodjournal.org/content/76/5/1015.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.