From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
High Doses of Intravenous Ig Inhibit In Vitro Uptake of C4 Fragments Onto
Sensitized Erythrocytes
By Milan Basta, Louis F. Fries, and Michael M. Frank
We have recently reported that intravenous lg (IVlg) inhibits
uptake of activated C3 fragments onto antibody-sensitized
red blood cells (RBCs). To elucidate the mechanism by which
lVlg exerts its effect on the complement system, we examined the possible interference with the C4 step of the
classical complement cascade. We examined the capacity of
autologous serum containing high concentrations of human
lVlg t o deposit C4 fragments onto model targets (guinea pig
and/or human erythrocytes sensitized with rabbit antiguinea pig/human erythrocyte IgG antibody). C4 binding was
quantified with radiolabeled anti-C4. Guinea pig serum with
added lVlg suppressed C4 uptake onto IgG-sensitized guinea
pig erythrocytes at all time points (0,5,15, and 30 minutes).
Using sera of guinea pigs treated with increasing doses of
IVlg, this effect was shown t o be dose-responsive. Serum
from a patient treated with lVlg showed reduced C4 uptake
onto sensitized homologous RBCs. In comparison with the
serum from the same patient before lVlg therapy was administered, levels were decreased almost t o background. C4
functional titers in those two samples were not different. C3
uptake was studied in parallel with C4 t o compare the degree
of inhibition using sera with increasing doses of lVlg in both
the human and guinea pig system. C3 and C4 inhibition
curves completely overlapped. Our findings suggest that lVlg
is an effective inhibitor of deposition of early complement
activation products (C4b. C3b) onto target surfaces and may
indicate interference of lVlg with multiple sites of complement activation.
This is a US government work. There are no restrictions on
its use.
I
C3; and C4 is also well known to bind readily to IgG. To
clarify the way IVIg exerts its effect on the complement
system, we examined the possible interference of IVIg with
the C4 step of the classical pathway of complement activation.
MMUNOGLOBULINS modified for intravenous use
(IVIg) have been introduced relatively recently.’ These
products are currently approved for use in the therapy of
idiopathic thrombocytopenic purpura (ITP) and primary
immunodeficiencies with associated hypogammaglobulinemia. However, it appears from numerous anecdotal reports
that IVIg has a substantial short- and long-term therapeutic
potential in a variety of autoimmune diseases, but the
underlying mechanisms are not understood. Several theories have been introduced, such as competition of infused
IVIg for Fc receptors with subsequent inhibition of IgG-Fcmediated cellular functions?+ Other theories include modulation of autoantibody production’ and direct effect on
lymphoid cells.“8
We have established another potential effect of IVIg
through interference with the complement c a ~ c a d e .Two
~~’~
in vivo models of complement-dependent immune injury
were used: clearance of IgM-sensitized guinea pig erythrocytes from the circulation of guinea pigsg and Forssman
shock,” a lethal response in guinea pigs injected with rabbit
anti-Forssman antibody. In both systems, pretreatment of
animals with IVIg significantly modulated immune damage
mediated by activated complement fragments. In addition,
in vitro studies were performed examining the uptake of C3
on antibody-sensitized guinea pig erythrocytes. Data from
these experiments suggest that supraphysiologic levels of
IVIg act in part by preventing C3 fragments from binding to
targets cells. Fragments of activated C4 bind covalently to
target surfaces by chemical reactions identical to those of
From the Laboratory of Clinical Investigations, National Institutes
of Allergy and Infectious Diseases, National Institutes of Health,
Bethesda, MD.
Submitted June 22, 1990; accepted September 19, 1990.
Address reprint requests to Milan Basta, MD, NIH, Bldg 10, Room
llN228, 9000 Rockville Pike, Bethesda, MD 20892.
The publication costs of this article were defrayed in pari 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.
This is a US govemment work There are no restrictions on its use.
0006-4971/91/7702-OO12$0.0OlO
376
MATERIALS AND METHODS
Animals. Female guinea pigs weighing 300 to 400 g (National
Institutes of Health “multipurpose”-non-inbred) were obtained
from the National Cancer institute, Frederick, MD.
Antibodies. Rabbit IgM anti-guinea pig erythrocyte antibody
was fractioned from the serum of a rabbit immunized IV with
guinea pig red blood cells (RBCs). A 42.7% ammonium sulfate
precipitate from the immune serum was redissolved in phosphatebuffered isotonic saline (PBS, pH 7.4) and chromatographed on a
Sephadex G-200 (Pharmacia Fine Chemicals, Freehold, NJ) column. IgM-containing fractions were pooled and then passed over a
concanavalin A (con A) agarose affinity column. IgM contained in
the 0.5 mol/L glucose eluate from the con-A agarose affinity
column was concentrated and further subjected to ultracentrifugation on a 10% to 40% sucrose gradient. Fractions containing IgM
were pooled and stored directly in the sucrose solution at 4°C.
Rabbit polyclonal IgG anti-guinea pig erythrocyte antibody,
anti-guinea pig C3, and anti-guinea pig C4 were purified from the
sera of rabbits immunized with guinea pig erythrocytes, guinea pig
C3, and guinea pig C4, respectively. IgG fractionation involved
sequential ammonium sulfate precipitation, Sephacel G-200 sieving chromatography, and diethyloamino ethanol (DEAE) anionexchange chromatography. Rabbit antihuman RBC IgG was purchased from Cappel (West Chester, PA). Antihuman C4 IgG
polyclonal antibody was purified from a burro immunized with
purified human C4. Antihuman C3 IgG polyclonal antibody was
obtained by immunizing a goat with purified human C3. IgG
fractionation was performed by octanoic acid precipitation.”
Radiolabeling. Radioiodination of rabbit anti-guinea pig C3,
rabbit anti-guinea pig C4, burro antihuman C4, and goat antihuman C3 was accomplished with iodobeads following the directions
supplied by the manufacturer. Respective polyclonal IgG, 300 fig,
was incubated with 1mCi of lZsIand three iodobeads for 30 minutes
at room temperature. Free iodine was separated from proteinbound isotope by passing the labeled material through a prepacked
PDlO (Sephadex G-25M) column (Pharmacia Fine Chemicals,
Freehold, NJ) in isotonic PBS, pH 7.4.
Complement components. Guinea pig fourth component of
complement was purchased from Cordis Laboratories (Miami,
Blood, Vol77, No 2 (January 151,1991: pp 376-380
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
IVlG AND C4 UPTAKE
377
FL). Guinea pig C3 was purified to homogeneity by a modification
of the procedure of Hammer et all2 designed for human C3.
Human C3 and human C4 were also purified by the method of
Hammer et a1.”
Complement titrations. Hemolytic complement titrations of human C4 were performed using standard published procedures.”
Other reagents. Human serum Ig for IV use in 10%maltose, pH
4.25 (Gammimune), and human serum albumin 25% solution for
IV administration were purchased from Cutter Biologicals (Berkeley, CA). Maltose hydrate grade I was obtained from Sigma
Chemical Co (St Louis, MO). Sodium
iodide was purchased
from Amersham Corp (Arlington Heights, IL). Iodobeads were
purchased from Pierce Chemical Co (Rockford, IL).
Buffers. Isotonic Veronal-buffered saline containing 0.1% gelatin, 0.15 mmol/L Ca2+,and 1 mmol/L Mgz+ (GVBS”), isotonic
Veronal-buffered saline containing 0.1% gelatin, and 10 mmol/L
EDTA buffer were prepared as previously de~cribed.’~
In vitro C4 uptake studies. To obtain blood, a guinea pig was
anesthetized by a subcutaneous injection of a cocktail containing
0.1 mL of Rompun (Xylazine; Mobay Corp, Shawnee, KS) and 0.5
mL of Vetalar (Ketamine HCI; Parke-Davis Co, Morris Plains,
NJ). Cardiac puncture was then performed. RBCs were separated
from EDTA plasma, washed, and standardized. Three tenths of 1
mL of guinea pig RBCs suspension at 2.7 x 10’ cells/mL was
incubated with 0.3 mL of diluted rabbit IgG or IgM anti-guinea pig
erythrocytes antibody, or 0.3 mL of buffer for 30 minutes at 37°C.
After two washes in GVBS2+,tubes were placed on ice. Pellets of
aliquoted EA were resuspended in 500 pL of normal guinea pig
serum (NGPS), NGPS substituted with IVIg (10 mgltube), and
NGPS substituted with human albumin in 10% maltose (10
mgltube). E pellet was resuspended in NGPS. Immediately after
the pellets have been resuspended, duplicate 60-pL samples were
taken from each tube and transferred to corresponding tubes with
2 mL ice-cold EDTA buffer (“0”
time points). The tubes were
then placed in a water bath at 37°C. At 5-, 15-, and 30-minute time
points, duplicate 60-pL samples were removed from each tube and
transferred to EDTA wash tubes. The intact erythrocytes in these
samples were pelleted and the extent of lysis was determined by
measuring free hemoglobin in the corresponding supernatants
spectrophotometrically. The pellets were then washed and resuspended in 200 pL of GVBS2+.Each tube received the amount of
radiolabeled anti-C3 or antLC4 equal to 2 x lo6cpm and incubated
on melting ice for 60 minutes with occasional shaking. After two
further washes in ice-cold EDTA buffer, the pellets were counted
in a y counter. Where appropriate, counts were corrected for
calculated percentage of lysis.
20000
01
0
C4 uptake onto ZgG-sensitized erythrocytes. Guinea pig
erythrocytes sensitized with rabbit anti-guinea pig erythrocyte antibody were incubated with NGPS diluted 1:4,
guinea pig serum substituted with 10 mg human IVIg, and
guinea pig serum with added 10 mg of human albumin for
IV use in 10% maltose, pH 4.25. At various time points,
aliquots were removed to quantitate C4 binding by uptake
of lZrIanti-guinea pig C4 onto the sensitized erythrocytes.
At all time points IVIg suppressed C4 uptake by 50% (Fig
1). Some C4 was deposited on antibody-sensitized cells
incubated briefly in NGPS at O”C, reflecting the known
ability of guinea pig complement to allow uptake of C4 onto
sensitized erythrocytes at 0°C. NGPS with added albumin in
maltose had no effect on C4 uptake in comparison with the
NGPS.
1
I
20
30
40
T i m e (min)
Fig 1. The effect of lVlg on C4 uptake by IgG-sensitized guinea pig
erythrocytes. Guinea pig erythrocytes were sensitized with rabbit
anti-guinea pig erythrocyte IgG antibodies and then incubated with
NGPS diluted 1:4 l-U-1,
NGPS substituted with 10 mg/tube of
human lVlg (-O-),
and NGPS with added human albumin for IV use
in 10% maltose (-0-).
Aliquots were removed et 0-, 5-, 15.. and
30-minute time points for quantitation of C4 binding by anti-guinea
pig C4 IgG antibody labeled with ‘-1. Inhibition of 50% at all time
pointswas observed.
C4 uptake onto ZgM-sensitized guinea pig erythrocytes.
IgM-sensitized erythrocytes were used as targets and C4
uptake kinetics were quantified with anti-guinea pig C4
labeled with ’% In the presence of NGPS diluted 1:4, C4
uptake reaches the maximum in 5 minutes and sharply
decreases at the 15-minute time point. Over the next 15
minutes, C4 on the surface of targets shows tendency for a
further decay. IVIg serum suppressed the C4 binding by
20% at all time points, but did not alter the shape of the
curve (Fig 2). The effect was specific for IVIg, because
albumin/maltose had no effect on C4 uptake relative to
NGPS.
4000
I
1
3000
E
RESULTS
I
IO
Q
2000
0
1000
O!
0
I
I
I
I
10
20
30
40
Time (min)
Fig 2. The effect of lVlg on C4 uptake by IgM-sensitizedguinea pig
erythrocytes. Guinea pig erythrocytes were sensitized with antiguinea pig IgM antibodies and then incubated with NGPS diluted 1:4
(-0-1,
NGPS containing lVlg at 10 mg/tube I-U-),
and NGPS
Aliquots were removed
with human albumin in 10% maltose (-O-).
at 0,5, 15, and 30 minutes and C4 binding quantified by uptake of ”I
anti-guinea pig C4. At all time points lVlg inhibited C4 uptake onto
sensitized erythrocytesby approximately20%.
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
370
BASTA, FRIES, AND FRANK
The effect of increasing doses of Mg on C4 uptake onto IgG
targets. Three guinea pigs were injected with increasing
doses of IVIg (600 mgkg, 1,200 gkg, and 1,800 mgkg).
Three hours after the injection of IVIg, a sample of blood
from each animal was obtained by cardiac puncture. Guinea
pig erythrocytes were then sensitized with rabbit antiguinea pig erythrocyte IgG antibody. Pellets of aliquoted
EA were resuspended in undiluted NGPS and undiluted
sera of animals treated with increasing doses of IVIg. After
a 5-minute incubation at 37"C, aliquots of cells were
removed for C4 uptake quantitation by radiolabeled antiguinea pig C4 antibody. Stepwise increase of C4 uptake
inhibition (50%, 76%, and 90%) was observed, proportional to the increase of IVIg dose in guinea pig serum,
meaning that C4 uptake is inhibited in a dose-responsive
manner (Fig 3).
C4 uptake inhibition by serum from an Mg-treated patient.
Human RBCs were sensitized with rabbit antihuman RBC
IgG antibody and incubated with the undiluted serum from
a patient treated with IVIg at 600 mgkg for 3 consecutive
days and the same patient's serum sample before IVIg
therapy. After 5-minute incubation at 37"C, aliquots of cells
were taken for C4 uptake quantitation by antihuman C4
IgG antibody labeled with lUI. In comparison with the
sample before IVIg therapy, posttreatment serum decreased C4 uptake onto sensitized homologouserythrocytes
to almost background (Fig 4). C4 functional titer in those
two samples were not different, meaning that therapy itself
had no effect on complement levels in serum.
Parallel C3lC4 uptake studies. C3 uptake was studied in
parallel with C4 to compare the degree of inhibition using
sera with added increasing doses of IVIg in both the human
and guinea pig system. Guinea pig erythrocytes were
sensitized with rabbit anti-guinea pig RBC IgG antibody.
Pellets of aliquoted antibody-sensitized erythrocytes were
resuspended in NGPS and guinea pig serum substituted
2000
I
T
I
20000
:
1
10000
V
0
Fig 4. The effect of serum from IVlg-treated patient on C4 binding
to sensitized human erythrocytes. Serum samples were collected
from a patient before and after 3day therapy with iVlg at 600 mglkg.
Human erythrocytesof " 0 + type (same as patient's) were sensitized
with the serum before (D) and after (UI) lVlg therapy; nonsensitized
erythrocytes(n)were incubated with autologous human serum. After
a 5-minute incubation at 37°C. aiiquots of cells were removed for C4
uptake quantification by burro antihuman "i labeled C4. C4 uptake
was almost completely blocked in the presenceof post-lVlg serum.
with increasing doses of IVIg (5 mg, 10 mg, 20 mg, and 25
mgltube). After a 5-minute incubation at 37"C, aliquots of
RBCs were taken for C3/C4 uptake quantitation using
rabbit anti-guinea pig C3/C4 antibodies labeled with '%I. It
was found that with increasing doses of IVIg, both C3 and
C4 uptake inhibition increased to the same extent, so that
their curves completely overlapped. Similar experiments
were performed in the human system; human erythrocytes
were sensitized with rabbit antihuman RBC IgG antibody.
Pellets of sensitized erythrocytes were resuspended in
autologous serum and the same serum substituted with
increasing doses of IVIg (5 mg, 10 mg, 20 mg, and 25
mg/tube). After 5-minute incubation at 37"C, aliquots of
RBCs were taken for C3/C4 uptake quantification with
respective radiolabeled ('=I) rabbit antibody. As in the
guinea pig system, percent of uptake inhibition increased
proportionally to the increase of added IVIg; uptake
inhibition was the same both for C3 and C4 at all dose
points, so that the inhibition curves were superimposable
(Fig 5).
DISCUSSION
E
1000
0
Fig 3. The effect of increasing doses of lvlg on the uptake of C4
onto IgG-sensitized guinea pig erythrocytes. Three guinea pigs were
given 600 mg/kg (a), 1200 mg/kg (C3)and 1.800 mglkg (a)of IVlg,
respectively. Serum from each animal was incubated with IgGsensitized guinea pig erythrocytes (GPEA); controls were GPEA
incubated wlth NGPS (D) and unsensitized erythrocytesin NGPS (0).
After a 5-minute incubation at 37°C. aliquots were removed and C4
uptake quantified by radiolabeledantiguinea pig c4.Percent inhibition (relative to positive control) increased with increasing doses of
wig.
The propensity of IgG to serve as a covalent acceptor for
nascent C3 fragments is well known. C3b produced during
complement activation by soluble immune complexes or
bacteria binds to IgG immune complexes or "innocent
bystander" IgG molecules.14 Studies by Berger et all5
showed that IVIg-inhibited deposition of C3b onto antibodysensitized erythrocytes in a system using serum diluted 40to 50-fold, and the uptake of trypsin-generated C3 fragments onto particles has been efficiently blocked by fluidphase IgG.I6
C3b fragments coating antibody-sensitized targets are
important for in vivo clearance of immune complexes.
Using clearance of IgM-sensitized erythrocytes as an in vivo
model free of IgG-Fc receptor interactions, we explored a
potential in vivo action of IVIg through interference with
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
IVIG AND C4 UPTAKE
Qe
379
20
0
0
10
20
30
I VIG(mg/ml)
Fig 5. Parallel C3/C4 uptake studies in the human serum system.
Human RBCs were sensitized with rabbit antihuman erythrocyte IgG
antibody. Sensitized RBCs were incubated for 5 minutes at 37°C with
undiluted autologous human serum and autologous human serum to
which different doses of lVlg were added (5 mg, 10 mg, 20 mg, and 25
me). C4 uptake was calculated as percent inhibition relative to the
maximum uptake observed in autologous human serum without IVlg.
Inhibition curves for C3 (4-)
and C4 (e)
were identical in
shape and magnitude.
complement fragment deposition onto targets.' We found
that the clearance of IgM-sensitized guinea pig erythrocytes
was significantly retarded in guinea pigs treated with
human IVIg preparation. In vitro C3 uptake studies suggested that IVIg produces a kinetic depression of C3 uptake
onto IgM-sensitized targets. Next, we used Forssman shock,
another model of complement-mediated tissue damage
that is IgG dependent but free of IgG-Fc interactions. By
treating the animals with high doses of IVIg we were able to
prevent death in 40% of animals and prolong survival
fivefold in those animals that eventually succumbed to an
otherwise lethal dose of anti-Forssman antibody." No
control animal, treated with human/guinea pig albumin and
maltose vehicle survived Forssman shock. In vitro C3
uptake onto IgG-sensitized erythrocytes was almost completely inhibited in an IVIg-treated animal's serum. We
suggested that one of the mechanisms of action of IVIg is
interference with complement fragment deposition onto
antibody-coated targets.
In previous studies we studied blockade of C3 fragment
uptake on erythrocytes in detail, and examined binding of
C3 to erythrocytes presensitized with antibody and C1 and
C4 of the classical complement pathway. Specifically, preloading of targets with C1 and C1 + C4 did not affect the
capacity of IVIg to block C3 uptake onto these particular
targets. We concluded that IVIg exerts the major effect at
C3 convertase step. The approach we used to address the
issue was indirect, using limited amounts of C1 and C4
compared with what is available in plasma. Therefore, we
felt that the effect of IVIg at early steps of complement
activation by classical pathway requires careful re-examination. C4 was chosen for study because of many similarities
between C3 and C4. Both are activated by cleavage of a
small peptide from the a chain exposing a buried reactive
internal thiolester bond. Both are capable of binding
covalently to targets and both have a propensity to bind
IgG. Similar to o u r previous C3 uptake studies, we found
that IVIg had much more pronounced effect on C4 uptake
by IgG-sensitized erythrocytes than IgM-coated targets.
For that reason, in all subsequent experiments we used
IgG-sensitized erythrocytes to evaluate C4 uptake. We
showed that this process is inhibited by IVIg in a doseresponsive manner. As can be concluded from parallel
C3/C4 uptake studies in both human and guinea pig
systems, IVIg inhibits C3 and C4 uptake to the same
degree. These findings imply that IVIg interferes with the
classical complement cascade at multiple sites. It would be
of interest to determine which of these steps is of greater
importance in control of complement activation by this
important therapeutic modality.
REFERE NCES
1. Imbach P, Barandun S, d'Apuzzo V, Rossi E, Vest M,
Baumgartner C, Morel A, Wagner HP: High-dose intravenous
gammaglobulin for idiopathic thrombocytopenic purpura in childhood. Lancet 1:1228,1981
2. Kurlander RJ, Ellison DM, Hall J: The blockade of Fc
receptor mediated clearance of immune complexes in-vivo by
monoclonal antibody (2.462) directed against Fc receptors on
murine leukocytes. J Immunol133:855,1984
3. Fehr J, Hofmann V, Kappeler U: Transient reversal of
thrombocytopenia in idiopathic thrombocytopenic purpura by high
dose intravenous gamma globulin. N Engl J Med 306:1254,1982
4. Kimberly RP, Salomon JE, Bussel JB, KuntzCrow M, Higartner Mw: Modulation of mononuclear phagocyte function by
intravenous gamma-globulin. J Immunol132:745,1984
5. Rossi F, Sultan Y, Kazatchkine MD: Spontaneous and
therapeutic suppression of autoimmune response to factor VI11 by
anti-idiotypic antibodies, in Morel1 AA, Nydeger UE (eds): Clinical Use of Intravenous Immunoglobulin. San Diego, CA, Academic, 1986, p 421
6. Durandy AA, Fleisher A, Griscelli C Dysfunction of pokeweed
mitogen-stimulated T and B lymphocyte responses induced by
gammaglobulin therapy. J Clin Invest 67:867,1981
7. Dammacco F, Iodice G, Campobasso N: Treatment of adult
patients with idiopathic thrombocytopenic purpura with intravenous immunoglobulin: Effects on circulating T-subsets and PWMinduced antibody synthesis in vitro. Br J Haematol62125,1986
8. h u n g DY, Burns JC, Newberg JW,Geha RS: Reversal of
lymphocyte activation in vivo in the Kawasaki syndrome by intravenous gammaglobulin. J Clin Invest 79:468,1987
9. Basta M, Langlois PF, Marques M,Frank MM, Fries L F
High-dose intravenous immunoglobulin modifies complementmediated in vivo clearance. Blood 74:326,1989
10. Basta M, Kirshbom P, Frank MM, Fries LF: Mechanism of
therapeutic effect of high-dose intravenous immunoglobulin.Attenuation of acute, complement-dependent immune damage in a
guinea pig model. J Clin Invest 84:1974,1989
11. Steinbuch A, Audrin R: The isolation of functionally active
components of the human complement system. Arch Biochem
Biophys 134229,1969
12. Hammer CH, Wirtz GH, Renfer L, Gresham HD, Tack BFJ:
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
380
A large scale isolation of functionally active components of the
human complement system. J Biol Chem 265:3995,1981
13. Gaither TA, Frank MM: Complement, in Henry JB (ed):
Clinical Diagnosis and Management by Laboratory Methods.
Philadelphia, PA, Saunders, 1979, p 1253
14. Jacobs RJ,Reichlin M: Generation of low M.W. C3-bearing
immunoglobulin in human serum. J Immunol1302775,1983
BASTA, FRIES, AND FRANK
15. Berger M, Rosenkrantz P, Brown CY: Intravenous and
standard immune serum globulin preparations interfere with
uptake of '"I-C3 onto sensitized erythrocytes and inhibit hemolytic
complement activity. Clin Immunol Immunopathol34227,1985
16. Capel PJA, Groeneboer 0, Grosveld G, Pondman KWJ:
The binding of activated C3 to polysaccharides and immunoglobulins. J Immunol121:2566,1978
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
1991 77: 376-380
High doses of intravenous Ig inhibit in vitro uptake of C4 fragments
onto sensitized erythrocytes
M Basta, LF Fries and MM Frank
Updated information and services can be found at:
http://www.bloodjournal.org/content/77/2/376.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.