Regular Article

RED CELLS, IRON, AND ERYTHROPOIESIS

A hyperactive Mpl-based cell growth switch drivesmacrophage-associated erythropoiesis through anerythroid-megakaryocytic precursorEyayu Belay,1-3 Chris P. Miller,2,4 Amanda N. Kortum,2,4 Beverly Torok-Storb,3 C. Anthony Blau,2,4 and David W. Emery1,2

1Department of Medicine, Division of Medical Genetics and 2Institute for Stem Cell and Regenerative Medicine, University of Washington, Seattle, WA;3Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA; and 4Department of Medicine, Division of Hematology, University of

Washington, Seattle, WA

Key Points

• Increasing receptor stability ofan Mpl-based cell growthswitch improves ex vivoexpansion from cord bloodCD341 cells.

• Expansion includes Epo-independent, macrophage-associated erythropoiesisfrom a novel erythroid-megakaryocytic precursorpopulation.

Several approaches for controlling hematopoietic stem and progenitor cell expansion,

lineage commitment, and maturation have been investigated for improving clinical inter-

ventions. We report here that amino acid substitutions in a thrombopoietin receptor (Mpl)–

containing cell growth switch (CGS) extending receptor stability improve the expansion

capacityofhumancordbloodCD341cells in theabsenceofexogenouscytokines.Activation

of this CGS with a chemical inducer of dimerization (CID) expands total cells 99-fold, eryth-

rocytes 70-fold, megakaryocytes 0.5-fold, and CD341 stem/progenitor cells 4.4-fold by 21

days of culture. Analysis of cells in these expanded populations identified a CID-dependent

bipotent erythrocyte-megakaryocyte precursor (PEM) population, and a CID-independent

macrophage population. The CD235a1/CD41a1 PEM population constitutes up to 13% of

the expansion cultures, can differentiate into erythrocytes or megakaryocytes, exhibits very

littleexpansioncapacity, andexistsat very low levels inunexpandedcordblood.TheCD2061

macrophage population constitutes up to 15% of the expansion cultures, exhibits high-

expansion capacity, and is physically associated with differentiating erythroblasts. Taken

together, these studies describe a fundamental enhancement of the CGS expansion platform,

identify a novel precursor population in the erythroid/megakaryocytic differentiation pathway of humans, and implicate an erythropoietin-

independent,macrophage-associated pathway supporting terminal erythropoiesis in this expansion system. (Blood. 2015;125(6):1025-1033)

Introduction

The ability to control the expansion, lineage commitment, and matu-rationofhematopoietic stemandprogenitor cells (HSPCs), ina specificand regulated fashion, would provide a powerful tool for clinical inter-vention. The initial promise of recombinant cytokines for this pur-pose has been limited by their association with deleterious off-targeteffects.1-3 Currently, recombinant cytokines have proven useful formobilizing HSPCs for collection by apheresis,4 treating anemia asso-ciated with chronic kidney disease and chemotherapy,5 and treatingcancer-associated neutropenia.6Cytokines supportHSPCcell survivaland proliferation in vitro during transduction with recombinant viralvectors,7 and support limited ex vivo expansion for improving out-comes in cord blood transplantation.8 Genetic engineering strategiesbased on drug resistance,9 or enhanced HSPC self-renewal,10 providea means of controlling the expansion of specific cell populations, butrequire the use of cytotoxicdrugs for selectionor geneswithoncogenicpotential, raising both off-target and safety concerns.

Wehave been investigating an alternative approach for regulatinghematopoietic cell expansionanddifferentiationbasedon theobservation

that signaling by many cytokine receptors is triggered by binding of2 receptor molecules by a single cytokine molecule. By fusing theintracellular signaling domain of these receptors to an artificial dimer-izationdomain, it is possible to bring receptor binding, and thus receptorsignaling, under control of a small-drug molecule called a chemical in-ducer of dimerization (CID).3Artificial cell growth switch (CGS) recep-tors of this type encoding the signaling domain of the thrombopoietin(TPO) receptor (Mpl) have proven especially useful for the regulatedexpansion of selected hematopoietic lineages in multiple settings.11-23

The 635-aa nativeMpl protein, also knownasCD110 andTPO-receptor,is a major regulator of megakaryocyte and platelet formation andhas also been implicated in HSPC maintenance.24-26 Ex vivo cultureand in vivo transplantation studies with constitutively active viralvectors expressing the artificially dimerizable version of thisMpl-basedCGS receptor in HSPCs from mice,13-15 dogs,16,17 and humans18-23

demonstrated an unexpected and disproportionate effect of CID-mediated expansion on primitive erythroid cells, and to a lesser extent Tand B lymphocytes, as well as megakaryocytes and platelets. In every

Submitted February 7, 2014; accepted October 14, 2014. Prepublished online

as Blood First Edition paper, October 24, 2014; DOI 10.1182/blood-2014-02-

555318.

The online version of this article contains a data supplement.

There is an Inside Blood Commentary on this article in this issue.

The publication costs of this article were defrayed in part by page charge

payment. Therefore, and solely to indicate this fact, this article is hereby

marked “advertisement” in accordance with 18 USC section 1734.

© 2015 by The American Society of Hematology

BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6 1025

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

instance, expansion was limited to cells transduced with the viral vector,and was reversible upon withdraw of the CID. Studies with high vectordoses and highly purifiedHSPCpopulations provided evidence thatthis CGS system was capable of expanding HSPCs from human cordblood.21,22 However, most studies with cord blood CD341 cells inculture, and all transplantation studies in mice and dogs, showed noevidence for CGS-mediated expansion of primitive HSPCs. Further-more, efforts to use this system for cell expansion from adult sourcesof human HSPCs have also met with limited success.19

Althoughphysiological levelsofTpo/Mpl signaling result inHSPCquiesence,25,26 superphysiological doses of Tpo induce HSPC repli-cation in mice.26 Based on this observation, we hypothesized thatthe inability of this CGS system to expand primitive HSPC in mostsettings, and especially from adult human HSPCs, was the result ofinadequate levels of CGS receptor signaling. To test this hypothesis,we substituted sequences in theMpl signaling domain of the CGS re-ceptor known to be involved in degradation of the human Mpl recep-tor, and assessed the expansion potential of the resulting constructsin human HSPC cultures. We describe here the capacity of one of theseconstructs to significantly improve the ex vivo expansion of both ma-ture and immature hematopoietic populations from cord bloodCD341

cells. These studies also revealed a CD235a1/CD41a1 precursorpopulation capable of differentiating into both erythrocytes andmega-karyocytes similar to a population reported to arise during stress he-matopoiesis inmice.27,28 This bipotent precursor populationwas foundto undergo erythropoietin (Epo)–independent terminal erythropoiesisin contact with macrophages.

Methods

Lentiviral vectors

The self-inactivating CGS lentiviral vector LMFMPG was reported previ-ously.23 The CGS receptor cassette is transcribed from the constitutively activemurine stemcell virus (MSCV) promoter and is composed of a hybrid sequenceencoding the modified binding domain FKBP(F36V) linked to complementaryDNA (cDNA) encoding the intracellular domain of mouse Mpl. This hybridprotein also contains a myristoylation domain to target the inner cell membraneand a hemagglutinin (HA) tag. The green fluorescent protein (GFP) cassette istranscribed from the constitutively activePGK gene promoter. The hyperactivevariants were created by introducing the following amino acid substitutions(coordinates based on the mouse Mpl sequence): K544F, K564F, and Y582F.Lentiviral vector lots were generated by cotransfection of 293T cells andconcentrated 100-fold before use as described.29 Vector titers were determinedby flow cytometric analysis of GFP expression in transduced HT1080 cells.

Ba/F3 transduction and expansion cultures

Murine Ba/F3 cells were vector-transduced and expanded in the absenceof CID in RPMI 1640 media supplemented with 10% fetal bovine serum(RPMI/FBS) and 5%WEHI-conditionedmedium as a source of interleukin-3(IL-3) at 37°C, 5% CO2. After confirming similar rates of gene transfer usingflow cytometry for GFP (wild type [wt], 40%; K*K* (K544F and K564F),44%; Y* (Y582F), 39%; and K*K*Y*, 43%), cells were washed to removeIL-3 and then cultured in RPMI/FBS supplemented with the indicated con-centrations of the CID drug AP1903, a derivative of AP20187 developedfor clinical use (a gift from ARIAD Pharmaceuticals Inc). After 5 days,cell proliferation was assessed by the MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) assay as previ-ously described.11

Western blot analysis

Ba/F3 cell lines transduced with wt and Y* (Y582F) CGS vectors were ex-pandedwith 100 nMAP20187 until all cells expressed vector GFP. Equal num-bers of cells were then plated in RPMI/FBS supplemented with cycloheximideat 0, 1, and 5 mg/mL. After 4 and 8 hours, cell lysates were prepared in thepresenceof protease andphosphatase inhibitors, resolvedon4%to12%Bis-Trisgels, transferred to polyvinylidene difluoride membranes (NuPage system; LifeTechnologies), and probedwith a primary antibody to the CGS receptor HA tagfollowed by an anti-mouse horseradish peroxidase–conjugated secondary anti-body. Blots were developed and exposed using enhanced chemiluminescenceand Hyperfilm (GE Healthcare). Band densities were quantified using Image Jsoftware (http://rsb.info.nih.gov/ij/). Analysis was only performed on this CGSvariant because it provided the best expansion in primary CD341 cell cultures.

Primary CD341 purification, transduction, and

expansion cultures

Anonymized human cord blood was obtained from the Puget Sound BloodCenter Cord Blood Program following donor consent. Mononuclear cells wereenriched for CD341 cells by immunomagnetic separation (Miltenyi Biotec Inc)to an average 84% 6 5% purity. Fresh CD341 cells were prestimulated inIscove modified Dulbecco medium supplemented with 10% FBS (IMDM/FBS), 50 ng/mL recombinant human (rh) stem cell factor (rhSCF), 20 ng/mLrhIL-6, 20 ng/mL rhTPO, and 100 ng/mL rhFlt3-L at 37°C, 5% CO2. Allcytokines were purchased from PeproTech. After 3 hours, the cultures werefurther supplemented with 5 mg/mL polybrene and CGS vectors at amultiplicity of infection of 10, and incubated overnight. The cells were thenwashed and replated in IMDM/FBS and 100 nMAP20187without additionalcytokines at a dose of;53 105 cells permL per well. At days 3, 7, 11, 14, 21,and 26 cells were collected, washed, and either saved for analysis or replatedin freshmedium at a target concentration of 53 105 cells per mL.Anonymizedgranulocyte–colony-stimulating factor (G-CSF) mobilized peripheral bloodCD341 cells were obtained from the Fred Hutchinson Cancer ResearchCenter Hematopoietic Cell Processing Core following donor consent, andwere transduced and cultured in the same fashion as the cord blood CD341

cells, except that they were prestimulated for 24 hours to allow for recoveryfrom cryopreservation.

Cell analysis

Cells were stained with fluorochrome-conjugated antibodies (CD34–phycoerythrin [PE] Cy7a, CD38-PE, CD41a-PE, CD42b–allophycocyanin[APC], CD45RA-APC, CD90-APC, CD123-PECy7a, and CD235a-APC;BD Biosciences), and analyzed for antibody staining and vector GFPexpression on a FACSCanto 6-color flow cytometer (BD Biosciences) usingFloJo software. Data were collected from a target cell count $30 000, deadcells were excluded by 7-aminoactinomycin D (7-AAD) staining, nucleatedcells were gated based on light scatter, and gates were determined by un-transduced, unstained, or isotype controls. Sorting was performed on aFACSAria cell sorter (BD Biosciences). Cytospins were prepared by sus-pending cells in 200 mL of phosphate-buffered saline (PBS)/2% FBS, andcentrifuging onto slides at 72 g for 5 minutes. Air-dried slides were Wright-Giemsa stained at the Core Center of Excellence in Hematology cell char-acterization resource at the Fred Hutchinson Cancer Research Center andimaged at3200 magnification.

Secondary cultures

Colony-forming progenitor assays were performed using MethoCult H4034Optimum (StemCell Technologies). Plates were incubated at 37°C, 5% CO2,and scored for colony formation and classification for up to 18 days. To assesserythroid differentiation potential, sorted cells were cultured in IMDM/FBSsupplemented with 2 IU/mL rhEpo 6 50 ng/mL rhSCF. To assess mega-karyocyte differentiation potential, sorted cells were cultured in IMDM sup-plemented with 20% bovine serum albumin, insulin and transferrin (BIT)serum substitute (StemCell Technologies), 30 ng/mL rhTPO, 50 ng/mLrhIL-6, 13.5 ng/mL rhIL-9, 6 1 ng/mL rhSCF as described.30

1026 BELAY et al BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

Results

Derivation and validation of hyperactive variants of the

Mpl-based CGS

Toward the goal of improving the signaling activity of ourMpl-basedCGS (Figure 1A), we introduced single amino acid substitutions at3 specific sites in the Mpl intracellular domain shown by others tobe involved in degradation of the homologous human Mpl receptor(Figure 1B). This included 2 sites involved in protein ubiquitination(K544F and K564F, K*K*),31 and 1 site bound by the accessoryprotein AP2 leading to receptor endocytosis (Y582F, Y*).32 Theresulting variants were introduced into a lentiviral vector coexpress-ing a GFP reporter gene (Figure 1C),22 and assessed for CGS activityin factor-dependent Ba/F3 cells. As seen in Figure 1D, all 3 variantsreduced the threshold for CID responsiveness (from50 nM to 25 nM)and increased the proliferative magnitude of CID responsiveness(from threefold to nearly sixfold at 100 nM CID) compared with thevector containing the wt Mpl signaling domain. Western blottingconfirmed that the AP2 binding site–deficient variant increasedthe half-life of the CGS receptor protein by 34%, from 4.3 to 5.8hours (Figure 1E).

Hyperactive CGS variants improve expansion from human

CD341 cells

Culture studies with human cord blood CD341 cells also demon-strated a pronounced improvement in the proliferative magnitude ofCID responsiveness for all 3 of the hyperactive CGS variants(Figure 2A). In particular, the Y*CGS variant increased the total celloutput, from 33-fold to 99-fold. Based on an average initial trans-duction rate of ;25%, this translates to a degree of expansion forcells transduced with the Y* CGS vector of nearly 400-fold.Tracking of the percentage of cells expressing vector GFP furtherconfirmed that this increase in total cell output was due to the accel-erated expansion of vector-transduced cells, peaking at roughly 90%of all cells (Figure 2B). Cytospins confirmed the expansion productsgenerated by both the wt and Y* CGS vectors were dominated bycells of the erythroid lineage characterized by dark blue cytoplasm,dark nuclei, and prominent Golgi, representing different stages ofmaturation, as well as macrophages characterized by blue nuclei,pale foamy cytoplasm, and a small nucleus to cytoplasm ratio(Figure 2C). Titration studies demonstrated that increasing the doseof the wt CGS vector increased the amount of expansion, but thiseffect plateaued at a level that was still statistically less than theamount of expansion achieved with a standard dose of the Y* CGSvariant (supplemental Figure 1, see supplemental Data available atthe Blood Web site). This in turn suggests that the Y* variant im-proves expansion not only through increased receptor stability, butalso through a functional improvement in receptor signaling. Theimproved activity of the Y* CGS vector also allowed for a greaterthan fourfold improvement in the expansion from adult hematopoi-etic CD341 cells compared with the wt CGS vector (supplementalFigure 2A-B), a cell population which is relatively resistant to CGS-mediated expansion.

Characterization of CGS expansion products

Based on the starting cell number, CD235a1/CD41a2 erythroid cellswere expanded 70-fold by the Y*CGS vector compared with only21-fold by the wt CGS vector (Figure 3A). Although the output ofCD41a1/CD42b1megakaryocyteswas only 0.5-foldwhen compared

with the total number of starting cells, a comparison of this specificsubset of cells at day 3 vs day 11 of culture revealed a 22-fold ex-pansion for the Y* CGS vector, compared with fivefold for the wtCGSvector (Figure 3B).ThewtCGSvector provided for a stable levelof CD341 cells in these cultures, whereas the Y* CGS vector ex-panded the number of CD341 cells 4.4-fold by 21 days of culture(Figure 3C). This was correlated with a maintenance of about half of

Figure 1. Derivation and validation of hyperactive CGS variants. (A) Schematic

comparing the principal components of the Tpo receptor Mpl and the hybrid Mpl-

based CGS receptor. Tpo and CID not to scale. (B) Schematic indicating the po-

sitions of the amino acid substitutions introduced into the Mpl signaling domain of the

CGS receptor. (C) Design of the lentiviral vectors expressing the CGS variants Y*

(Y582F), K*K* (K544F and K564F), and K*K*Y*. Coordinates are based on amino

acid sequence of mouse Mpl. (D) Hyperactive variants improve CGS-mediated

expansion of Ba/F3 cells. Murine Ba/F3 cells were transduced with lentivectors

expressing the CGS containing the wt, Y*, K*K*, or K*K*Y* Mpl signaling domains.

After confirming similar rates of gene transfer, cells were cultured in the indicated

concentration of the CID AP1903. After 5 days, cell proliferation was assessed by the

MTS assay. Data represent the mean6 SD of triplicate determinations. P values are

based on the t test across all concentrations. (E) The hyperactive variant Y* im-

proves CGS receptor stability. Ba/F3 cells transduced with the wt or Y* CID

lentivectors and expanded in 100 nM CID AP20187 were treated with CYH for 4 and

8 hours, and the level of wt and Y* CID receptor proteins was compared with

untreated controls (0 mg/mL CYH) by western blotting using an HA tag antibody. The

relative levels of CGS protein compared with the untreated controls are shown at the

bottom. DLTR, self-inactivating long-terminal repeat; CYH, cycloheximide; GFP,

GFP reporter gene; MSCVpro, MSCV promoter; PGKpro, PGK gene promoter; SD,

standard deviation.

BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6 HYPERACTIVE CELL GROWTH SWITCH 1027

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

the initial progenitor colony-forming potential for the Y* CGS vector(Figure 3D), whereas the number of colony-forming cells declined toundetectable levels with the wt CGS vector (data not shown). Duringthe course of this analysis, we noted the expansion of a novel cellpopulation in these cultures staining for both CD235a and CD41a.Expression of CD235a (glycophorin A) is strictly associated witherythrocytes, erythroblasts, and their immediate precursors,33,34

whereas CD41a (integrin aIIb) is associated with megakaryocytesand platelets,35 as well as some other more primitive cells.36,37 Boththe absolute level of expansion (19-fold) and the period of positiveexpansion (17 days) for this CD235a1/CD41a1 population wasgreater with the Y* CGS vector compared with the wt CGS vector(Figure 3E). The Y* CGS vector also expanded CD235a1/CD41a1

cells from adult hematopoietic CD341 cells to a greater degree thanthe wt CGS vector (supplemental Figure 2C). As summarized inFigure 3F, the increased expansion from cord blood CD341 cellsprovided by the Y* CGS vector was distributed across all of the celltypes analyzed, with a moderate preference for CD235a1 erythroidcells at day 7, CD41a1/CD42b1 megakaryocytes at day 14, andCD235a1/CD41a1 cells at day 21 comparedwith thewtCGSvector.Additional studies demonstrated that all 3 of these populations werepredominantly positive for vector GFP, indicating their expansionwas CGS-dependent (supplemental Figure 3A). Furthermore, quan-titative polymerase chain reaction (PCR) analysis indicated that theresidualGFP-negativecellspresent in later-stageculturesdidnotcon-tain vector provirus, indicating that expansionof theCD235a2/CD41a2

population was CGS-independent (supplemental Figure 3B).

Characterization of the bipotent CD235a1/CD41a1

PEM population

To assess the expansion and differentiation potential of the novelCD235a1/CD41a1 population, cells from day 14 to 21 cord bloodCD341 cultures expandedwith theY*CGSvectorwere sorted basedon their expression pattern (Figure 4A). Subsequent clonogenicassays in methylcellulose demonstrated that almost all of the colony-forming progenitor cells were located in the CD235a2/CD41a2

population (Figure 4B), and that the few colonies that grew fromthe CD235a1/CD41a1 population frequently formed primitiveCFU-Mix colonies containing both CD235a1 erythroid- andCD41a1/CD42b1megakaryocyte-derived cells (Figure 4C). This isconsistent with the observation that almost all of the CD341 cellswere contained within the CD235a2/CD41a2 population. Secondarycultures in cytokine cocktails demonstrated that CD235a1/CD41a1

cells had the capacity to differentiate into both erythrocytes andmega-karyocytes (Figure 4D). This differentiation was not associated withexpansion of cell numbers, occurredmaximallywithin 2 days, andwasnot dependent on SCF (supplemental Figure 4). These are the sameproperties ascribedpreviously byothers to aTer1191/CD42d1bipotentpopulation arising during stress hematopoiesis in mice and nameda precursor for erythrocytes and megakaryocytes (PEM).27,28 Sub-sequent studies demonstrated that the CD235a1/CD41a1 cells in theCGS expansion cultures were predominantly negative for CD34 andCD123 (Figure 5A), both markers associated with primitive progen-itors. Analysis of mononuclear cells from 6 cord blood samples indi-cated the presence of a similar CD235a1/CD41a1 population at afrequency of 0.27%6 0.15% (range, 0.11%-0.49%; see Figure 5B forexample), or about half the frequency of PEM in bone marrow ofuntreatedmice.28 PEMcells from theCGSexpansion cultures and fromfresh cord blood were also predominantly negative for CD38, CD90,and the SCF receptor CD117, 3 othermarkers associatedwith primitive

Figure 2. Hyperactive variants improve CGS-mediated expansion from cord

blood CD341 cells. (A) Expansion of total cells. Cord blood-derived CD341 cells

were transduced with lentivectors expressing the wt, Y*, K*K*, and K*K*Y* variants

of the CGS and expanded in serum-containing medium supplemented with 100 nM

CID AP20187 and no additional cytokines. Cell numbers were determined by he-

mocytometer count with media changes on the indicated days, and are reported as

a fold-expansion based on the starting cell number. The maximum fold-expansion is

shown for the wt and Y* samples. P values are based on the t test across all time

points. P 5 .16 for K*K* vs K*K*Y*. (B) Expansion of transduced cells. The percent

of cells expressing vector GFP was also determined on the indicated days by flow

cytometry as a means of determining the degree of expansion for the vector-

transduced cells. Data represent the mean 6 standard error from 4 independent

experiments. P values are based on the t test across all time points. (C) Morphology

of expanded cells. Representative Wright-Giemsa–stained cytospins are shown

for the original cord blood mononuclear cells (CB MNC) used to prepare the

CD341 cells, and from day 14 and day 21 cultures expanded with the wt and Y*

CGS vectors.

1028 BELAY et al BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

HSPCs (supplemental Table 1). Real-time PCR analysis demonstratedthat PEM sorted from Y* CGS-expanded cord blood expressed genesassociated with both the erythroid and megakaryocytic lineages, in-cluding high levels of the erythropoietin receptor, low to moderatelevels of a-globin, high levels of CD41, but almost no detectable re-ceptor for the megakaryocyte-specific genes MPL or PF4, consistentwith the properties of PEM from mice (supplemental Figure 5).27 Thissame analysis also demonstrated that PEM from expansion culturesexpressed an even higher ratio of fetal g-globin than primary erythroidcells from cord blood (supplemental Figure 6). However, the overalllevels of all globin transcripts in the PEM were lower than the levelsfound inCD235a1/CD41a2cells collected fromcord blood, andhigherthan the levels found in CD235a1/CD41a2 cells collected from theCGS expansion cultures.

Characterization of macrophages in cord blood CD341

expansion cultures

Macrophages play a central role in normal erythrocyte maturation,forming the basis of erythroid islands found in normal bone mar-row.38 Cytospin analysis revealed the presence of macrophages inCGS expansion cultures, and that these macrophages were often phys-ically surrounded by maturing erythroblasts (Figure 6A). Immunoflu-orescent staining andflowcytometry (Figure 6B), aswell as sorting andcytospin studies (Figure 6C), confirmed the presence of activatedCD2061 macrophages in late-stage cultures expanded with the Y*CGS. Subsequent tracking studies demonstrated that the relativefrequency of these CD2061macrophages increases sharply around

day 7 of culture, coincident with the onset of active erythropoiesis,but then declines to a maintenance level of 2% to 3% of all cells inthe cultures (Figure 6D). Surprisingly, the relative frequency ofCD2061 macrophages expressing vector GFP remains unchangedthroughout the culture (Figure 6B,D), whereas the total number ofCD2061 cells expands about 110-fold by day 26 of culture (datanot shown). As such, it appears that CD2061 macrophages, or amacrophage progenitor/precursor populations, undergo a substan-tial degree of CGS-independent expansion in these cultures.

Discussion

Incorporating the Mpl signaling domain into the CGS platformallows for the expansion of a much wider range of hematopoieticlineages than can be achieved with native Tpo/Mpl. There are likely2main reasons for this difference: (1) nativeMpl is predominantly ex-pressed on primitive HSPCs and megakaryocytes, whereas the CGSvector is expressed constitutively; and (2) native Mpl signaling isdownregulated in part by internalization of dimerized receptors,whereas CGS receptor signaling occurs intracellularly. In addition,native Mpl signaling is difficult to block in vivo due to circulatingTpo, whereas CGS signaling can be tightly regulated by the additionor withdrawal of the CID while avoiding off-target effects. Also,unlike native Mpl, expansion by this CGS platform has an inherentbias for the erythroid lineage, followed bymegakaryocytes/platelets,a subset of myeloid cells, and when assessed in vivo, lymphoidcells.13-18,22,23 These properties constitute fundamental differences

Figure 3. Hyperactive CGS variant improves expansion of both mature and primitive cells. Expansion cultures were established with cord blood CD341 cells transduced with

lentivectors expressing the wt or Y* CGS as for Figure 2. Cell aliquots were analyzed by immunofluorescent staining and flow cytometry for (A) erythroid cells (CD235a1/CD41a2), (B)

megakaryocytic cells (CD41a1/CD42b1), (C) primitive stem/progenitor cells (CD341), and (E) bipotent erythroid/megakaryocytic precursor cells (CD235a1/CD41a1), later determined to be

PEMs. Data represent the mean6 standard error (SE) from 4 independent experiments for the fold-expansion based on the starting number of total cells. P values are based on the

t test across all time points. The maximum fold-expansion is shown for the wt and Y* samples based on each parameter. (D) Histograms showing the fold expansion of

progenitor colony-forming cells present in the cord blood CD341 expansion cultures transduced with the Y* lentivector and expanded for the indicated days. Colonies were

detected by plating cell aliquots in methylcellulose supplemented with a broad cytokine cocktail, and consisted predominantly of primitive and mature erythroid colonies,

granulocyte-macrophage colonies, and low levels of megakaryocyte colonies and colonies with mixed phenotypes. Data represents the mean 6 SE from 3 independent

experiments. (F) Histograms showing the relative frequency of the indicated cell populations present in the cord blood CD341 cultures transduced with the wt or Y* lentivectors

and expanded for 7, 14, and 21 days.

BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6 HYPERACTIVE CELL GROWTH SWITCH 1029

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

between native Mpl (including studies with viral vectors expressingnative Mpl),39 and our Mpl-based CGS platform. The studies pre-sented here suggest that theMpl-based CGS receptor is subject to thesame receptor endocytosis pathways governing native Mpl. How-ever, it appears that the link between blocking receptor degradationand an increased capacity for cellular proliferationmay bemore com-plicated than originally envisioned31,32 because the improved expan-sion achieved by the hyperactive variants cannot be fully recapitulatedby simply increasing the dose of the wt CGS receptor. Future studieswill be needed to identify the mechanism(s) underlying this increasein receptor signaling, to determine why the Y* variant performed sta-tistically better than the K*K*Y* variant, and to determine whethersimilar changes occur in the setting of native Mpl.

We found that thesemodifications to theMpl signaling domain ofthe CGS receptor substantially improved the overall activity of thisCGS receptor across all hematopoietic lineages, including CD341

cells and colony-forming progenitors. These effects are consistentwith the observation that treating mice with low doses of exoge-nous Tpo induced HSPC quiescence, whereas similar studies withsuperphysiological doses of Tpo resulted in HSPC expansion.25,26

Although the erythroid and megakaryocytic cells generated in theexpansion cultures describedherewere not directly assessed for func-tionality in vivo, previous studies inmice and dogs demonstrated thaterythrocytes and platelets generated by the CGS in vivo exhibit thesame half-lives as their normal counterparts and do not exhibit grossphysiological defects.13-18,22,23

The PEM population identified in cord blood and adult CD341

cultures expanded with the Y* CGS receptor shares several prop-erties with the PEM population described by others in associationwith stress hematopoiesis in mice.27,28 Both populations simulta-neously express cell surface markers typically associated with ter-minal erythropoiesis and megakaryopoiesis, but are negative formarkers typically associated with HSPCs (CD34/ScaI). Both pop-ulations exhibit the bipotent capacity to differentiate into eithererythrocytes ormegakaryocytes, very little progenitor colony-formingability, and gene expression patterns that represent a mix of erythroidand megakaryocytic profiles but with a clear erythroid bias. Finally,both populations are present, albeit at low levels (,1%) under steady-state conditions invivo.These similarities suggest that thePEMarisingin CGS-expanded cultures, as well as the erythrocytes dominatingthese cultures, arise through the mechanism(s) underlying stress he-matopoiesis. This is consistent with the retention of a high ratio offetal to adult hemoglobin in both the PEM and their progeny.

Based on the similarities to mouse PEM, and the characterizationpresented here, we propose that PEM represent an ontologically dis-tinct population in the erythroid/megakaryocytic arm of human hema-topoiesis. As summarized in Figure 740-45, PEM do not express theprimitive cell surfacemarker CD34 characteristic ofmegakaryocyte-erythrocyte progenitors (MEPs), colony-forming unit-megakaryocytes(CFU-Meg), or burst-forming unit-erythroid (BFU-E). However, PEMdo share the CD1232 phenotype characteristic of MEP, but not CFU-Meg, BFU-E, or CFU-erythroid (CFU-E). Based on these properties,

Figure 4. Differentiation potential of bipotent CD235a1/CD41a1 PEM. (A) Representative flow cytometric histogram demonstrating the CD235a1/CD41a1

immunofluorescent staining pattern of cord blood CD341 cells transduced with the Y* CGS lentiviral vector and expanded for 14 days in 100 nM AP20187. Gates were

set based on staining controls. The percent of cells in each quadrant are indicated. (B) Relative progenitor content of CD235a/CD41a subpopulations. Cells were sorted from

day 14 to 21 cord blood CD341 cultures expanded with the Y* CGS lentiviral vector based on presence or absence of CD235a and CD41a expression, and plated in

methylcellulose cultures supplemented with a broad cytokine cocktail. Data represent the mean6 SD from 3 independent experiments. P values are based on the t test vs the

unsorted control. (C) Example of a mixed progenitor colony generated from sorted CD235a1/CD41a1 PEM, showing the presence of both erythroid bursts (dark globules at

bottom) and nonerythroid cells (white colony at top). (D) Differentiation potential of PEM. CD235a1/CD41a1 PEM were sorted from day 14 cord blood CD341 Y* CGS

expansion cultures and transferred to media supporting either erythroid (Epo 1 SCF) or megakaryocytic (TPO, IL-6, IL-9, SCF) differentiation. After 6 days, the cells were

collected, prepared as cytospins, and stained by Wright-Giemsa. Left, sorted cells before secondary culture characterized morphologically as erythroid precursors; middle,

representative of erythroid cultures characterized morphologically as enucleated red blood cells and a few pre-erythrocytes; and right, representative of megakaryocyte

cultures characterized morphologically as megakaryocytes exhibiting mono- and polylobular nuclei and surrounded by a halo of platelets, and confirmed as CD41a1/CD42b1

by immunofluorescent staining and flow cytometry.

1030 BELAY et al BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

the lack of colony-forming ability, and the expression of markers andgenes associated with terminal erythropoiesis and megakaryopoiesis,we propose that PEM bypass the CFU-Meg and BFU-E/CFU-E clas-sical pathways, and instead arise directly from MEP as a shortcut togenerate red blood cells (RBCs) and megakaryocytes under conditionsof stress. The observation that PEM also constitute a major componentof the cultures expanded by the Y* CGS from adult-derived CD341

cells indicates that PEM are not restricted to cord blood, but insteadlikely represent a previously unidentified population in definitive hu-man hematopoiesis.

The importance of macrophages in these cultures remains unclear.They surge in number during the onset of large-scale erythropoiesis inthe expansion cultures, interact directly with maturing erythrocytes,and appear to form erythroid islands reminiscent of those found inbone marrow. Furthermore, these macrophages, or progenitors givingrise to these macrophages, are expanding in a CGS-independent man-ner in the CID-treated cultures, but not in CID-untreated cultures. Thisin turn suggests that the macrophages are expanding in response to theCGS-driven production of CFU-E and/or their physical interactionswith maturing erythrocytes arising from these CFU-E. As such, thisCGS-based culture system may prove valuable for dissecting boththe mechanisms supporting the macrophages at the heart of erythroidislands, and the mechanisms capable of supporting Epo-independentterminal erythropoiesis in this setting.

Optimizing the activity of theMpl-based CGS receptor and under-standing the cellular components of the associated expansion cultures

are critical for translating the CGS expansion platform to the clinic.The studies reported here describe a fundamental enhancement of theCGS expansion platform, identify a novel precursor population inthe erythroid/megakaryocyte differentiation pathway of humans thatconstitutes a major component of the CGS expansion cultures, and

Figure 5. Identification of PEM population in unexpanded cord blood. Cells

collected from day 14 cord blood CD341 Y* CGS expansion cultures (A) and fresh

CB MNCs (B) were analyzed by immunofluorescent staining and flow cytometry for

CD235a1/CD41a1 PEM and coexpression of the primitive hematopoietic markers

CD34 and CD123. Gates were set based on isotype staining controls (see examples

shown under the 2-color plots). The percent of cells in each quadrant or gate are indicated.

Figure 6. Macrophages are passengers in CGS expansion cultures. (A)

Cytopathology evidence for macrophages. Cytospins were prepared from day

14 cord blood CD341 Y* CGS expansion cultures and stained by Wright-Giemsa. The

representative slide shows the presence of morphologically distinct macro-

phage (Mac) and their tight physical association with developing erythroblasts

reminiscent of erythroid islands. The presence of an enucleated RBC is also

indicated. (B) Immunophenotyping evidence for macrophages. Cells from day

14 cord blood CD341 Y* CGS expansion cultures were stained for the activated

macrophage cell surface marker CD206 and analyzed by flow cytometry. The

representative histogram shows the presence of 5% CD2061 cells, and that only

20% of these also express viral vector GFP. Gates were set based on staining controls.

The percentage of cells in each quadrant is indicated. (C) Cytopathology of CD2061

cells. Cells expressing CD206 were sorted and analyzed by cytospin and Wright-

Giemsa staining. The representative slide shows a preponderance of large macro-

phages with large foamy cytoplasm and blue nuclei. (D) Contribution of macrophages to

expansion cultures. The relative frequency of CD235a1/CD41a2 erythroid cells and

activated CD2061 macrophages were tracked over time in cord blood CD341 cultures

expanded with the Y* CGS by immunofluorescent staining and flow cytometry. The

relative frequency of cells expressing vector GFP is shown for the 2 populations

early and late in culture. Data represent the mean 6 SE from 3 independent

experiments. P values are based on the t test across all time points.

BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6 HYPERACTIVE CELL GROWTH SWITCH 1031

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

point to a potentially important role for macrophages in these expan-sion cultures. Although some CGS-based cultures have been reportedto support the expansion of primitive hematopoietic cells,15,19,21,22 thecultures described here do not support the large-scale expansion ofprimitive HSPCs necessary for enhancing long-term engraftment.

However, the large-scale expansion of cells committed to the erythroid/megakaryocytic lineage could be considered as a means of transientlysupporting patients undergoing fully myeloablative HSPC transplanta-tion, especially with sources of cells such as cord bloodwhich typicallyengraft more slowly, as well as patients requiring repeated transfusiontherapy.

Acknowledgments

The authors thank Catherine M. Coy for assistance with the westernblotting, George E. Sale for interpretation of cytospins, and AriadPharmaceuticals for the CIDs AP20187 and AP1903.

This work was supported by the National Institutes of Health(grant RO1 DK74522 [C.A.B.], grant P30 DK56465 [B.T.-S.], andgrant UO1HL099993 [B.T.-S. and D.W.E.]), a grant from the TietzeFoundation (D.W.E.), and a grant from theAmerican Cancer Society(117682-MRSG-09-268-01-CCE [C.P.M.]).

Authorship

Contribution: E.B., C.P.M., B.T.-S., C.A.B., and D.W.E. designedexperiments and analyzed data; E.B., C.P.M., A.N.K., and D.W.E.performed experiments; E.B., C.P.M., and D.W.E. prepared thefigures; and E.B. and D.W.E. wrote the manuscript.

Conflict-of-interest disclosure: C.A.B. is named as an inventor ona patent for the CGS technology licensed to Ariad PharmaceuticalsInc through the University ofWashington. The remaining authorsdeclare no competing financial interests.

Correspondence: David W. Emery, Institute for Stem Cell andRegenerative Medicine, University of Washington, 850 RepublicanSt, Box 358056, Seattle, WA 98109; e-mail demery@uw.edu.

References

1. McKinney M, Arcasoy MO. Erythropoietin foroncology supportive care. Exp Cell Res. 2011;317(9):1246-1254.

2. Falzetti F, Aversa F, Minelli O, Tabilio A.Spontaneous rupture of spleen during peripheralblood stem-cell mobilisation in a healthy donor.Lancet. 1999;353(9152):555.

3. Miller CP, Blau CA. Using gene transfer tocircumvent off-target effects. Gene Ther. 2008;15(10):759-764.

4. Champlin RE, Schmitz N, Horowitz MM, et al;IBMTR Histocompatibility and Stem CellSources Working Committee and the EuropeanGroup for Blood and Marrow Transplantation(EBMT). Blood stem cells compared with bonemarrow as a source of hematopoietic cells forallogeneic transplantation. Blood. 2000;95(12):3702-3709.

5. Bohlius J, Tonia T, Schwarzer G. Twist and shout:one decade of meta-analyses of erythropoiesis-stimulating agents in cancer patients. ActaHaematol. 2011;125(1-2):55-67.

6. Newburger PE, Dale DC. Evaluation andmanagement of patients with isolatedneutropenia. Semin Hematol. 2013;50(3):198-206.

7. Larochelle A, Dunbar CE. Genetic manipulation ofhematopoietic stem cells. Semin Hematol. 2004;41(4):257-271.

8. Dahlberg A, Delaney C, Bernstein ID. Ex vivoexpansion of human hematopoietic stem andprogenitor cells. Blood. 2011;117(23):6083-6090.

9. Adair JE, Beard BC, Trobridge GD, et al.Extended survival of glioblastoma patients afterchemoprotective HSC gene therapy. Sci TranslMed. 2012;4(133):133ra57.

10. Watts KL, Adair J, Kiem HP. Hematopoietic stemcell expansion and gene therapy. Cytotherapy.2011;13(10):1164-1171.

11. Blau CA, Peterson KR, Drachman JG, SpencerDM. A proliferation switch for genetically modifiedcells. Proc Natl Acad Sci USA. 1997;94(7):3076-3081.

12. Jin L, Siritanaratkul N, Emery DW, et al. Targetedexpansion of genetically modified bone marrowcells. Proc Natl Acad Sci USA. 1998;95(14):8093-8097.

13. Jin L, Zeng H, Chien S, et al. In vivo selectionusing a cell-growth switch. Nat Genet. 2000;26(1):64-66.

14. Zeng H, Masuko M, Jin L, Neff T, Otto KG, BlauCA. Receptor specificity in the self-renewal anddifferentiation of primary multipotentialhemopoietic cells. Blood. 2001;98(2):328-334.

15. Weinreich MA, Lintmaer I, Wang L, Liggitt HD,Harkey MA, Blau CA. Growth factor receptors asregulators of hematopoiesis. Blood. 2006;108(12):3713-3721.

16. Neff T, Horn PA, Valli VE, et al. Pharmacologicallyregulated in vivo selection in a large animal.Blood. 2002;100(6):2026-2031.

17. Okazuka K, Beard BC, Emery DW, et al. Long-term regulation of genetically modified primaryhematopoietic cells in dogs [published correctionappears in Mol Ther. 2011;19(11):2102]. MolTher. 2011;19(7):1287-1294.

18. Richard RE, Wood B, Zeng H, Jin L,Papayannopoulou T, Blau CA. Expansion ofgenetically modified primary human hemopoieticcells using chemical inducers of dimerization.Blood. 2000;95(2):430-436.

19. Richard RE, Blau CA. Small-molecule-directedmpl signaling can complement growth factors toselectively expand genetically modified cord bloodcells. Stem Cells. 2003;21(1):71-78.

20. Richard RE, De Claro RA, Yan J, et al.Differences in F36VMpl-based in vivo selectionamong large animal models. Mol Ther. 2004;10(4):730-740.

21. Abdel-Azim H, Zhu Y, Hollis R, et al. Expansion ofmultipotent and lymphoid-committed humanprogenitors through intracellular dimerization ofMpl. Blood. 2008;111(8):4064-4074.

22. Parekh C, Sahaghian A, Kim W, et al. Novelpathways to erythropoiesis induced bydimerization of intracellular C-Mpl in humanhematopoietic progenitors. Stem Cells. 2012;30(4):697-708.

Figure 7. PEM population in the hierarchy of human hematopoiesis. The

schema indicates the proposed placement of PEM in the erythrocyte-megakaryocyte

differentiation arm of human hematopoiesis. The expression patterns of key CD anti-

gens are shown for principal cell populations. Curved arrow, self-renewal capacity;

straight arrow, differentiation path; broken arrows, minimal self-renewal or differen-

tiation potential. References for progenitor-specific markers: MEP33-35,40,41; CFU-

Meg30,33-35,41,42; CFU-E/BFU-E.33-35,41,43-45 CD, cluster of differentiation.

1032 BELAY et al BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

23. Nagasawa Y, Wood BL, Wang L, et al. Anatomicalcompartments modify the response of humanhematopoietic cells to a mitogenic signal. StemCells. 2006;24(4):908-917.

24. Kaushansky K. Historical review:megakaryopoiesis and thrombopoiesis. Blood.2008;111(3):981-986.

25. Qian H, Buza-Vidas N, Hyland CD, et al. Criticalrole of thrombopoietin in maintaining adultquiescent hematopoietic stem cells. Cell StemCell. 2007;1(6):671-684.

26. Yoshihara H, Arai F, Hosokawa K, et al.Thrombopoietin/MPL signaling regulateshematopoietic stem cell quiescence andinteraction with the osteoblastic niche. Cell StemCell. 2007;1(6):685-697.

27. Vannucchi AM, Paoletti F, Linari S, et al.Identification and characterization of a bipotent(erythroid and megakaryocytic) cell precursorfrom the spleen of phenylhydrazine-treated mice.Blood. 2000;95(8):2559-2568.

28. Sanchez M, Weissman IL, Pallavicini M, et al.Differential amplification of murine bipotentmegakaryocytic/erythroid progenitor andprecursor cells during recovery from acute andchronic erythroid stress. Stem Cells. 2006;24(2):337-348.

29. Dull T, Zufferey R, Kelly M, et al. A third-generation lentivirus vector with a conditionalpackaging system. J Virol. 1998;72(11):8463-8471.

30. Robert A, Cortin V, Garnier A, Pineault N.Megakaryocyte and platelet production fromhuman cord blood stem cells. Methods Mol Biol.2012;788:219-247.

31. Saur SJ, Sangkhae V, Geddis AE, Kaushansky K,Hitchcock IS. Ubiquitination and degradation ofthe thrombopoietin receptor c-Mpl. Blood. 2010;115(6):1254-1263.

32. Hitchcock IS, Chen MM, King JR, Kaushansky K.YRRL motifs in the cytoplasmic domain of thethrombopoietin receptor regulate receptorinternalization and degradation. Blood. 2008;112(6):2222-2231.

33. Rogers CE, Bradley MS, Palsson BO, Koller MR.Flow cytometric analysis of human bone marrowperfusion cultures: erythroid development andrelationship with burst-forming units-erythroid.Exp Hematol. 1996;24(5):597-604.

34. Hu J, Liu J, Xue F, et al. Isolation and functionalcharacterization of human erythroblasts at distinctstages: implications for understanding of normaland disordered erythropoiesis in vivo. Blood.2013;121(16):3246-3253.

35. Basch RS, Zhang XM, Dolzhanskiy A, KarpatkinS. Expression of CD41 and c-mpl does notindicate commitment to the megakaryocytelineage during haemopoietic development. Br JHaematol. 1999;105(4):1044-1054.

36. Woo KS, Goh RY, Kim SH, et al. Number ofmegakaryocytic progenitors and adhesionmolecule expression of stem cells predict plateletengraftment after allogeneic hematopoietic stemcell transplantation. Cytotherapy. 2007;9(6):555-561.

37. Mikkola HK, Fujiwara Y, Schlaeger TM, Traver D,Orkin SH. Expression of CD41 marks the initiationof definitive hematopoiesis in the mouse embryo.Blood. 2003;101(2):508-516.

38. Chasis JA, Mohandas N. Erythroblastic islands:niches for erythropoiesis. Blood. 2008;112(3):470-478.

39. Wicke DC, Meyer J, Buesche G, et al. Genetherapy of MPL deficiency: challenging balancebetween leukemia and pancytopenia. Mol Ther.2010;18(2):343-352.

40. Chen X, Skutt-Kakaria K, Davison J, et al. G9a/GLP-dependent histone H3K9me2 patterningduring human hematopoietic stem cell lineagecommitment. Genes Dev. 2012;26(22):2499-2511.

41. Manz MG, Miyamoto T, Akashi K, Weissman IL.Prospective isolation of human clonogeniccommon myeloid progenitors. Proc Natl Acad SciUSA. 2002;99(18):11872-11877.

42. De Bruyn C, Delforge A, Martiat P, Bron D. Exvivo expansion of megakaryocyte progenitor cells:cord blood versus mobilized peripheral blood.Stem Cells Dev. 2005;14(4):415-424.

43. Gainsford T, Roberts AW, Kimura S, et al.Cytokine production and function in c-mpl-deficient mice: no physiologic role for interleukin-3in residual megakaryocyte and plateletproduction. Blood. 1998;91(8):2745-2752.

44. Nakahata T, Okumura N. Cell surface antigenexpression in human erythroid progenitors:erythroid and megakaryocytic markers. LeukLymphoma. 1994;13(5-6):401-409.

45. Flygare J, Rayon Estrada V, Shin C, Gupta S,Lodish HF. HIF1alpha synergizes withglucocorticoids to promote BFU-E progenitorself-renewal. Blood. 2011;117(12):3435-3444.

BLOOD, 5 FEBRUARY 2015 x VOLUME 125, NUMBER 6 HYPERACTIVE CELL GROWTH SWITCH 1033

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom

online October 24, 2014 originally publisheddoi:10.1182/blood-2014-02-555318

2015 125: 1025-1033  

W. EmeryEyayu Belay, Chris P. Miller, Amanda N. Kortum, Beverly Torok-Storb, C. Anthony Blau and David erythroid-megakaryocytic precursormacrophage-associated erythropoiesis through an A hyperactive Mpl-based cell growth switch drives 

http://www.bloodjournal.org/content/125/6/1025.full.htmlUpdated information and services can be found at:

(625 articles)Red Cells, Iron, and Erythropoiesis    (546 articles)Platelets and Thrombopoiesis    (3297 articles)Hematopoiesis and Stem Cells   

(3000 articles)Free Research Articles    Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

  Copyright 2011 by The American Society of Hematology; all rights reserved.of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society

For personal use only.on March 11, 2015. by guest www.bloodjournal.orgFrom