hematopoietic stem cell transplantation - hindawi publishing

54
Hematopoietic Stem Cell Transplantation Bone Marrow Research

Upload: others

Post on 09-Feb-2022

6 views

Category:

Documents


0 download

TRANSCRIPT

Hematopoietic Stem Cell Transplantation

Bone Marrow Research

Hindawi Publishing CorporationBone Marrow ResearchVolume 2011, Article ID 570796, 15 pagesdoi:10.1155/2011/570796

Review Article

Future Perspectives: Therapeutic Targeting ofNotch Signalling May Become a Strategy in Patients ReceivingStem Cell Transplantation for Hematologic Malignancies

Elisabeth Ersvaer,1 Kimberley J. Hatfield,1 Hakon Reikvam,1 and Øystein Bruserud1, 2

1 Division of Hematology, Department of Medicine, Haukeland University Hospital and Institute of Medicine,University of Bergen, N-5021 Bergen, Norway

2 Medical Department, Haukeland University Hospital, N-5021 Bergen, Norway

Correspondence should be addressed to Øystein Bruserud, [email protected]

Received 26 July 2010; Accepted 30 August 2010

Academic Editor: Joseph H. Antin

Copyright © 2011 Elisabeth Ersvaer et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

The human Notch system consists of 5 ligands and 4 membrane receptors with promiscuous ligand binding, and Notch-initiatedsignalling interacts with a wide range of other intracellular pathways. The receptor signalling seems important for regulation ofnormal and malignant hematopoiesis, development of the cellular immune system, and regulation of immune responses. SeveralNotch-targeting agents are now being developed, including natural receptor ligands, agonistic and antagonistic antibodies, andinhibitors of intracellular Notch-initiated signalling. Some of these agents are in clinical trials, and several therapeutic strategiesseem possible in stem cell recipients: (i) agonists may be used for stem cell expansion and possibly to enhance posttransplantlymphoid reconstitution; (ii) receptor-specific agonists or antagonists can be used for immunomodulation; (iii) Notch targetingmay have direct anticancer effects. Although the effects of therapeutic targeting are difficult to predict due to promiscuousligand binding, targeting of this system may represent an opportunity to achieve combined effects with earlier posttransplantreconstitution, immunomodulation, or direct anticancer effects.

1. Introduction

The most important members of the human Notch sys-tem are the four Notch receptors and their five lig-ands. Notch-mediated signalling is important in embryonichematopoiesis and development of the immune system,regulation of the peripheral immune system, and devel-opment of hematological malignancies, especially T cellacute lymphoblastic leukemia (T-ALL) [1–3]. Thus, forpatients treated with allogeneic stem cell transplantationfor hematological malignancies, agonistic or antagonistictargeting of Notch signalling may become useful to (i)achieve more effective and safe antileukemic treatment andthereby reduce the risk of posttransplant relapse throughdirect targeting of the malignant cells, (ii) enhance T cellreconstitution and thereby reduce posttransplant immunedefects, and (iii) develop new immunomodulatory strategiesthat can reduce the risk of severe infections and severe

graft versus host disease (GVHD) without inhibition of graftversus leukemia (GVL) effects. Even a combination of theseeffects may become a possible treatment by careful selectionof molecular targets.

2. Notch Molecules, Notch Ligands, andDownstream Signalling

2.1. Notch and Notch Ligands. Humans possess the fourheterodimeric transmembrane Notch receptors Notch1-4that can bind the five transmembrane ligands Delta-like 1,3, and 4 (DLL1/3/4) and Jagged 1 and 2 (JAG1/2) (Figure 1).The receptor chains are cleaved by a furin-like protease inthe Golgi apparatus during their way to the cell surfacewhere they form heterodimeric receptors. These receptorsconsist of an extracellular subunit (NEC) with a distant partwith a variable number of glycosylated Epithelial growth

2 Bone Marrow Research

JAG1, JAG2

DLL1, DLL4DLL3

N1 N2 N3 N4

CR

EGF

DSL

RAM

ANK

TAD

PEST

Ligands Receptors

LIN

EGF

HD

Signal-receiving cell

Signal-initiating cell

Figure 1: Notch receptors and their ligands. Signal-initiating cellsexpress Notch ligands of the Delta-like (DLL1, DDL3, DLL4)or Jagged families (JAG1, JAG2). Common structural featuresof all ligands are the Epithelial growth factor-like (EGF) repeatsand the distal amino-terminal domain called DSL (Delta, Serrate,and Lag-2). DSL is involved in receptor binding. Additionally,JAG1 and JAG2 contain a proximal cysteine-rich (CR) domainbetween the EGF-like repeats and the plasma membrane. Inhumans there are four heterodimeric Notch receptors (Notch1-4;N1-N4). The extracellular Notch receptor domain contains EGF-like repeats, a cysteine-rich LIN-12 repeats (LIN domain) thatprevents ligand-independent activation, and the proximal het-erodimerization domain (HD). The cytoplasmic domain containsan RBP-J-associated molecule (RAM) domain (closest to the cellmembrane) followed by ankyrin repeats (ANK) that bind to theCSL (CBF1/RBP-Jκ/Suppressor of Hairless/LAG-1) transcriptionfactor, a transactivation domain (TAD; only Notch 1 and 2),and a PEST (proline, glutamic acid, serine, threonine) sequencethat is important for stabilization of the protein (adapted from[1]).

factor (EGF-) like repeats followed by LIN domains thatprevent ligand-independent activation. The transmembraneand cytoplasmic (NTM) subunit consists of the cytoplasmicRAM domain followed by ankyrine repeats that bind to theeffector transcription factor CBF1, two nuclear localizationsignals, a transactivation domain that is present only inNotch1 and Notch2, and finally a PEST sequence involvedin stabilization of the protein.

The five ligands also differ in their structure (Figure 1):the amino-terminal DSL domain (Delta, Serrate, and Lag-2) which is involved in receptor binding is common toall ligands; this is followed by a variable number of EGFrepeats; JAG1/2 contains an additional C-terminal cysteine-rich domain (CR). The Delta ligands seem to have twoactivities: to trans-activate Notch in neighboring cells andto cis-inhibit Notch in its own cells [5]. Glycosylation ofthe extracellular Notch domain modulates ligand-initiatedNotch signalling; for example, with regard to one experi-mental model DLL ligands were preferred over JAG ligandswhen the receptors contained N-Acetylglucosamine on theO-fucose residues in the EGF-like repeats [1, 6]. Notchreceptors seem to be promiscuous with regard to ligandbinding, although it should be emphasized that the ligandspecificity of the various receptors has not been characterizedin detail. Notch1, Notch2, and Notch3 can all be activatedby different ligands like DLL1, JAG1/2 [7, 8]; however, asmentioned, the ligand specificity of Notch1-4 has not beencharacterized in detail.

2.2. Canonical Intracellular Signalling. The first event incanonical Notch signalling (Figure 2) is ligand-receptorinteraction with initiation of two successive proteolyticcleavages of the receptors (at sites S2 and S3 by ADAM familyprotease and the γ-secretase, resp.) and thereby the release ofthe Notch IntraCellular Domain (NICD). NICD translocatesto the nucleus where it heterodimerizes with the DNA-binding transcription factor CBF1 (also named CSL or Rbp-j) and recruits other coactivators, including mastermind-likeproteins (MAML1-3) and the MED8-mediator transcriptionactivation complex; this leads to induction of transcrip-tional expression of target genes. The Notch-associated geneexpression profile will not depend on the receptor mediatingthe signal only since binding of different ligands to the samereceptor will in some cases have different functional effects[1, 9].

2.3. Noncanonical Signalling. Noncanonical Notch signallingis well documented [2, 10], but less characterized thanthe canonical pathway. There are probably three typesof noncanonical Notch signalling: Type I involves Notchligation and translocation of activation signals independentof CBF1 (NICD-dependent but CBF1-independent); TypeII involves activation of Notch target genes independentof S3 cleavage (NICD- and CBF1-independent); Type IIIinvolves CBF1-dependent gene activation without receptorcleavage and NICD release [10]. Several signalling pathwaysare involved, including Hedgehog, Jak/STAT, RTK, TGF,Wnt, PI3/Akt, mTor/Akt, JNK, MEK/ERK, and NFκB [2, 10].

Bone Marrow Research 3

ADAM-familyprotease (S2) γ-secretase

complex (S3)

NICD

NotchLigand

Target gene activation

CBF1

Co-activator

NICD MAML1

1

23

4

Hes

Hey

Notch

Signal-initiating cell

Cyt

opla

smN

ucl

eus

Signal-receiving cell

Figure 2: Notch canonical signalling pathway. The figure shows anoverview of the canonical Notch signalling pathway (adapted from[4]). The signalling is initiated by ligand binding to Notch receptor(1). This binding initiates two subsequent proteolytic cleavagesof the Notch receptor by the ADAM family protease (2) andthe γ-secretase, respectively (3). The Notch IntraCellular Domain(NICD) is thereby released to the cytoplasm (4) and translocatesto the nucleus where NICD heterodimerizes with the DNA-bindingtranscription factor CBF1 (also named CSL or Rbp-j). Additionalcoactivators are also recruited, including mastermind-like proteins(MAML1-3) ultimately leading to induction of transcriptionalexpression of downstream target genes, including those belongingto the Hes and Hey families.

Noncanonical Notch signalling seems important for main-tenance of lineage-restricted hematopoietic progenitors, andseveral of the mediators involved in this signalling are inaddition important in leukemogenesis as well as regulation ofcellular immune responses. The noncanonical pathway thusrepresents a point of crosstalk between other intracellularsignalling pathways.

3. Notch and Hematopoietic Progenitors

3.1. Notch in the Hematopoietic System. Notch1-mediatedsignals are essential for generation of definitive Hematopoi-etic stem cells (HSCs) during embryogenesis [11] though asdescribed in more detail below, canonical Notch signallingseems to be dispensable for HSC maintenance in adults.HSCs reside primarily in the bone marrow in a complexmicroenvironment consisting of stromal cells, microvesselsand extracellular matrix. These cells have the dual abilityto self-renew as well as being able to give rise to all thecells in the hematopoietic system. Mainly HSCs are ina quiescent state; that is, the cells are in G0/G1 of thecell cycle and do not proliferate. An important factor inregulating the fate of HSCs in terms of HSC quiescence,self-renewal and differentiation, is the surrounding stem cellmicroenvironment, the so-called stem cell niche. HSCs havebeen shown to be in close proximity to cells lining theendosteum as well as near the specialised blood vessels inbone marrow called sinusoids [12].

3.2. The Osteoblastic Stem Cell Niche. The osteoblastic niche(also referred to as the endosteal niche) [13, 14] and thevascular niche [12, 15] create a supportive environment forstem cells. Notch signalling is thought to be a key signallingpathway involved in maintenance and expansion of the HSCpool. In addition, an important role of Notch signalling inosteoblast and osteoclast homeostasis was recently described[16, 17]. Hematopoietic progenitor cells express Notchreceptors and are exposed to Notch ligands in the bonemarrow such as expression of JAG1 and DLL1 by osteoblasts[13, 18]. In a study by Calvi et al., parathyroid hormonestimulation of osteoblasts in mice resulted in inducedosteoblastic proliferation with increased expression of JAG1and a Notch1-mediated expansion of HSCs [13, 19]. Theseobservations identified Notch as an important component ofthe stem cell niche that supports osteoblastic HSC regulation.However, further studies of osteoblastic regulation of HSCsvia the Notch pathway have yielded conflicting results. Usingserial transplantation studies, long-term reconstitution ofHSCs was shown to be impaired after inhibition of Notchsignalling [20]. In contrast, inactivation of neither JAG1nor Notch1 impaired HSC maintenance in conditionalknockout mouse models [21]. In a study by Maillard etal., Notch signalling was blocked by elimination of CBF1and expression of dominant negative MAML mutants, andcanonical Notch signalling was shown to be dispensable forthe maintenance of long-term (LT) HSCs in vivo [22].

3.3. The Endothelial Stem Cell Niche. Endothelial cells pro-mote HSC expansion and self-renewal in vitro and are shownto have an important role in engraftment of HSCs and recon-stitution of hematopoiesis in vivo [23]. Inhibition of VEGFR-2 signalling in sinusoidal endothelial cells impaired vascularrecovery and hematopoietic reconstitution following irradi-ation in mice [23]. Thus, hematopoietic regeneration aftermyeloablation depends on vascular recovery and endothelialcell function, and Notch has been implicated in cell-cell

4 Bone Marrow Research

interactions between HSCs and endothelial cells that regulateHSC function. In vivo, sinusoidal endothelial cells expressNotch ligands JAG1 and JAG2 [24], and Notch-activatedHSCs have been visualised in close proximity to the bonemarrow vasculature [25]. A recent study used angiogenicmodels to demonstrate that Notch signalling via endothelialcells plays a role in regulation of HSCs in the vascular niche[24]. Increased expression of Notch ligand on endothelialcells after stimulation with soluble kit ligand stimulated theexpansion of repopulating CD34−Flt3−cKit+Lineage−Sca1+

LT-HSCs at the expense of reducing differentiation, andserial transplantation assays demonstrated that these LT-HSCs retained their self-renewal ability. Furthermore, in acoculture model, endothelial cells failed to expand HSCsderived from Notch1-/Notch2-deficient mice.

3.4. Notch as a Part of an Interactive Cell Signalling Network.Notch-initiated signalling is part of an interacting network ofintracellular signalling pathways. The noncanonical activa-tion of Notch signalling represents a crosstalk between Notchsignalling and other intracellular signalling pathways (seeabove). Interactions between Notch and the Wnt pathwayhave been best characterized, but other interactions withvarious pathways have also been described.

(i) Wnt-initiated signalling is mediated through thedownstream β-catenin [26]. The Wnt and Notchpathways seem to act in synergy to maintain the stemcell pool [26, 27]. The crosstalk between these twopathways seems to occur at both the intracellularlevel and between cells in the stem cell niche. Firstly,members of the Wnt pathway regulate the expressionof established Notch target genes, and inhibition ofWnt signalling affects the expression of both Wnt andNotch target genes [20, 28]. Secondly, Wnt signallingcan affect the expression of Notch1 as well as HoxB4[29]. The HoxB4 transcription factor is importantfor HSC self-renewal and expansion by inducingthe expression of genes preferentially expressed byHSCs and downregulating genes associated withmyeloid differentiation [30, 31]. Finally, an exampleof extrinsic crosstalk between these two pathwaysin the stem cell niche is the induced expression ofNotch ligands by activated β-catenin in stromal cellswhich thereby induce-Notch-mediated intracellularsignalling in adjacent HSCs [32].

(ii) Notch signalling becomes a part of a more extensivenetwork through its crosstalk with the Wnt pathwaythat interacts with several other intracellular path-ways [27], including (i) Hedgehog signalling [33],(ii) Prostaglandin E2 signalling; animal experimentssuggest that this crosstalk is dependent on a pro-tein kinase A-dependent mechanism that connectsthe pathways via β-catenin [34], (iii) Transforminggrowth factor-β (TGF-β) and Bone morphogenicprotein (BMP) signaling which targets the commonintracellular mediator Smad4 that directly interactswith members of the Hox transcription factor family[26, 35], and (iv) Angiopoietin-1/Tie2 signalling

which is also important in HSCs [36]; this signallingtargets Cdh2 (N-cadherin) [37] that seems to activateβ-catenin signalling through protein kinase B (Akt-)dependent mechanisms [38].

(iii) Hey2 is a transcription factor that seems to actdownstream of Notch in primitive hematopoieticcells, and studies in zebrafish suggest that its expres-sion is maintained by Hedgehog as well as Vascularendothelial growth factor signalling [39].

(iv) Several members of the NF-κB family (includingp65, p50, RelB, and c-Rel) are under transcriptionalcontrol by Notch-initiated signaling, and decreasedlevels were found in Notch-1 antisense transgenic(Notch-AS-Tg) mice [40]. NF-κB is an importantregulator of the expression of several chemokines,and Notch-initiated signalling may thereby affectchemotaxis and cell trafficking [41, 42].

(v) The Ets transcription factor Er71 seems to be acommon downstream target both for the Wnt, Notchand BMP signalling pathways [43].

These observations clearly illustrate that Notch signallingis part of an extensive network of interacting pathways. Thesepathways are important for normal HSCs, and several ofthem are also important in the development of myeloidmalignancies. Besides Notch signalling pathways (see below),the extensive network of interacting pathways includes theWnt pathway [44, 45], Ang-1/Tie2 [46], HoxB4 [44, 47],Hedgehog signalling [44], BMP [35], NF-κB [48], and TGF-β/Smad4 [35, 49] signaling pathways. Thus, Notch signallingis a part of an extensive network involving several interactingpathways both in normal and leukemic hematopoietic cells.

4. Notch Signalling in the Immune System

4.1. The Role of Notch in T Cell Development. Notchsignalling is directly involved in the regulation of thymicT cell development with Notch1 acting as a key receptorresponsible both for the lineage commitment and inhibitionof other differentiation directions [1]. The DLL4 ligandis expressed by thymic epithelial cells and is essentialfor T lineage commitment [50] (Table 1). The αβ T celldevelopment depends on Notch signalling, and transitionthrough the β-selection checkpoint is then dependent onboth Notch signalling [1] as well as CXCL12 ligation ofCXCR4 with initiation of PI3K signalling [51]. Notch1expression is downregulated after β-selection [1]. In contrast,the γδ T cell development seems less dependent on Notchsignalling.

4.2. Effects of Notch on Peripheral T Cell Subsets. Naive Tcells exit the thymus and migrate to the periphery wherethey mediate immune responses after antigenic recognitiontogether with adequate costimulation. Activated naive CD4+

T helper cells (Th0 cells) can differentiate towards Th1,Th2, Th9, Th17, and Th22 helper cells, or they mayalternatively develop into induced regulatory T (iTreg) cellsthat act together with thymus-derived natural T regulatory

Bone Marrow Research 5

Table 1: Notch ligand-receptor interaction: a summary of possible interactions involved in normal hematopoiesis, T cell development, andregulation of the peripheral T cell system.

SIGNAL-INITIATING CELL SIGNAL-RECEIVING CELL

Cell type Ligand Cell type Receptor References

Bone marrow stromal cells Bone marrow stem cells

Osteoblasts (endosteal niche)JAG1 Sca-1+, c-kit+, Lin− Notch1 [13]

JAG1, JAG2 Sca-1+, c-kit+, CD117+ Notch1, Notch2 [52]

Endothelial cells (vascular niche) JAG1, JAG2, DLL4, DLL1 Sca-1+, c-kit+, Lin− Notch1, Notch2 [24]

Thymic epithelial cells (TECs) T cell progenitors

TECs DLL4 Thymocytes Notch1 [1, 50]

Antigen-presenting cells Peripheral T cells

APC DLL1 Tc Notch1, Notch2 [1]

DC DLL1, DLL4 Th1 Notch3 [1, 4, 53]

DC JAG1, JAG2 Th2 Notch1, Notch2 [1, 4, 53]

APC DLL1, DLL4 Th17 Notch [1, 54]

pDC DLL4 Th1 IL-10+ Notch [55]

Abbreviations: Delta-like (DLL); Jagged (JAG); Thymic epithelial cells (TECs); Antigen presenting cells (APC); Dendritic cells (DC); plasmacytoid DC (pDC);Cytotoxic T cells (Tc); Helper T cells (Th); Interleukin-10 (IL-10).

(nTreg) cells to inhibit immune responses. Activation ofnaive CD8+ T cells leads to the differentiation towardscytotoxic T lymphocytes (Tc, or also called CTLs). A detailedlist of possible interactions between ligand-presenting andreceptor-expressing cells involved in normal hematopoiesis,T cell development, and T cell activation is given in Table 1.

4.2.1. Tc Cells. The transcriptional regulator eomesodermin(Eomes) regulates the expression of perforin and granzyme Bin CD8+ cytotoxic T cells [56]. Notch1 seems to directly reg-ulate expression of Eomes as well as perforin and GranzymeB by binding to their promoters, and γ-secretase inhibitors(GSIs) thereby attenuate in vitro T cell cytotoxicity. Inaddition, Notch2-ICD seems to cooperate with CREB1 in theregulation of granzyme B expression [57].

4.2.2. Th1 and Th2 Cells. Th1 cells produce IFNγ whileTh2 cells produce IL-4, IL-5, and IL-13 as their signaturecytokines. Jagged ligands expressed by APCs are importantfor Th2 differentiation whereas DLL ligands (DLL1 and/orDLL4) seem to promote Th1 and inhibit Th2 differentiation,but the additional molecular events in this differentiationhave not been characterized [1]. Thus, it is not knownwhich Notch receptors mediate the DLL-induced Th1 celldifferentiation signal. Although canonical Notch signallingdoes not seem to be essential [1], Notch3 signalling ispossibly also involved in Th1 differentiation [58]. On theother hand, Th2 cell differentiation seems to involve CBF1and IL-4, as well as the Th2-specific transcription factorGata3 that is a Notch target gene [58].

4.2.3. Th17 Cells. These cells represent a proinflammatorysubset distinct from Th1 cells; their signature cytokines areIL-17A, IL-21, and IL-22, and they express the Th17-specifictranscription factor RORγ. DLL ligands might play a role inthe generation of Th17 cells [1], and recent in vitro studies

[54] suggest that DLL4 inhibits Th2 cytokine production,contributes to Th17 differentiation, and upregulates RORcexpression. Both the RORc and IL-17 gene promoters thenseem to be direct targets for Notch-initiated signalling.

4.2.4. Treg Cells. Both natural (nTreg) and peripherallyinduced Treg (iTreg) cells are important for downregulationof immune responses. FoxP3 is a Treg-specific transcriptionfactor, and the cells typically release IL-10. Notch ligands(usually the Jagged family) increase Treg cell differentiationin vitro [1, 4, 59, 60], but this differentiation is not Notch-dependent because Notch loss-of-function mutant mice donot lack Treg cells [1]. Notch1 signalling seems to contributeto the FoxP3 expression [61], and Notch3 receptors areincreased on murine CD4+CD25+ Treg cells [62]; exposureto JAG2-overexpressing hematopoietic progenitors seems toincrease the expression of Notch3 and FoxP3 in Treg cells[63].

4.3. Notch and Autoimmunity. There are few reports ofNotch in human autoimmune diseases. Recently Jiao et al.[64] reported increased expression of Notch2/3/4 in Thcells from patients with active rheumatoid arthritis, andthe increased expression of Notch3 was mainly detected inactivated T cells. These patients also show increased nucleartranslocation of NICD in Th cells as well as increasedexpression of the Notch target gene HES-1 [64]. Anotherrecent study [65] reported increased expression of Notch1/3and decreased Notch2 together with downregulated DLL1among peripheral blood mononuclear cells from patientswith autoimmune thrombocytopenia. Finally, Sodsai et al.[66] suggest that defective Notch1 upregulation during Tcell activation is important for increased disease activityin patients with systemic lupus erythematosus. However, itshould be emphasized that a major part of these studies onlyincluded a description of Notch/Notch ligand expressionin immunocompetent cells; it is therefore difficult to judge

6 Bone Marrow Research

whether this altered expression is directly involved in dis-ease development/progression or only represents secondary(innocent bystander or secondary) effects that may not beclinically important.

The importance of Notch signalling in autoimmunediseases has been investigated more in detail in murinemodels of autoimmune disorders, and these results aresummarized in Table 2. The contribution of Notch signallinghas especially been investigated in experimental autoim-mune encephalomyelitis (EAE), an experimental model ofmultiple sclerosis. Th1 and Th17 cells are important forthe development of this disease, and inhibitions of Notch3,γ-secretase, or DLL1 inhibit proinflammatory Th1/Th17responses and improve disease symptoms. In contrast, DLL2-initiated signalling increases symptoms whereas JAG1 resultsin improvement. Furthermore, Notch1-induced signallingalso seems important for the development of autoimmunityin other organs, but these effects seem to differ depending onthe experimental model. Finally, Notch-induced signallingwith induction of Treg cells can inhibit the development ofautoimmune diabetes in various disease models. It should beemphasized that several of these conclusions are based on theobserved effects when using specific Notch inhibition in theexperimental models (like specific neutralizing antibodies)as described in detail in Table 2; these effects of specificinhibitors demonstrate that Notch-initiated signalling isdirectly involved in the regulation of experimental murineautoimmunity.

Many clinical and laboratory features of GVHD, espe-cially in its chronic form, resemble those of autoimmunediseases, and the pathophysiological mechanisms also seemto show similarities [72, 73]. Autoimmune phenomena canbe seen after both auto- and allotransplantation, and themost common manifestations seem to be thyroid disease andautoimmune cytopenias. Certain manifestations of GVHDhave been postulated to represent a specific loss of toleranceto self structures. Taken together, these observations suggestthat the role of Notch signalling in autoimmune diseases isrelevant also for development of GVHD.

5. Immune Reconstitution after StemCell Transplantation

Immunological reconstitution after stem cell transplantationhas been extensively reviewed previously [74–78], and themost important observations both in allogeneic and autol-ogous transplantation are (i) early lymphoid reconstitutionis associated with a decreased risk of relapse, suggestingthat antileukemic immune reactivity is mediated early aftertransplantation [79–82] and (ii) a long-lasting quantitativeCD4 T cell defect can persist for several months posttransplant [75, 76, 78].

The quantitative CD4+ T cell defect after allotransplan-tation may last for several months, but early normalizationseems more common after reduced intensity conditioning[78]. Infusion of a high number of CD4+ T cells and NaturalKiller (NK) T cells seems to be associated with a betterprognosis [83], an observation supporting the hypothesis

that early antileukemic immune reactivity is important.Recovery of dendritic cells occurs more slowly [84], and anabnormal ratio between various dendritic cell subsets maypersist for months after transplantation [84]. Low dendriticcell counts one month post transplant also seem to bean independent adverse prognostic factor for the overallsurvival [85].

In autotransplanted patients the immunological recon-stitution differs between patients receiving peripheral bloodand bone marrow autografts [86]. Mobilized stem cells arenow most commonly used, and during the first posttrans-plant months the patients generally show early recoveryof CD8+ T cells, CD14+ monocytes, and CD56+ NK cells[86–88]. Circulating dendritic cells are usually normalizedrelatively early, although differences in dendritic cell subsetcomposition may persist for several months also in thesepatients [88]. The T cell defect is detected after 6 months formost of these patients and may last for more than a year [87],and it is mainly due to reduced naive CD3+CD4+CD45RA+

T cells, but reduced CD8+ naive T cells can also be seen[87, 88].

6. Notch in Hematological Malignancies

6.1. T-Lineage Acute Lymphoblastic Leukemia. ALL is charac-terized by accumulation of immature lymphoblast of eitherB or T cell lineage origin in bone marrow and eventuallyother lymphoid organs. T-ALL accounts for approximatelyone third of all cases [89]. Notch involvement in T-ALL wasfirst described in patients with the rare t(7;9) (q43;q34.3)translocation that leads to the expression of a cytoplasmicform of the Notch1 receptor with constitutive activity [90].However, the most common Notch abnormality in T-ALLis mutations in the Notch1 alleles that result in constitutiveactivation of the pathway; this is seen in more than halfof the patients [91]. These Notch1 mutations are locatedat specific hotspots and affect critical negative regulatoryelements of the protein. The molecular mechanisms bywhich aberrant Notch1 signalling contributes to T-cell trans-formation are not yet fully understood. Oncogenic Notch1probably cooperates with oncogenic transcription factorssuch as c-Myc [92], E2A-PBX [93], and Ikaros [94], butthe aberrant Notch1 signalling is not sufficient for leukemictransformation [95]. Observations in animal models suggestthat even nonmutational Notch1 activation contributes toleukemogenesis [96], probably through activation of c-Mycthat is a direct downstream target of Notch1 [92]. Finally, theprognostic impact of Notch1 mutations was demonstrated inrecent clinical studies where the mutations were associatedwith good prognosis both in children [97] and adults [98].

6.2. Acute Myeloid Leukemia. The prevalence of Notchmutation in AML is probably less than 5% [99, 100], andNotch ligation in AML cells has diverse or only minoreffects [101]. AML cells seem to express JAG1, Notch1 andNotch2 [102–104]. In hematopoiesis, Notch1, drives myeloiddifferentiation through the expression of transcriptionalfactor PU.1. Results obtained from Chen et al. in [104]

Bone Marrow Research 7

Table 2: Therapeutic targeting of Notch in murine autoimmunity in vivo.

Disease and intervention Therapeutic effect References

Experimental autoimmune encephalomyelitis (EAE; model of multiple sclerosis)

γ-secretase inhibitorInhibition of disease-associated Th1 responses and

improvement of symptoms[4, 58, 67]

Notch1 neutralizing antibodies No effect on Th1 and Th17 responses [4, 58]

Notch3 neutralizing antibodiesDecreased Th1 and Th17 responses, inhibition of the ability of

myelin-primed T cells to transfer the disease[4, 58]

DLL1 neutralizing antibodies Reduced Th1 responses and EAE symptoms [4, 68]

Activating DLL1-Fc fusion protein Increased Th1 responses and EAE symptoms [4, 68]

Neutralizing JAG1 antibodies EAE disease progression [4, 68]

Activating JAG1-Fc fusion protein Improvement of EAE symptoms [4, 68]

Experimental hepatitis

γ-secretase inhibitorReduced Notch1 signalling and FoxP3 expression, spontaneoushepatic lymphocyte infiltration consistent with autoimmune

hepatitis (C57BL/6 mice)[61]

Murine diabetes

Lck-Notch3-IC transgenic mice

Up regulation of the generation and function of CD4+CD25+

Treg. The mice failed to develop streptozotocin-inducedautoimmune diabetes. Adoptive transfer of the lck-Notch3-ICtransgenic CD4+ cells to wild-type recipients prevented the

progression of the disease.

[62]

Exposure to JAG2-expressinghematopoietic progenitor cells

Activation of Notch3 signalling with increased Tregproliferation and prevention of diabetes in NOD mice.

[63]

Multiorgan autoimmune disease

Loss of functional mutation in the Itchubiquitin ligase

This ligase is involved in Notch1 degradation; homozygousmice develop an autoimmune-like disease mainly affecting

lungs, skin, and lymphoid organs.[2, 69–71]

Abbreviations: Experimental autoimmune encephalomyelitis (EAE), Delta-like (DLL), Jagged (JAG), Helper T cells (Th).

show that the Notch1 gene and protein expression weredecreased in human AML samples in comparison withnormal hematopoietic stem cells. This decrease of Notch1expression was associated with a concordant downregulationin PU.1, suggestive of impeded PU.1-mediated myeloid sig-nalling and thus contributing to AML leukemogenesis [104].However, gene expression profiling of primary human AMLcells has identified a subgroup of patients with recurringmutations in Notch [105, 106]; the expression profile ofthese patients seems to be mainly determined by silencingof the CEBPA gene through promoter hypermethylation[106]. The CEBPA gene encodes for the transcription factorCCAAT/enhancer-binding protein alpha (C/EBPalfa); thisgene is mutated in approximately 10% of AML cases [107].AML cells with silenced CEBPA gene and Notch mutationscluster together [106].

6.3. Chronic Lymphocytic Leukemia. Chronic lymphocyticleukemia (CLL) is characterized by detection of malignantCD5+CD19+ B cells in blood, bone marrow, and eventuallyother lymphoid organs. Several recent studies suggest thatthe Notch system is important also in B-CLL. Firstly,B-CLL cells express high levels of Notch2 that regulatethe expression of antiapoptotic CD23a [108]. Secondly,Notch1/2 and the ligands JAG1/2 are also constitutivelyexpressed in B-CLL [109] and are then associated with

resistance to apoptosis [109]. Upregulation of Notch1 isobserved during treatment with the MDM2/p53 inhibitorNutlin-3 and possibly represents a feedback mechanisminvolved in restrain of the Nutlin-3 effects [110].

7. Possible Strategies forNotch Targeting in Patients Treated withStem Cell Transplantation

7.1. Therapeutic Tools for Inhibition of Canonical NotchSignalling. An overview of possible therapeutic tools isgiven in Table 3. The tools include ligands, agonistic andantagonistic antibodies, stimulatory fusion proteins, andinhibitors of intracellular signalling. Inhibition of the γ-secretase activity with general downregulation of Notchsignalling has been used in experimental in vitro studies[111]. Unless NICD is translocated to the nucleus, the NICDform of Notch is ubiquitinated and thereafter degradedby the proteasomes; and proteasomal inhibitors may thusenhance Notch signalling. Various proteasomal inhibitors arenow used in the treatment of hematologic malignancies, andthey are also tried as immunosuppressive agents [112–115],but it is not known whether inhibition of Notch signallingcontributes to their clinical effects.

The results summarized in Table 3 suggest that if adetailed characterization of the immune system is available,

8 Bone Marrow Research

Table 3: Potential molecular tools for targeting of Notch signalling.

Molecular tool Observations in clinical or experimental studies References

Natural receptor ligands

JAG2-expressing hematopoieticprogenitor cells

Activation of Notch3 signalling with increased Tregproliferation and prevention of diabetes in NOD mice.

[63]

Fusion proteins of natural ligands andFc-Ig

DLL1-IgG-Fc has been used for expansion of human umbilicalcord stem cell expansion; cells caused no unexpected toxicityand contributed to long-term engraftment. Could be used for

transplantation.

[116]

JAG1- and DLL2-Fc fusion proteinsBoth types of fusion proteins have shown immunomodulatory

effects in experimental murine disease models.[4, 68]

Receptor- or ligand-directed antibodies

Agonistic antibodiesAgonistic antibodies have been identified for their reactivity

against Notch2 or Notch3.[117, 118]

Antagonistic antibodies directed againstNotch

Several antagonistic Notch1-, Notch2- or Notch3-directedantibodies have been tested in vitro and in vivo.

[4, 58, 117,119, 120]

Antagonistic antibodies directed againstNotch ligands

Both DLL- and JAG1- specific antibodies showimmunomodulatory effects in murine disease models.

[4, 68]

DLL1 neutralizing antibodies Reduced Th1 responses and improvement of EAE symptoms. [4, 68]

DLL4 neutralizing antibodiesIn CSC-driven colon and breast xenograft models, anti-human

DLL4-blocking antibodies inhibited tumour growth andreduced tumour-initiating cell frequencies.

[121]

Antibodies that inhibit thetranscription-regulating complex

Development of such antibodies could inhibit parts of theNotch-initiating effects and possibly limit the toxicity.

[122]

Activating JAG1-Fc fusion protein Improvement of EAE symptoms. [4, 68]

Activating DLL2-Fc fusion protein Increased Th1 responses and progression of EAE symptoms. [4, 68]

Inhibition of intracellular signalling

γ-secretase inhibitorReduced Notch1 signalling and FoxP3 expression in Treg cells in

murine disease models.[58, 61, 67]

Proteasome inhibitors Inhibition of noncanonical Notch signalling.[2, 10, 123,124]

Inhibition of the PI3K-Akt-mTORpathway

Inhibition of noncanonical Notch signalling. [2, 10, 48]

Peptide that inhibits assembly of thetranscription-regulating complex

The small hydrocarbond-staple peptide SAHM1 inhibits Notchsignalling in vitro.

[122]

Abbreviations: Experimental autoimmune encephalomyelitis (EAE), Delta-like (DLL), Jagged (JAG), Helper T cells (Th), Cancer stem cell (CSC).

it would be possible to design therapeutic strategies basedon stimulation or inhibition of selected Notch-mediatedeffects. Especially in allograft recipients, Notch inhibitionmay offer the opportunity to combine immunosuppressiveGVHD prophylaxis with direct antileukemic effects.

7.2. Notch-Driven Stem Cell Expansion Effects on Posttrans-plant Myeloid Reconstitution. The delayed hematopoieticstem cell engraftment commonly seen after cord bloodtransplantation is probably due to an inadequate numbersof progenitor cells in the graft [116]. In a recent phase Iclinical trial, CD34+ cord blood cells were cultured ex vivowith the extracellular DLL1 domain fused to the Fc domainof human IgG. After 16 days and following myeloablativeconditioning, the expanded cells were transplanted togetherwith an unmanipulated allograft. No unexpected toxicity wasobserved, and the patients showed early myeloid reconstitu-tion with a shortened time until peripheral blood neutrophilcounts ≥0.5 × 109/L (16 days compared with 26 days

for the controls). In four of these patients, the neutrophilreconstitution was attained at a time when at least 80% of thecells were derived from the manipulated grafts, and for twopatients, long-term persistence of these cells was documentedafter 180 and 240 days. For other patients, persisting cellswere derived from the unmanipulated graft. Thus, targetingof Notch can be used for ex vivo stem cell expansion ofallogeneic stem cells. Whether a similar methodologicalapproach can be used for ex vivo expansion of autologousstem cells in poor mobilizers has not been investigated, butthe use of a CXCR4 antagonist would at present be thefirst alternative for such patients [125]. Another possibletherapeutic strategy may be in vivo expansion of stemand progenitor cells by administration of the Delta1-IgGpreparation.

7.3. Notch Targeting and Posttransplant Lymphoid Reconsti-tution. Early lymphoid reconstitution after stem cell trans-plantation is associated with decreased relapse risk in several

Bone Marrow Research 9

hematological malignancies [126]. The study by Delaneyet al. [116] demonstrated that ex vivo stem cell expansionreduced the time until neutrophil reconstitution, but themanipulated grafts did not contain T cells, and long-termT cell engraftment was always derived from the unmanip-ulated grafts (see above). Notch-induced signalling is alsoimportant for T lymphopoiesis, but it is not known whetheralternative ex vivo or in vivo strategies for Notch targetingcan be used to increase lymphopoiesis and thereby shortenthe time until lymphoid reconstitution and thereby shortenthe posttransplant CD4 defect. Such a strategy may becomeuseful in autotransplanted patients to increase posttransplantantileukemic T cell reactivity, whereas it would be moredifficult to use in allotransplanted patients with the risk ofsevere and potentially lethal GVHD.

7.4. Notch Targeting of Malignant Hematopoietic Cells—The Initial Clinical Experience. Allogeneic and autologousstem cell transplantation is mainly used in the treatmentof hematologic malignancies, and Notch signalling seemsimportant in these diseases, especially T-ALL (see above).Inhibitors of the γ-secretase activity have been developed,but the initial clinical Phase I studies in T-ALL patientsshowed a low efficiency and severe gastrointestinal toxicity[127]. Several other Notch-targeting drugs are now beingdeveloped for use in clinical phase 1-2 trials, and one ofthem has also been investigated in a phase 3 trial [128]. Theseagents are mainly γ-secretase inhibitors that are tried in thetreatment of various cancers. Whether these inhibitors willhave an acceptable toxicity and higher efficiency has to beaddressed in future studies.

7.5. Targeting Noncanonical Notch Signalling—Possible Mech-anisms for the Antileukemic Effects of Several Targeted Ther-apies. Several drugs may affect the expression of Notch-targeted genes through inhibition of the noncanonicalpathways such as HSP90, HDAC, PI3K/Akt/mTOR, andproteasomal inhibitors [2, 10]. Some of these drugs mayhave combined effects; for example, proteasomal inhibitorsmay alter noncanonical signalling through NFκB inhibitiontogether with decreased degradation of the NICD forminvolved in canonical signalling. Furthermore, specific PI3Kinhibitors are now evaluated in clinical studies [129]. ThePI3K-Akt pathway is upstream to mTOR, and the mTORinhibitor rapamycin is used for immunosuppression afterallotransplantation and is also being investigated as ananticancer agent in hematologic malignancies. Thus, Notchinhibition may contribute to the efficiency of several newanticancer and/or immunosuppressive-targeted therapeu-tics.

7.6. Notch Targeting in Stem Cell Recipients: Immunosuppres-sion versus Immunostimulation. As can be seen from Table 2,Notch targeting can be used both for immunostimulationand immunosuppression in experimental autoimmunity,and these observations may be relevant also for humanGVHD [72, 73]. Thus, Notch signalling seems importantboth for T lymphopoiesis and for regulation of the peripheral

T cell system. Notch agonists may thus become useful toenhance T cell reconstitution after both allogeneic andautologous stem cell transplantation. Early lymphoid recon-stitution is then associated with decreased risk of cancerrelapse, and earlier T cell reconstitution would possiblyfurther reduce the relapse risk. T cell defects are in additionassociated with an increase of severe opportunistic infectionsespecially in allotransplant recipients [130, 131], and earlyreconstitution may also reduce this risk. On the other hand,enhancement of T cell reconstitution after allotransplanta-tion has to be balanced against a possible risk of inducingsevere and potentially lethal GVHD.

A second possibility could be to use Notch-targetingtherapy to modulate the function of peripheral T cells.Immunostimulatory agonists could then be used to enhanceantileukemic immune reactivity after autologous stem celltransplantation. Clinical studies have demonstrated thatantileukemic T cells can be detected in autotransplantedleukemia patients and that this reactivity can possibly beenhanced by vaccination therapy [126]. Immunostimula-tory Notch targeting may then increase this antileukemicreactivity and possibly increase the efficiency of anticancervaccines.

A third strategy is to consider immunomodulatorystrategies to reduce the risk of severe GVHD after allotrans-plantation. Yvon et al. [132] overexpressed the JAG1 ligandin alloantigen-presenting B cells and observed induction ofTreg cells from CD45-RA+ T cells; these allospecific Treg cellscaused a specific inhibition of proliferative and cytotoxic Tcell responses against the priming alloantigens. Thus, Notchagonists may be used to induce specific tolerance againstalloantigens, and both ex vivo generation of immunoregu-latory cells and in vivo administration of agonists should beconsidered. An alternative would be to use Notch inhibitionfor suppression of effector T cells (see Table 2). However,the use of Notch inhibition in targeting the peripheral Tcell system seems less attractive because this approach mayinterfere with lymphoid reconstitution and aggravates theposttransplant T cell defects (see above).

7.7. Mesenchymal Stromal Cells. Multipotent mesenchymalstromal cells (MSCs), also called mesenchymal stem cells,are able to differentiate into a variety of cell types includ-ing osteoblasts, chondrocytes, and adipocytes [133]. Thesecells are important components of the bone marrow HSCniche and can support HSC maintenance and engraftment[134]. Intriguingly, MSCs have also been shown to haveimmunomodulatory properties which are of value in aclinical setting with regard to treatment of GVHD afterhematopoietic stem cell transplantation (HSCT). More-over, cotransplantation of HSCs and MSCs can facilitatehematopoietic engraftment and was shown to acceleratelymphocyte recovery in clinical HSCTs [135]. The exactmechanisms of how MSCs contribute to hematopoieticreconstitution remain unclear though both immunomod-ulatory effects as well as effects on HSC self-renewalcapacity are assumed and Notch signalling has been impli-cated in these effects. Studies of Notch function have

10 Bone Marrow Research

revealed that Notch signalling affects various differentiationcapabilities of MSC, including differentiation in directionof osteoblasts [17, 136, 137]. Notch signalling in bonemarrow is suggested to maintain a pool of mesenchymalprogenitors by suppressing osteoblast differentiation [16]. Inaddition, Notch signalling has been identified as a possiblepathway involved in osteogenic differentiation of MSCsinduced by soluble mediators derived from endothelial cells[138].

The infusion of MSCs has been tried in the treat-ment of GVHD. There seems to be a consensus thatthese cells are immunomodulatory, but the initial clini-cal studies have shown conflicting results with regard tothe efficiency of MSC in the treatment of GVHD [139–141]. Human bone marrow-derived MSCs were found toexpress high levels of functionally active toll-like receptors(TLR) 3 and 4, and these cells had an immunosuppressiveeffect on T-cell proliferation after ligation of either TLR3or TLR4. Suppression of T-cell activation was inhibitedby neutralization of JAG1 and inhibition of γ-secretaseactivity, thus implying a role of impaired Notch receptorsignalling in T cells [142]. Additional mechanisms possiblyinvolved in MSC-induced immunomodulation could beinteractions with the NK cell system, inhibition of dendriticcell differentiation, or modulation of the humoral system[139].

To conclude, even though additional studies are definitelyneeded, these studies suggest that Notch signalling may beimportant both for the development of supportive cells instem cell niches and for the immunomodulatory/GVHD-suppressing effect of the MSC.

7.8. Concluding Remarks. The Notch ligand/receptor systemis important for (i) development and regulation of the Tcell system and (ii) regulation of normal as well as leukemichematopoiesis. The final biological effects of Notch targetingin stem cell recipients are difficult to predict, and dependboth on the involved ligand(s) and receptors, and signallingthrough the canonical intracellular pathway is modulated bynoncanonical signalling. The interactions between Notch-initiated signalling and several other intracellular signallingpathways further make it difficult to predict the finaleffect of Notch-targeted therapy. Pharmacological toolsfor targeting of Notch-mediated signalling are now beingdeveloped. However, because the effects of Notch-targetedtherapy are difficult to predict a more detailed study ofthe post-transplant hematopoiesis as well as the T cellsystem is necessary before clinical studies of these agentsin transplant recipients can be designed. However, suchstudies should be encouraged because Notch targeting mayrepresent a unique possibility to combine enhancement ofreconstitution, immunomodulation, and direct anticancertreatment.

Acknowledgment

The scientific work of the authors is supported by theNorwegian Cancer Society and Helse-Vest.

References

[1] F. Radtke, N. Fasnacht, and H. R. MacDonald, “Notchsignaling in the immune system,” Immunity, vol. 32, no. 1,pp. 14–27, 2010.

[2] C. Talora, A. F. Campese, D. Bellavia et al., “Notch signalingand diseases: an evolutionary journey from a simple begin-ning to complex outcomes,” Biochimica et Biophysica Acta,vol. 1782, no. 9, pp. 489–497, 2008.

[3] A. R. Sandy and I. Maillard, “Notch signaling in thehematopoietic system,” Expert Opinion on Biological Therapy,vol. 9, no. 11, pp. 1383–1398, 2009.

[5] D. Sprinzak, A. Lakhanpal, L. Lebon et al., “Cis-interactionsbetween Notch and Delta generate mutually exclusive sig-nalling states,” Nature, vol. 465, no. 7294, pp. 86–90, 2010.

[6] N. Haines and K. D. Irvine, “Glycosylation regulates Notchsignalling,” Nature Reviews Molecular Cell Biology, vol. 4, no.10, pp. 786–797, 2003.

[7] K. Shimizu, S. Chiba, N. Hosoya et al., “Binding ofDelta1, Jagged1, and Jagged2 to Notch2 rapidly inducescleavage, nuclear translocation, and hyperphosphorylationof Notch2,” Molecular and Cellular Biology, vol. 20, no. 18,pp. 6913–6922, 2000.

[8] K. Shimizu, S. Chiba, T. Saito, K. Kumano, and H. Hirai,“Physical interaction of Delta1, Jagged1, and Jagged2 withNotch1 and Notch3 receptors,” Biochemical and BiophysicalResearch Communications, vol. 276, no. 1, pp. 385–389, 2000.

[4] D. Amsen, A. Antov, and R. A. Flavell, “The different facesof Notch in T-helper-cell differentiation,” Nature ReviewsImmunology, vol. 9, no. 2, pp. 116–124, 2009.

[9] I. Van de Walle, M. De Smet, E. Waegemans et al., “TheSerrate-like ligand Jagged2 acts as a Delta-like Notch ligandduring hematopoietic cell fate decisions,” Submitted.

[10] R. Sanalkumar, S. B. Dhanesh, and J. James, “Non-canonicalactivation of Notch signaling/target genes in vertebrates,”Cellular and Molecular Life Sciences, vol. 67, no. 17, pp. 2957–2968, 2010.

[11] A. Robert-Moreno, L. Espinosa, J. L. de la Pompa, and A.Bigas, “RBPjκ-dependent Notch function regulates Gata2and is essential for the formation of intra-embryonichematopoietic cells,” Development, vol. 132, no. 5, pp. 1117–1126, 2005.

[12] M. J. Kiel, O. H. Yilmaz, T. Iwashita, O. H. Yilmaz,C. Terhorst, and S. J. Morrison, “SLAM family receptorsdistinguish hematopoietic stem and progenitor cells andreveal endothelial niches for stem cells,” Cell, vol. 121, no. 7,pp. 1109–1121, 2005.

[13] L. M. Calvi, G. B. Adams, K. W. Weibrecht et al., “Osteoblasticcells regulate the haematopoietic stem cell niche,” Nature, vol.425, no. 6960, pp. 841–846, 2003.

[14] D. Visnjic, Z. Kalajzic, D. W. Rowe, V. Katavic, J. Lorenzo, andH. L. Aguila, “Hematopoiesis is severely altered in mice withan induced osteoblast deficiency,” Blood, vol. 103, no. 9, pp.3258–3264, 2004.

[15] H.-G. Kopp, S. T. Avecilla, A. T. Hooper, and S. Rafii, “Thebone marrow vascular niche: home of HSC differentiationand mobilization,” Physiology, vol. 20, no. 5, pp. 349–356,2005.

[16] M. J. Hilton, X. Tu, X. Wu et al., “Notch signaling maintainsbone marrow mesenchymal progenitors by suppressingosteoblast differentiation,” Nature Medicine, vol. 14, no. 3, pp.306–314, 2008.

Bone Marrow Research 11

[17] F. Engin, Z. Yao, T. Yang et al., “Dimorphic effects of Notchsignaling in bone homeostasis,” Nature Medicine, vol. 14, no.3, pp. 299–305, 2008.

[18] F. N. Karanu, B. Murdoch, L. Gallacher et al., “The Notchligand Jagged-1 represents a novel growth factor of humanhematopoietic stem cells,” Journal of Experimental Medicine,vol. 192, no. 9, pp. 1365–1372, 2000.

[19] J. Zhang, C. Niu, L. Ye et al., “Identification of thehaematopoietic stem cell niche and control of the niche size,”Nature, vol. 425, no. 6960, pp. 836–841, 2003.

[20] A. W. Duncan, F. M. Rattis, L. N. DiMascio et al., “Integrationof Notch and Wnt signaling in hematopoietic stem cellmaintenance,” Nature Immunology, vol. 6, no. 3, pp. 314–322,2005.

[21] S. J. C. Mancini, N. Mantei, A. Dumortier, U. Suter, H.R. MacDonald, and F. Radtke, “Jagged1-dependent Notchsignaling is dispensable for hematopoietic stem cell self-renewal and differentiation,” Blood, vol. 105, no. 6, pp. 2340–2342, 2005.

[22] I. Maillard, U. Koch, A. Dumortier et al., “Canonicalnotch signaling is dispensable for the maintenance of adulthematopoietic stem cells,” Cell Stem Cell, vol. 2, no. 4, pp.356–366, 2008.

[23] A. T. Hooper, J. M. Butler, D. J. Nolan et al., “Engraft-ment and reconstitution of hematopoiesis is dependent onVEGFR2-mediated regeneration of sinusoidal endothelialcells,” Cell Stem Cell, vol. 4, no. 3, pp. 263–274, 2009.

[24] J. M. Butler, D. J. Nolan, E. L. Vertes et al., “Endothelial cellsare essential for the self-renewal and repopulation of Notch-dependent hematopoietic stem cells,” Cell Stem Cell, vol. 6,no. 3, pp. 251–264, 2010.

[25] J. M. Butler, H. Kobayashi, and S. Rafii, “Instructive role ofthe vascular niche in promoting tumour growth and tissuerepair by angiocrine factors,” Nature Reviews Cancer, vol. 10,no. 2, pp. 138–146, 2010.

[26] U. Blank, G. Karlsson, and S. Karlsson, “Signaling pathwaysgoverning stem-cell fate,” Blood, vol. 111, no. 2, pp. 492–503,2008.

[27] F. J. T. Staal and T. C. Luis, “Wnt signaling in hematopoiesis:crucial factors for self-renewal, proliferation, and cell fatedecisions,” Journal of Cellular Biochemistry, vol. 109, no. 5,pp. 844–849, 2010.

[28] J. J. Trowbridge, A. Xenocostas, R. T. Moon, and M. Bhatia,“Glycogen synthase kinase-3 is an in vivo regulator ofhematopoietic stem cell repopulation,” Nature Medicine, vol.12, no. 1, pp. 89–98, 2006.

[29] T. Reya, A. W. Duncan, L. Ailles et al., “A role for Wntsignalling in self-renewal of haematopoietic stem cells,”Nature, vol. 423, no. 6938, pp. 409–414, 2003.

[30] H. M. Lee, H. Zhang, V Schulz, D. P. Tuck, and B. G. Forget,“Downstream targets of HOXB4 in a cell line model ofprimitive hematopoietic progenitor cells,” Blood, vol. 116, pp.720–730, 2010.

[31] J. Jiang, H. Yu, Y. Shou et al., “Hemgn is a direct transcrip-tional target of HOXB4 and induces expansion of murinemyeloid progenitor cells,” Blood, vol. 116, pp. 711–719, 2010.

[32] T. Yamane, T. Kunisada, H. Tsukamoto et al., “Wnt signalingregulates hemopoiesis through stromal cells,” Journal ofImmunology, vol. 167, no. 2, pp. 765–772, 2001.

[33] C. Cerdan and M. Bhatia, “Novel roles for Notch, Wnt andHedgehog in hematopoesis derived from human pluripotentstem cells,” International Journal of Developmental Biology,vol. 54, no. 6-7, pp. 955–964, 2010.

[34] W. Goessling, T. E. North, S. Loewer et al., “Genetic inter-action of PGE2 and Wnt signaling regulates developmentalspecification of stem cells and regeneration,” Cell, vol. 136,no. 6, pp. 1136–1147, 2009.

[35] N. Wang, H.-G. Kim, C. V. Cotta et al., “TGFβ/BMP inhibitsthe bone marrow transformation capability of Hoxa9 byrepressing its DNA-binding ability,” EMBO Journal, vol. 25,no. 7, pp. 1469–1480, 2006.

[36] F. Arai, A. Hirao, M. Ohmura et al., “Tie2/angiopoietin-1signaling regulates hematopoietic stem cell quiescence in thebone marrow niche,” Cell, vol. 118, no. 2, pp. 149–161, 2004.

[37] A. Wilson, M. J. Murphy, T. Oskarsson et al., “c-Myc controlsthe balance between hematopoietic stem cell self-renewal anddifferentiation,” Genes and Development, vol. 18, no. 22, pp.2747–2763, 2004.

[38] J. Zhang, G. J. Woodhead, S. K. Swaminathan et al., “Corticalneural precursors inhibit their own differentiation via N-cadherin maintenance of beta-catenin signaling,” Develop-mental cell, vol. 18, no. 3, pp. 472–479, 2010.

[39] J. M. Rowlinson and M. Gering, “Hey2 acts upstream ofNotch in hematopoietic stem cell specification in zebrafishembryos,” Blood, vol. 116, no. 12, pp. 2046–2056, 2010.

[40] P. Cheng, A. Zlobin, V. Volgina et al., “Notch-1 regulatesNF-κB activity in hemopoietic progenitor cells,” Journal ofImmunology, vol. 167, no. 8, pp. 4458–4467, 2001.

[41] O. Bruserud and A. O. Kittang, “The Chemokine System inExperimental and Clinical Hematology,” Current Topics inMicrobiology and Immunology, vol. 870, pp. 3–12, 2010.

[42] ∅. Bruserud, A. Ryningen, A. M. Olsnes et al., “Subclassifi-cation of patients with acute myelogenous leukemia basedon chemokine responsiveness and constitutive chemokinerelease by their leukemic cells,” Haematologica, vol. 92, no.3, pp. 332–341, 2007.

[43] D. Lee, C. Park, H. Lee et al., “ER71 acts downstream of BMP,Notch, and Wnt signaling in blood and vessel progenitorspecification,” Cell Stem Cell, vol. 2, no. 5, pp. 497–507, 2008.

[44] A. Sengupta, D. Banerjee, S. Chandra et al., “Deregulationand cross talk among Sonic hedgehog, Wnt, Hox andNotch signaling in chronic myeloid leukemia progression,”Leukemia, vol. 21, no. 5, pp. 949–955, 2007.

[45] C. Muller-Tidow, B. Steffen, T. Cauvet et al., “Translocationproducts in acute myeloid leukemia activate the Wnt signal-ing pathway in hematopoietic cells,” Molecular and CellularBiology, vol. 24, no. 7, pp. 2890–2904, 2004.

[46] H. Reikvam, K. J. Hatfield, P. Lassalle, A. Olsnes Kittang,E. Ersvær, and ∅. Bruserud, “Targeting the angiopoietin(Ang)/Tie-2 pathway in the crosstalk between acute myeloidleukaemia and endothelial cells: studies of Tie-2 blockingantibodies, exogenous Ang-2 and inhibition of constitutiveagonistic Ang-1 release,” Expert Opinion on InvestigationalDrugs, vol. 19, no. 2, pp. 169–183, 2010.

[47] M. S. Thakar, X.-B. Zhang, B. C. Beard et al., “Transmis-sion and expansion of HOXB4-induced leukemia in twoimmunosuppressed dogs: implications for a new canineleukemia model,” Experimental Hematology, vol. 37, no. 10,pp. 1157–1166, 2009.

[48] H. Reikvam, A. M. Olsnes, B. T. Gjertsen, E. Ersvar, andO. Bruserud, “Nuclear factor-B signaling: a contributor inleukemogenesis and a target for pharmacological interven-tion in human acute myelogenous leukemia,” Critical Reviewsin Oncogenesis, vol. 15, no. 1-2, pp. 1–36, 2009.

12 Bone Marrow Research

[49] ∅. Bruserud, “Acute myelogenous leukemia blasts as acces-sory cells during T lymphocyte activation: possible implica-tions for future therapeutic strategies,” Leukemia, vol. 13, no.8, pp. 1175–1187, 1999.

[50] U. Koch, E. Fiorini, R. Benedito et al., “Delta-like 4 is theessential, nonredundant ligand for Notchl during thymic Tcell lineage commitment,” Journal of Experimental Medicine,vol. 205, no. 11, pp. 2515–2523, 2008.

[51] M. L. Janas, G. Varano, K. Gudmundsson, M. Noda, T.Nagasawa, and M. Turner, “Thymic development beyond β-selection requires phosphatidylinositol 3-kinase activation byCXCR4,” Journal of Experimental Medicine, vol. 207, no. 1,pp. 247–261, 2010.

[52] B. R. Chitteti, Y.-H. Cheng, B. Poteat et al., “Impact ofinteractions of cellular components of the bone marrowmicroenvironment on hematopoietic stem and progenitorcell function,” Blood, vol. 115, no. 16, pp. 3239–3248, 2010.

[53] D. Amsen, J. M. Blander, G. R. Lee, K. Tanigaki, T. Honjo, andR. A. Flavell, “Instruction of distinct CD4 T helper cell fatesby different notch ligands on antigen-presenting cells,” Cell,vol. 117, no. 4, pp. 515–526, 2004.

[54] S. Mukherjee, M. A. Schaller, R. Neupane, S. L. Kunkel,and N. W. Lukacs, “Regulation of T cell activation byNotch ligand, DLL4, promotes IL-17 production and Rorcactivation,” Journal of Immunology, vol. 182, no. 12, pp. 7381–7388, 2009.

[55] N. Kassner, M. Krueger, H. Yagita et al., “Cutting edge:plasmacytoid dendritic cells induce IL-10 production in Tcells via the Delta-like-4/Notch axis,” Journal of Immunology,vol. 184, no. 2, pp. 550–554, 2010.

[56] O. H. Cho, H. M. Shin, L. Miele et al., “Notch regulatescytolytic effector function in CD8+ T cells,” Journal ofImmunology, vol. 182, no. 6, pp. 3380–3389, 2009.

[57] Y. Maekawa, Y. Minato, C. Ishifune et al., “Notch2 integratessignaling by the transcription factors RBP-J and CREB1 topromote T cell cytotoxicity,” Nature Immunology, vol. 9, no.10, pp. 1140–1147, 2008.

[58] M. Jurynczyk, A. Jurewicz, C. S. Raine, and K. Selmaj,“Notch3 inhibition in myelin-reactive T cells down-regulatesprotein kinase Cθ and attenuates experimental autoimmuneencephalomyelitis,” Journal of Immunology, vol. 180, no. 4,pp. 2634–2640, 2008.

[59] G. F. Hoyne, I. Le Roux, M. Corsin-Jimenez et al., “Serrate1-induced Notch signalling regulates the decision betweenimmunity and tolerance made by peripheral CD4+ T cells,”International Immunology, vol. 12, no. 2, pp. 177–185, 2000.

[60] S. Vigouroux, E. Yvon, H.-J. Wagner et al., “Induction ofantigen-specific regulatory T cells following overexpressionof a Notch ligand by human B lymphocytes,” Journal ofVirology, vol. 77, no. 20, pp. 10872–10880, 2003.

[61] J. B. Samon, A. Champhekar, L. M. Minter et al., “Notchiand TGF21 cooperatively regulate Foxp3 expression and themaintenance of peripheral regulatory T cells,” Blood, vol. 112,no. 5, pp. 1813–1821, 2008.

[62] E. Anastasi, A. F. Campese, D. Bellavia et al., “Expressionof activated Notch3 in transgenic mice enhances generationof T regulatory cells and protects against experimentalautoimmune diabetes,” Journal of Immunology, vol. 171, no.9, pp. 4504–4511, 2003.

[63] H. Kared, H. Adle-Biassette, E. Foıs et al., “Jagged2-expressing hematopoietic progenitors promote regulatory T

cell expansion in the periphery through notch signaling,”Immunity, vol. 25, no. 5, pp. 823–834, 2006.

[64] Z. Jiao, W. Wang, M. Guo et al., “Expression analysis ofNotch-related molecules in peripheral blood T helper cellsof patients with rheumatoid arthritis,” Scandinavian Journalof Rheumatology, vol. 39, no. 1, pp. 26–32, 2010.

[65] D. Ma, J. Dai, X. Zhu et al., “Aberrant expression of Notchsignaling molecules in patients with immune thrombocy-topenic purpura,” Annals of Hematology, vol. 89, no. 2, pp.155–161, 2010.

[66] P. Sodsai, N. Hirankarn, Y. Avihingsanon, and T. Palaga,“Defects in Notch1 upregulation upon activation of T cellsfrom patients with systemic lupus erythematosus are relatedto lupus disease activity,” Lupus, vol. 17, no. 7, pp. 645–653,2008.

[67] L. M. Minter, D. M. Turley, P. Das et al., “Inhibitors ofγ-secretase block in vivo and in vitro T helper type 1polarization by preventing Notch upregulation of Tbx21,”Nature Immunology, vol. 6, no. 7, pp. 680–688, 2005.

[68] W. Elyaman, E. M. Bradshaw, Y. Wang et al., “Jagged1 anddelta1 differentially regulate the outcome of experimentalautoimmune encephalomyelitis,” Journal of Immunology, vol.179, no. 9, pp. 5990–5998, 2007.

[69] M. A. McGill and C. J. McGlade, “Mammalian Numbproteins promote Notch1 receptor ubiquitination and degra-dation of the Notch1 intracellular domain,” Journal ofBiological Chemistry, vol. 278, no. 25, pp. 23196–23203, 2003.

[70] C. M. Hustad, W. L. Perry, L. D. Siracusa et al., “Moleculargenetic characterization of six recessive viable alleles of themouse agouti locus,” Genetics, vol. 140, no. 1, pp. 255–265,1995.

[71] L. E. Matesic, D. C. Haines, N. G. Copeland, and N. A.Jenkins, “Itch genetically interacts with Notch1 in a mouseautoimmune disease model,” Human Molecular Genetics, vol.15, no. 24, pp. 3485–3497, 2006.

[72] A. Tyndall and F. Dazzi, “Chronic GVHD as an autoimmunedisease,” Best Practice and Research: Clinical Haematology,vol. 21, no. 2, pp. 281–289, 2008.

[73] T. Daikeler and A. Tyndall, “Autoimmunity followinghaematopoietic stem-cell transplantation,” Best Practice andResearch: Clinical Haematology, vol. 20, no. 2, pp. 349–360,2007.

[74] J. Storek, M. Geddes, F. Khan et al., “Reconstitution of theimmune system after hematopoietic stem cell transplantationin humans,” Seminars in Immunopathology, vol. 30, no. 4, pp.425–437, 2008.

[75] J. Storek, “Immunological reconstitution after hematopoieticcell transplantation—its relation to the contents of the graft,”Expert Opinion on Biological Therapy, vol. 8, no. 5, pp. 583–597, 2008.

[76] T. J. Fry and C. L. Mackall, “Immune reconstitution followinghematopoietic progenitor cell transplantation: challenges forthe future,” Bone Marrow Transplantation, vol. 35, no. 1, pp.S53–S57, 2005.

[77] O. Bruserud and O. Wendelboe, “Biological treatment inacute myelogenous leukaemia: how should T-cell targetingimmunotherapy be combined with intensive chemother-apy?” Expert Opinion on Biological Therapy, vol. 1, no. 6, pp.1005–1016, 2001.

[78] M. Jimenez, G. Ercilla, and C. Martınez, “Immune recon-stitution after allogeneic stem cell transplantation withreduced-intensity conditioning regimens,” Leukemia, vol. 21,no. 8, pp. 1628–1637, 2007.

Bone Marrow Research 13

[79] L. F. Porrata, M. A. Gertz, D. J. Inwards et al., “Early lym-phocyte recovery predicts superior survival after autologoushematopoietic stem cell transplantation in multiple myelomaor non-Hodgkin lymphoma,” Blood, vol. 98, no. 3, pp. 579–585, 2001.

[80] L. F. Porrata and S. N. Markovic, “Timely reconstitutionof immune competence affects clinical outcome followingautologous stem cell transplantation,” Clinical and Experi-mental Medicine, vol. 4, no. 2, pp. 78–85, 2004.

[81] R. Parkman, G. Cohen, S. L. Carter et al., “Successful immunereconstitution decreases leukemic relapse and improves sur-vival in recipients of unrelated cord blood transplantation,”Biology of Blood and Marrow Transplantation, vol. 12, no. 9,pp. 919–927, 2006.

[82] D. H. Kim, S. K. Sohn, D. I. Won, N. Y. Lee, J. S. Suh, andK. B. Lee, “Rapid helper T-cell recovery above 200 × 106/lat 3 months correlates to successful transplant outcomesafter allogeneic stem cell transplantation,” Bone MarrowTransplantation, vol. 37, no. 12, pp. 1119–1128, 2006.

[83] J. Vela-Ojeda, M. A. Garcıa-Ruiz Esparza, E. Reyes-Maldonado et al., “Clinical relevance of NK, NKT, anddendritic cell dose in patients receiving G-CSF-mobilizedperipheral blood allogeneic stem cell transplantation,” Annalsof Hematology, vol. 85, no. 2, pp. 113–120, 2006.

[84] M. D. Porta, G. M. Rigolin, E. P. Alessandrino et al., “Den-dritic cell recovery after allogeneic stem-cell transplantationin acute leukemia: correlations with clinical and transplantcharacteristics,” European Journal of Haematology, vol. 72, no.1, pp. 18–25, 2004.

[85] C. Talarn, A. Urbano-Ispizua, R. Martino et al., “Kineticsof recovery of dendritic cell subsets after reduced-intensityconditioning allogeneic stem cell transplantation and clinicaloutcome,” Haematologica, vol. 92, no. 12, pp. 1655–1663,2007.

[86] J. E. Talmadge, E. Reed, K. Ino et al., “Rapid immunologicreconstitution following transplantation with mobilizedperipheral blood stem cells as compared to bone marrow,”Bone Marrow Transplantation, vol. 19, no. 2, pp. 161–172,1997.

[87] P. Dreger, K. Viehmann, N. von Neuhoff et al., “Autograftingof highly purified peripheral blood progenitor cells followingmyeloablative therapy in patients with lymphoma: a prospec-tive study of the long-term effects on tumor eradication,reconstitution of hematopoiesis and immune recovery,” BoneMarrow Transplantation, vol. 24, no. 2, pp. 153–161, 1999.

[88] A. Galy, S. Rudraraju, R. Baynes, and J. Klein, “Recoveryof lymphocyte and dendritic cell subsets after autologousCD34+ cell transplantation,” Bone Marrow Transplantation,vol. 25, no. 12, pp. 1249–1255, 2000.

[89] C.-H. Pui, L. L. Robison, and A. T. Look, “Acute lymphoblas-tic leukaemia,” The Lancet, vol. 371, no. 9617, pp. 1030–1043,2008.

[90] L. W. Ellisen, J. Bird, D. C. West et al., “TAN-1, thehuman homolog of the Drosophila Notch gene, is broken bychromosomal translocations in T lymphoblastic neoplasms,”Cell, vol. 66, no. 4, pp. 649–661, 1991.

[91] A. P. Weng, A. A. Ferrando, W. Lee et al., “Activatingmutations of NOTCH1 in human T cell acute lymphoblasticleukemia,” Science, vol. 306, no. 5694, pp. 269–271, 2004.

[92] A. P. Weng, J. M. Millholland, Y. Yashiro-Ohtani et al.,“c-Myc is an important direct target of Notch1 in T-cell acute lymphoblastic leukemia/lymphoma,” Genes andDevelopment, vol. 20, no. 15, pp. 2096–2109, 2006.

[93] B. J. Feldman, T. Hampton, and M. L. Cleary, “A carboxy-terminal deletion mutant of Notch1 accelerates lymphoidoncogenesis in E2A-PBX1 transgenic mice,” Blood, vol. 96,no. 5, pp. 1906–1913, 2000.

[94] L. J. Beverly and A. J. Capobianco, “Perturbation of Ikarosisoform selection by MLV integration is a cooperative eventin NotchIC-induced T cell leukemogenesis,” Cancer Cell, vol.3, no. 6, pp. 551–564, 2003.

[95] D. Allman, F. G. Karnell, J. A. Punt et al., “Separationof Notch1 promoted lineage commitment and expan-sion/transformation in developing T cells,” Journal of Experi-mental Medicine, vol. 194, no. 1, pp. 99–106, 2001.

[96] J. R. Gothert, R. L. Brake, M. Smeets, U. Duhrsen, C. G.Begley, and D. J. Izon, “NOTCH1 pathway activation is anearly hallmark of SCLT leukemogenesis,” Blood, vol. 110, no.10, pp. 3753–3762, 2007.

[97] S. Breit, M. Stanulla, T. Flohr et al., “Activating NOTCH1mutations predict favorable early treatment response andlong-term outcome in childhood precursor T-cell lym-phoblastic leukemia,” Blood, vol. 108, no. 4, pp. 1151–1157,2006.

[98] V. Asnafi, A. Buzyn, S. Le Noir et al., “NOTCH1/FBXW7mutation identifies a large subgroup with favorable outcomein adult T-cell acute lymphoblastic leukemia (T-ALL): aGroup for Research on Adult Acute Lymphoblastic Leukemia(GRAALL) study,” Blood, vol. 113, no. 17, pp. 3918–3924,2009.

[99] T. Palomero, K. McKenna, J. O-Neil et al., “Activatingmutations in NOTCH1 in acute myeloid leukemia andlineage switch leukemias,” Leukemia, vol. 20, no. 11, pp.1963–1966, 2006.

[100] L. Fu, H. Kogoshi, N. Nara, and S. Tohda, “NOTCH1mutations are rare in acute myeloid leukemia,” Leukemia andLymphoma, vol. 47, no. 11, pp. 2400–2403, 2006.

[101] S. Tohda, H. Kogoshi, N. Murakami, S. Sakano, and N. Nara,“Diverse effects of the Notch ligands Jagged1 and Delta1 onthe growth and differentiation of primary acute myeloblasticleukemia cells,” Experimental Hematology, vol. 33, no. 5, pp.558–563, 2005.

[102] S. Tohda and N. Nara, “Expression of Notch1 and Jagged1proteins in acute myeloid leukemia cells,” Leukemia andLymphoma, vol. 42, no. 3, pp. 467–472, 2001.

[103] G.-H. Li, Y.-Z. Fan, X.-W. Liu et al., “Notch signalingmaintains proliferation and survival of the HL60 humanpromyelocytic leukemia cell line and promotes the phos-phorylation of the Rb protein,” Molecular and CellularBiochemistry, vol. 340, no. 1-2, pp. 7–14, 2010.

[104] P.-M. Chen, C.-C. Yen, W.-S. Wang et al., “Down-regulationof Notch-1 expression decreases PU.1-mediated myeloiddifferentiation signaling in acute myeloid leukemia,” Inter-national Journal of Oncology, vol. 32, no. 6, pp. 1335–1341,2008.

[105] P. J. M. Valk, R. G. W. Verhaak, M. A. Beijen et al., “Prog-nostically useful gene-expression profiles in acute myeloidleukemia,” New England Journal of Medicine, vol. 350, no. 16,pp. 1617–1628, 2004.

[106] B. J. Wouters, M. A. Jorda, K. Keeshan et al., “Distinct geneexpression profiles of acute myeloid/T-lymphoid leukemiawith silenced CEBPA and mutations in NOTCH1,” Blood, vol.110, no. 10, pp. 3706–3714, 2007.

[107] T. Pabst and B. U. Mueller, “Complexity of CEBPA dysreg-ulation in human acute myeloid leukemia,” Clinical CancerResearch, vol. 15, no. 17, pp. 5303–5307, 2009.

14 Bone Marrow Research

[108] R. Hubmann, J. D. Schwarzmeier, M. Shehata et al., “Notch2is involved in the overexpression of CD23 in B-cell chroniclymphocytic leukemia,” Blood, vol. 99, no. 10, pp. 3742–3747,2002.

[109] E. Rosati, R. Sabatini, G. Rampino et al., “Constitutively acti-vated Notch signaling is involved in survival and apoptosisresistance of B-CLL cells,” Blood, vol. 113, no. 4, pp. 856–865,2009.

[110] P. Secchiero, E. Melloni, M. G. Di Iasio et al., “Nutlin-3 up-regulates the expression of Notch1 in both myeloidand lymphoid leukemic cells, as part of a negative feedbackantiapoptotic mechanism,” Blood, vol. 113, no. 18, pp. 4300–4308, 2009.

[111] T. Palomero, K. C. Barnes, P. J. Real et al., “CUTLL1, a novelhuman T-cell lymphoma cell line with t(7;9) rearrangement,aberrant NOTCH1 activation and high sensitivity to γ-secretase inhibitors,” Leukemia, vol. 20, no. 7, pp. 1279–1287,2006.

[112] C. Stapnes, B. T. Gjertsen, H. Reikvam, and ∅. Bruserud,“Targeted therapy in acute myeloid leukaemia: current statusand future directions,” Expert Opinion on InvestigationalDrugs, vol. 18, no. 4, pp. 433–455, 2009.

[113] J. J. Everly, R. C. Walsh, R. R. Alloway, and E. S. Woodle,“Proteasome inhibition for antibody-mediated rejection,”Current Opinion in Organ Transplantation, vol. 14, no. 6, pp.662–666, 2009.

[114] M. J. Everly, J. J. Everly, B. Susskind et al., “Proteasome inhibi-tion reduces donor-specific antibody levels,” TransplantationProceedings, vol. 41, no. 1, pp. 105–107, 2009.

[115] L. H. Mattingly, R. A. Gault, and W. J. Murphy, “Use ofsystemic proteasome inhibition as an immune-modulatingagent in disease,” Endocrine, Metabolic and Immune Disor-ders: Drug Targets, vol. 7, no. 1, pp. 29–34, 2007.

[116] C. Delaney, S. Heimfeld, C. Brashem-Stein, H. Voorhies, R. L.Manger, and I. D. Bernstein, “Notch-mediated expansion ofhuman cord blood progenitor cells capable of rapid myeloidreconstitution,” Nature Medicine, vol. 16, no. 2, pp. 232–236,2010.

[117] K. Li, Y. Li, W. Wu et al., “Modulation of notch signaling byantibodies specific for the extracellular negative regulatoryregion of NOTCH3,” Journal of Biological Chemistry, vol. 283,no. 12, pp. 8046–8054, 2008.

[118] K. Sugimoto, Y. Maekawa, A. Kitamura et al., “Notch2signaling is required for potent antitumor immunity in vivo,”Journal of Immunology, vol. 184, no. 9, pp. 4673–4678, 2010.

[119] M. Aste-Amezaga, N. Zhang, J. E. Lineberger et al., “Charac-terization of notch1 antibodies that inhibit signaling of bothnormal and mutated notch1 receptors,” PLoS One, vol. 5, no.2, article e9094, 2010.

[120] Y. Wu, C. Cain-Hom, L. Choy et al., “Therapeutic antibodytargeting of individual Notch receptors,” Nature, vol. 464, no.7291, pp. 1052–1057, 2010.

[121] T. Hoey, W.-C. Yen, F. Axelrod et al., “DLL4 blockade inhibitstumor growth and reduces tumor-initiating cell frequency,”Cell Stem Cell, vol. 5, no. 2, pp. 168–177, 2009.

[122] R. E. Moellering, M. Cornejo, T. N. Davis et al., “Directinhibition of the NOTCH transcription factor complex,”Nature, vol. 462, no. 7270, pp. 182–188, 2009.

[123] C. Stapnes, A. P. Døskeland, K. Hatfield et al., “Theproteasome inhibitors bortezomib and PR-171 have antipro-liferative and proapoptotic effects on primary human acutemyeloid leukaemia cells,” British Journal of Haematology, vol.136, no. 6, pp. 814–828, 2007.

[124] A. G. Clementz and C. Osipo, “Notch versus the proteasome:what is the target of gamma-secretase inhibitor-I?” BreastCancer Research, vol. 11, no. 5, p. 110, 2009.

[125] G. Calandra, G. Bridger, and S. Fricker, “CXCR4 in clinicalhematology,” Current Topics in Microbiology and Immunol-ogy, vol. 870, pp. 173–191, 2010.

[126] K. Liseth, M. Sjo, K. Paulsen, ∅. Bruserud, and E. Ersvaer,“Early pre-engraftment, functional, in vitro responsive-ness of T lymphocytes in allotransplanted, acute leukemiapatients: proliferation and release of a broad profile ofcytokines, possibly predictive of graft-versus-host disease,”European Cytokine Network, vol. 21, no. 1, pp. 40–49, 2010.

[127] D. Deangelo, R. Stone, and L. Silverman, “A phase I clinicaltrial of the notch inhibitor MK-0752 in patients with T-cell acute lymphoblastic leukemia/ lymphoma (T-ALL) andother leukemias,” Journal of Clinical Oncology, vol. 24, article6585, 2006.

[128] A. Pannuti, K. Foreman, P. Rizzo et al., “Targeting Notch totarget cancer stem cells,” Clinical Cancer Research, vol. 16, no.12, pp. 3141–3152, 2010.

[129] E. Weisberg, R. Barrett, Q. Liu, R. Stone, N. Gray, and J. D.Griffin, “FLT3 inhibition and mechanisms of drug resistancein mutant FLT3-positive AML,” Drug Resistance Updates, vol.12, no. 3, pp. 81–89, 2009.

[130] J. Zaia, L. Baden, M. J. Boeckh et al., “Viral diseaseprevention after hematopoietic cell transplantation,” BoneMarrow Transplantation, vol. 44, no. 8, pp. 471–482, 2009.

[131] K. A. Marr, E. Bow, T. Chiller et al., “Fungal infectionprevention after hematopoietic cell transplantation,” BoneMarrow Transplantation, vol. 44, no. 8, pp. 483–487, 2009.

[132] E. S. Yvon, S. Vigouroux, R. F. Rousseau et al., “Overex-pression of the Notch ligand, Jagged-1, induces alloantigen-specific human regulatory T cells,” Blood, vol. 102, no. 10, pp.3815–3821, 2003.

[133] M. Dominici, K. Le Blanc, I. Mueller et al., “Minimalcriteria for defining multipotent mesenchymal stromal cells.The International Society for Cellular Therapy positionstatement,” Cytotherapy, vol. 8, no. 4, pp. 315–317, 2006.

[134] W. A. Noort, A. B. Kruisselbrink, P. S. In’t Anker et al.,“Mesenchymal stem cells promote engraftment of humanumbilical cord blood-derived CD34+ cells in NOD/SCIDmice,” Experimental Hematology, vol. 30, no. 8, pp. 870–878,2002.

[135] L. M. Ball, M. E. Bernardo, H. Roelofs et al., “Cotransplanta-tion of ex vivo-expanded mesenchymal stem cells accelerateslymphocyte recovery and may reduce the risk of graft failurein haploidentical hematopoietic stem-cell transplantation,”Blood, vol. 110, no. 7, pp. 2764–2767, 2007.

[136] J. M. Weber and L. M. Calvi, “Notch signaling and the bonemarrow hematopoietic stem cell niche,” Bone, vol. 46, no. 2,pp. 281–285, 2010.

[137] Y.-H. Hu, D.-Q. Wu, F. Gao, G.-D. Li, and X.-C. Zhang,“Notch signaling: a novel regulating differentiation mecha-nism of human umbilical cord blood-derived mesenchymalstem cells into insulin-producing cells in vitro,” ChineseMedical Journal, vol. 123, no. 5, pp. 606–614, 2010.

[138] F. A. Saleh, M. Whyte, P. Ashton, and P. G. Genever,“Regulation of mesenchymal stem cell activity by endothelialcells,” Stem Cells and Development. In press.

[139] K. Le Blanc and O. Ringden, “Immunomodulation bymesenchymal stem cells and clinical experience,” Journal ofInternal Medicine, vol. 262, no. 5, pp. 509–525, 2007.

Bone Marrow Research 15

[140] S. Paczesny, S. W. Choi, and J. L. M. Ferrara, “Acutegraft-versus-host disease: new treatment strategies,” CurrentOpinion in Hematology, vol. 16, no. 6, pp. 427–436, 2009.

[141] T. Toubai, S. Paczesny, Y. Shono et al., “Mesenchymalstem cells for treatment and prevention of graft-versus-hostdisease after allogeneic hematopoietic cell transplantation,”Current Stem Cell Research and Therapy, vol. 4, no. 4, pp. 252–259, 2009.

[142] F. Liotta, R. Angeli, L. Cosmi et al., “Toll-like receptors 3 and4 are expressed by human bone marrow-derived mesenchy-mal stem cells and can inhibit their T-cell modulatory activityby impairing notch signaling,” Stem Cells, vol. 26, no. 1, pp.279–289, 2008.

Hindawi Publishing CorporationBone Marrow ResearchVolume 2011, Article ID 849387, 7 pagesdoi:10.1155/2011/849387

Review Article

Transplantation for Congenital Bone Marrow Failure Syndromes

Kenji Morimoto,1 Theodore B. Moore,1 Gary Schiller,2 and Kathleen M. Sakamoto1, 3

1 Division of Hematology/Oncology, Department of Pediatrics, David Geffen School of Medicine at UCLA, University of California,10833 Le Conte Avenue, Los Angeles, CA 90095-1752, USA

2 Division of Hematology/Oncology, Department of Medicine, University of California, Los Angeles, CA, USA3 Gwynne Hazen Cherry Memorial Laboratories and Mattel Children’s Hospital, Jonsson Comprehensive Cancer Center,Molecular Biology Institute, California NanoSystems Institute, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA

Correspondence should be addressed to Kenji Morimoto, [email protected]

Received 2 August 2010; Accepted 26 October 2010

Academic Editor: Axel R. Zander

Copyright © 2011 Kenji Morimoto et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Congenital bone marrow failure syndromes (BMFSs) are relatively rare disorders characterized by aberrant development in oneor more hematopoietic lineages. Genetic alterations have now been identified in most of these disorders although the exact roleof the molecular defects has yet to be elucidated. Most of these diseases are successfully managed with supportive care, however,treatment refractoriness and disease progression—often involving malignant transformation—may necessitate curative treatmentwith hematopoietic stem cell transplantation. Due to the underlying molecular defects, the outcome of transplantation for BMFSmay be dramatically different than those associated with transplantation for more common diseases, including leukemia. Givenrecent improvements in survival and molecular diagnosis of bone marrow failure syndrome patients presenting at adult ageswithout physical stigmata, it is important for both pediatric and adult hematologists to be aware of the possible diagnosis of BMFsyndromes and the unique approaches required in treating such patients.

1. Introduction

Bone marrow failure syndromes (BMFSs) are a constellationof disorders characterized by abnormal hematopoiesis andoften accompanied by assorted physical findings and a strik-ingly increased risk for hematologic malignancy. Historically,patients presented with significant cytopenias and physicalstigmata during childhood. However, it is now evident thatBMFS posses a broad range of phenotype penetrance andan increasing number of patients are being identified inadulthood, occasionally presenting with leukemia or aplasticanemia. Some BMFS disorders are associated with defectiveDNA repair or maintenance pathways and possess increasedsensitivity to DNA-damaging agents including chemother-apy and radiation. These sensitivities have been clearlydemonstrated by the significantly increased chemotherapy-related and stem cell transplant-related morbidity andmortality in BMFS patients. It is therefore important forboth pediatric and adult hematologists to be aware of thepresentation, management, complications, and approachesto treat BMFS. We review here transplantation approaches

and outcomes in patients with Fanconi Anemia, Diamond-Blackfan anemia, dyskeratosis congenita, severe congeni-tal neutropenia, and Shwachman-Diamond syndrome toillustrate their relative risks for malignancy and reportedutility of varying transplant regimens. The reporting ofcumulative transplant outcomes will hopefully add clarity tothe indications for transplantation in these disorders.

2. Fanconi Anemia

Fanconi Anemia (FA) is a disease of abnormal DNA repairresulting in abnormal hematopoiesis, variable anatomicphenotypes, and an increased predisposition to malignancy.13 FANC genes have been identified that are responsiblefor forming an ubiquitinating nuclear core complex, actingas a chromatin binding cofactor for other DNA repairenzymes, or exerting DNA helicase activity all in response toDNA intrastrand cross-linking damage. The gene FANCD1was previously identified as BRCA2, the breast cancersusceptibility gene 2.

2 Bone Marrow Research

Physical findings may include short stature and skeletalabnormalities (classically involving radius or thumb), devel-opmental delay, abnormal kidneys, skin pigmentation/cafeau lait spots, and triangular facies in 60%–75% of patients.Hematologic cytopenias are almost always present at diagno-sis and may include uni-, bi-, or trilineage involvement. Therisk of early bone marrow failure appears to correlate withthe presence of certain physical abnormalities and high-riskpatients face a 7% annual incidence of bone marrow failure[1]. Conversely, patients without physical findings or subtlephysical abnormalities are at less risk for early bone marrowfailure and may thus present in adulthood. Ten percent ofFA patients may present at greater than 16 years of age [1].Fanconi Anemia patients are at increased risk of developingmalignancies including acute myeloid leukemia (AML),myelodysplastic syndrome (MDS), head/neck cancers, andgynecologic malignancies; the relative risk of developingspecific solid tumors and AML/MDS ranges from severalhundred- to several thousand-fold compared to the generalpopulation [2, 3].

Given the high risk of bone marrow failure andclonal abnormalities, FA patients are often considered forhematopoietic stem cell transplantation. Initial transplan-tation attempts with antigen-matched donors resulted in80% mortality at 3 years when conditioned with cyclophos-phamide 200 mg/kg and 500 cGy thoraco-abdominal irra-diation (TAI) [4]. In vitro studies of FA primary cellsdemonstrated increased sensitivity to cyclophosphamide[5] and irradiation [6]. Subsequent transplants have thusused reduced intensity conditioning regimens and achievedincreased survival and decreased graft versus host disease(GVHD) severity. At present, transplant from a sibling-matched donor remains the optimum therapy for FA. Arecent series of 30 patients without demonstrable clonality ormalignancy receiving sibling-matched donor bone marrowsused a conditioning regimen of ATG, cyclophosphamide20 mg/kg (40 mg/kg for one 5/6 sibling match), 400 cGy TAI(450 cGy for lone 5/6 sibling match), and GVHD prophylaxisof cyclosporine and steroids. The authors reported a 97%neutrophil engraftment rate and 77% five-year overall sur-vival with an incidence of 20% acute GVHD (aGVHD) and7% chronic GVHD (cGVHD) [7].

In contrast, Wagner et al. reported 98 FA patients (78HLA identical matches) who underwent an unrelated bonemarrow transplantation and were conditioned with TBI(450 cGy), cyclophosphamide (20–40 mg/kg), and ATG ±fludarabine. The addition of fludarabine (used in post-1998transplants) with T-cell depletion significantly enhancedneutrophil engraftment rates (89%), decreased aGVHDGrade II–IV probability (16%), and increased adjusted 100-day survival (76% versus 35%) and 3-year survival (52%versus 13%) compared to nonfludarabine regimens. Overallmortality within the fludarabine-conditioned patients wassecondary to graft failure (15%), GVHD (11%), and organfailure (11%) [8]. Gluckman et al. reported the outcome of93 unrelated umbilical cord stem cell transplant recipientswith half of patients receiving a 6/6 or 5/6 matched cord,the other half receiving 2- or 3-HLA mismatched cords.Sixty one percent of patients received fludarabine-based

regimens, 80% received TBI, and 81% received anti-T celltherapy. Median average cell count was 1.9 × 105 CD34+cells/kg and 4.9 × 107 TNC/kg. Neutrophil engraftment atthe relatively late time-point of 60 days was achieved by 60%(median time to neutrophil recovery for these patients was23 days), incidence of acute GVHD Grade II–IV 32% (noaGVHD observed in matched cords), and a 3-year overallsurvival of 40% (50% in patients receiving fludarabine-based conditioning) [9]. Patients receiving 6/6, 5/6, and 3-4/6 matched cords had 3-year OS of 74%, 48%, and 25%respectively. Of the 56 deaths, mortality was largely due toinfection (31%), graft failure (21%), and aGVHD (12.5%).

Acknowledging the increased risks irradiation may posefor FA patients, an international retrospective study of 148sibling-matched patients receiving irradiation, cyclophos-phamide ± ATG versus cyclophosphamide alone or in com-bination with ATG, fludarabine, or busulfan reported similarneutrophil engraftment rates by day 28 (94% versus 89%,resp.) and day 100 (92% each), similar incidence of GradeII–IV acute GVHD (23% versus 21%, resp.), and similaradjusted probability of 5-year survival (78% versus 81%,resp.) with similar mortality from infection, GVHD, andorgan failure [10]. Consistent with these findings, Locatelli etal. analyzed the utility of cyclophosphamide and fludarabineconditioning in 31 sibling-matched HSCT patients with areported estimated 8-year OS of 87% [11]. A radiation dosereduction study at the University of Minnesota reported100% engraftment with 300 cGy TBI in 22 FA patientsreceiving MUD transplants (although both patients receiving150 cGy experienced secondary graft failure) [12].

Transplantation remains the front-line therapy for bonemarrow failure in FA patients. Sibling-matched donors havetraditionally been the donor although MUD BMT andUCSCT outcomes have improved with the incorporationof fludarabine into the conditioning regimen. Current datasupport the feasibility of reducing radiation dosages fromthe standard conditioning regimens in MUD transplants,and perhaps elimination of radiation in sibling-matchedtransplants.

3. Diamond Blackfan Anemia

Diamond Blackfan Anemia (DBA) is a congenital disorderconsisting of red cell aplasia and assorted physical abnor-malities including short stature, craniofacial abnormalities,and radial irregularities. Consistent with other congenitalBMF syndromes, DBA patients have an increased risk ofdeveloping malignancies. A longitudinal followup of 72 DBApatients reported 4 cases of AML ranging 15–30 years afterthe diagnosis of DBA [13]. Osteogenic sarcoma has alsobeen reported in DBA patients [14]. Over 50% of DBApatients possess a mutation in a ribosomal protein, althoughthe exact pathophysiology of the disease remains unclear.Patients typically present in infancy with a macrocyticanemia [15]; erythrocytes have disrupted maturation andpossess elevated adenosine deaminase levels [16]. There is aspectrum of anemia severity and red blood cell transfusionrequirements. Steroids are the front-line medical therapy

Bone Marrow Research 3

and approximately 70%–80% of DBA patients will initiallyhave a transfusion free response [17, 18]. Only 20% retaina sustained remissions off of steroids and thus steroid-related complications and transfusion-induced iron overloadremain significant problems.

Case reports of HSCT in DBA date back to 1976[19] but large studies are limited. A retrospective reviewanalyzed transplants from 8 sibling matched versus 12MUD donors in 20 patients, resulting in a 5-year overallsurvival of 87.5% and 28%, respectively [20] (excluding onedeath from osteogenic sarcoma in the MUD arm). Theseresults are confounded by differences in radiation containingconditioning regimens. Roy et al. [21] reported IBMTRregistry outcomes of 61 transplanted DBA patients (medianage 7 years), with two-thirds receiving transplants frommatched sibling donors and 20% from MUD (remainderreceived nonsibling family donor). 95% of sibling-matchedrecipients received myeloablative, cyclophosphamide-basedconditioning, and 45% of MUD recipients received TBI-based conditioning. All patients received bone marrowstem cells. GVHD prophylaxis consisted of cyclosporine,methotrexate, or both in 90% of the patients. Ninety-twopercent and 75% of all patients demonstrated neutrophiland platelet engraftment by Day 100, respectively with nodifference between the two arms (although Day 28 engraft-ment rates were significantly higher in the sibling-matchedarm). Total nonengraftment incidence was 9%. Cumulativeincidence of acute GVHD grade II–IV was 28% and cGVHDwas 19% with overall incidence being too small to detectdifferences between the two donor types. Heavily transfusedpatients (receiving greater than 50 transfusions) had slowerengraftment rates (at Day 28 for neutrophil recover and Day60 platelet recovery) although this did not impact total Day100 engraftment percentages. Overall survival at 1 and 3years was significantly superior with HLA-identical siblingdonors versus matched, unrelated donors (78% versus 45%,and 76% versus 39%, resp.). Of the 23 deaths, the mostprevalent causes of mortality were infection (31%), graftfailure (27%), and interstitial pneumonia (13%). Only onepatient died due to GVHD. The use of irradiation, thenumber of transfusions, time duration between diagnosisand transplant, and age at transplant were not associatedwith differences in survival. Data regarding iron overloadstatus at time of transplant were not available.

The poor outcomes of MUD transplantation are likelydue to the relatively high number of antigen-based matchesinvolved in these reports. In the relatively more HLA-homogenous Japanese population, nineteen Japanese DBApatients [18] receiving transplants from varying sources(42% sibling matched, 32% matched unrelated, 21% mis-matched unrelated, 5% mismatched related) and condi-tioned with cyclophosphamide plus irradiation in 68% andwithout irradiation in 32% had (excluding one early death)100 day neutrophils engraftment rates of 90%, with bothnonengrafting patients receiving unrelated, 4/6 cord bloodtransplants. GVHD prophylaxis was largely cyclosporine-based; incidence of acute GVHD grades II–IV was 25%and only one case of chronic GVHD was reported. Overallsurvival at 5 years was 79%; 5-year failure-free survival was

100% in recipients of bone marrow transplants compared to40% of cord-blood recipients. Observable outcome differ-ences between HLA-identical sibling BMT and MUD BMTwere limited to recovery time of platelets although overallengraftment rates were not significantly different. Addition-ally, recent mention of the DBA Registry data reports earlyoverall survival rate of 85% in MUD transplants performedsince 2000, far superior to more dated publications [22].

Historically, upfront transplantation for DBA was largelyrestricted to transfusion-requiring patients with availablesibling-matched donors. It is now evident that MUD trans-plantation is an increasingly feasible option although moredata and experience are required before identifying its exactrole for DBA. Limiting transfusion exposures, controllingiron overload, minimizing steroid-induced adverse effects,and monitoring for malignancy continue to be of crucialimportance in the care of these patients.

4. Dyskeratosis Congenita

Dyskeratosis congenita (DC) is a failure in telomere main-tenance resulting in the classic triad of abnormal skinpigmentation, nail dystrophy, and leukoplakia of the oralmucosa as well as lacrimal duct stenosis, pulmonary fibrosis,and bone marrow failure. Mutations in the telomere main-tenance complexes (including DKC1, TERC, TERT, NOP10,NHP2, TINF2) result in shortened telomeres especially instem cells undergoing high proliferative rates. Abrogation ofhematopoietic stem cell replicative potential results in bonemarrow failure in 85% of patients and is the cause of death in80% [23]. Similar to Fanconi Anemia patients, DC patientsare at increased risk for hematologic and solid malignancieswith a cumulative incidence approaching 50% at 50 years ofage [24, 25].

Initial HSCT attempts using myeloablative conditioningresulted in high morbidity and mortality, notably frompulmonary fibrosis [26]. Several case reports of successfultransplantation using nonmyeloablative regimens have beenpublished [27, 28]. A series of six patients receiving unrelateddonor bone marrow (one 8/8, one 7/8), unrelated doublecords (4/6 for both cords in one patient, and 4/6 and 5/6for two other patients), and HLA-identical related peripheralblood stem cells used nonmyeloablative conditioning withcyclophosphamide, fludarabine, alemtuzumab, and 200 cGyTBI [29]. GVHD prophylaxis consisted of cyclosporine andmycophenolate. Five patients engrafted (one patient diedprior to engraftment at 1 month); two out of six patientsdied from infectious complications, two patients developedaGVHD (skin, gut) Grade II–IV, and one patient developedchronic skin GVHD; overall survival was 67% with medianfollow-up of 2 years. The authors’ review of published casereports using non-myeloablative conditioning in transplan-tation for DC (excluding the authors’ six reported patients)yielded 18 patients (11 related bone marrow transplants,5 unrelated bone marrow transplants, 2 unrelated cordtransplants). 91% of related transplant recipients were alivewith a median follow-up of two years. 40% of matched-unrelated transplant recipients were alive with a median

4 Bone Marrow Research

follow-up of 15 months; there was a high incidence of GVHDin the 60% of patients that died. Transplantation usingreduced intensity conditioning is therefore indicated for DCgiven its high probability of bone marrow failure and death,with expected two year post-transplant survival rates of 40–90% depending upon stem cell source.

5. Severe Congenital Neutropenia

Severe congenital neutropenia (SCN) is a constellationof syndromes consisting of arrested myeloid developmentresulting in neutropenia. The genetic pathology is largelydue to mutations in ELA2 and GFI1; a subset of patientshave mutations in the Hax1 gene (Kostmann’s neutropenia)and rare, congenital X-linked mutations in CSF3R [30].The natural history of the disease includes infectious com-plications and, consistent with other bone marrow failuresyndromes, an elevated risk of myelodysplastic and leukemictransformation. Approximately 90% of patients respond toGCSF administration with a subsequent decrease in sepsis-related mortality to almost 1% per year during the firstdecade of life [31]. However, GCSF poor and nonrespon-ders remain at particularly high risk for hematopoieticdysplasia/malignancy with a cumulative incidence of 21%following 10 years of GCSF therapy that was thoughtto dramatically climb thereafter [31]. However recentlyupdated North American registry data suggest a relativeplateau in the hematologic malignancy risk after 10–15 years,similar to that seen in Fanconi Anemia and dyskeratosiscongenita [32]. It is thought that the SCN pathology anddegree of severity particularly predisposes to malignancyas the incidence of MDS/AML transformation at 10 yearswas only 15% for patients without neutropenia on GCSFversus 34% for patients with persistent neutrophil countsbelow 2100/microliter despite high doses (greater than8 mcg/kg/day) of GCSF. Although it remains controversialif prolonged GCSF therapy predisposes to malignant trans-formation it is notable that large cohorts of other patientsreceiving chronic GCSF therapy for cyclic and idiopathicneutropenia have not reported malignant transformation[33]. Furthermore the risk of malignant transformation inSCN patients receiving high doses of GCSF does not appearto be greater than other bone marrow failure syndromeswith high malignancy rates such as Fanconi Anemia anddyskeratosis congenita [32].

Zeidler et al. reported 11 SCN patients without malig-nancy who received HSCT [34]; 8 received sibling-matchedtransplants (5 BMT, 1 cord, 1 BMT and cord, 1 PBSC)and 10 received myeloablative conditioning with busul-fan/cyclophosphamide ± ATG, thiotepa, or melphalan.All patients receiving myeloablative conditioning engraftedregardless of source although 40% of tested patients hadgraft chimerism (30%–84% donor) at 6–12 months post-transplant. Acute GVHD grade II–IV was present in 2of the unrelated BMT recipients. Overall survival of the10 patients receiving myeloablative regimens was 80% atmedian follow-up of 10 months. The French SCN registryreported HSCT outcomes on 5 SCN patients without

malignant transformation. Donors included unrelated cords(2) and unrelated bone marrows (2), and related bonemarrow (1). All received myeloablative regimens with 80%engraftment and 40% mortality due to infection at one yearpost-transplant [35]. In contrast, 18 Japanese patients withSCN without malignant transformation underwent trans-plantation from sibling-matched (9) or matched-unrelateddonors (9) following myeloablative conditioning in 12 casesand non-myeloablative in the remaining 6 patients. Althoughfour patients encountered primary graft failure (includingtwo sibling matched donor transplants) and received asecond transplant, 16 of the 18 patients are considereddisease-free with a median follow-up of 6 years [36].

Hematopoietic stem cell transplantation in SCN patientsin the setting of MDS/AML carries high mortality (60%–80% as reported in case series [37] and 1998 SCNIR registrydata) but is the only curative option. Choi et al. [37] reported6 SCN patients with transformation to MDS (2) or AML (4)at a median age of 13 years old; patients received sibling-matched (1), matched-unrelated (3), or single mismatchunrelated (2) donors and were conditioned with busulfan-based (5) or TBI-based (1) regimens. The two MDS patientsdid not receive induction chemotherapy, engrafted, and weredisease-free from 2 to 4 years posttransplant, although bothremain on immunosuppression for chronic GVHD. The fourAML patients received induction chemotherapy and weretransplanted in complete remission; those receiving single-mismatched grafts had primary graft failure. Ultimately allfour patients died: two died from chronic GVHD, one diedfrom primary graft failure, and one died from relapsedAML. The French SCN registry [35] reported four SCNpatients transplanted for MDS (3) and ALL (1) who receivedunrelated cords, bone marrows, or related bone marrow aftermyeloablative conditioning. All four patients engrafted andthree survived (one died from septic shock) with medianfollowup of 2 years. One patient required a second transplantfor relapsed MDS/AML. Of note, 2 of the 3 MDS patients didnot receive induction chemotherapy prior to transplant.

Transplantation is therefore indicated for patients unre-sponsive to GCSF and should be strongly considered for mildresponse to high doses of GCSF given this group’s risk ofmalignancy and the disparity between transplant survivalprior to malignant transformation compared to afterwards.The utility of induction chemotherapy is unknown inpatients with MDS, as successful donor engraftment andrelapse-free survival without induction chemotherapy havebeen demonstrated [35].

6. Shwachman-Diamond Syndrome

Shwachman-Diamond Syndrome (SDS) is an autosomalrecessive disorder associated with mutation in the Shwach-man Bodian Diamond Syndrome gene, SBDS, involved inribosomal biogenesis and mitotic spindle association. Thedisorder is classically associated with neutropenia (intermit-tent or chronic) and exocrine pancreas deficiency. However,patients often have multiple hematopoietic cell lineageinvolvement with anemia and thrombocytopenia seen in

Bone Marrow Research 5

up to 80% of patients [38] and stromal cell abnormalitieshave also been reported [39]. Other findings include skeletal,cardiac, endocrine, and renal abnormalities.

Similar to Fanconi Anemia and dyskeratosis congenita,SDS patients have an increased risk for the development ofMDS and AML. The French SCN registry recorded a 19%cumulative incidence of MDS/AML by 20 years of age in 55SDS patients; this incidence increased to 36% by 30 yearsof age [40]. In contrast, the NCI’s IBMF registry of 17 SDSpatients followed for a cumulative duration of 274 person-years did not observe any malignancy development [25];further registry followup with more patients is necessaryto more accurately identify SDS patients’ true malignancypredisposition.

Donadieu et al. reported the outcomes of 10 SDS patientsreceiving HSCT; 5 of the patients were transplanted forbone marrow failure without malignancy [41]. Four receivedunrelated BMT (either 9/10 or 10/10 HLA-matched) and onereceived a sibling-matched BMT. Myeloablative conditioningutilized busulfan-based (3) or TBI-based (2) regimens, with3 patients also receiving ATG. All five patients engrafted andwere alive at a median of 8 years with no chronic GVHDgrade III-IV. The study also reported five SDS patientstransplanted for MDS (4) and leukemic (1) transformationwith sibling-matched donors (3) and 9/10 MUD (2) [41].Myeloablative conditioning was with busulfan (3) or TBI (2)regimens. Only one patient received induction chemother-apy. Four patients died; two died of sepsis and organ toxicityprior to 40 days posttransplant, one patient relapsed at 5months, and one died of respiratory complications at 1.7years post-transplant. Other myeloablative regimens havereported high transplant-related mortality with significantlyelevated transplant-related cardiac toxicity [42]. A review of27 case-reported SDS patients transplanted using myeloabla-tive conditioning demonstrated a 18% mortality in patientstransplanted for cytopenias(s) versus a 53% mortality rate inpatients with active dysplasia/malignancy [43].

Reduced-intensity conditioning has been successfullyused in SDS patients. Sauer et al. used a conditioningregimen of fludarabine, treosulfan, and melphalan in threerecipients transplanted for pancytopenia (2) and MDS (1)[44]. Two patients were alive at 16 and 24 months post-transplant without respiratory complications; one patienttransplanted for pancytopenia died at Day +98 due toidiopathic pneumonia syndrome. Bhatla et al. reported seventransplanted patients conditioned with alemtuzumab, flu-darabine, and melphalan in varying states of hypocellularity[45]. One patient was transplanted for AML postinductionchemotherapy, another for MDS. 57% received sibling-matched marrow stem cells; the remaining three patientsreceived unrelated peripheral blood or bone marrow stemcells. All 7 patients engrafted and were alive with medianfollowup of 1.5 years. There was no acute GVHD GradeIII-IV using prophylactic treatment with cyclosporine andmethotrexate/steroids (depending on stem cell source).

Transplantation in SDS patients with active dysplasia ormalignancy is associated with significantly higher mortality.The risk of malignancy is elevated although to what degreeis unclear. Therefore, close surveillance of these patients is

advised for early detection of worsening cytopenias withclonal evolution. Reduced-intensity conditioning regimenshave been successfully used in limited series with goodengraftment rates and decreased transplant-related mortal-ity.

7. Conclusions

The bone marrow failure syndromes constitute a het-erogeneous group of molecular pathologies that share aphenotype of hematopoietic disruption. Hematopoietic stemcell transplantation remains the only curative treatmentoption at present but the indications for transplant aredependent upon the predisposition to severe bone marrowfailure/malignancy transformation and must be balancedagainst the known increased transplant-related complicationrates seen in such patients. Furthermore, it is importantfor physicians to recognize the potential for underlyingBMFS in patients, regardless of age, presenting with aplasiaand to tailor the diagnostic workup and potential trans-plantation regimen accordingly. Of note, BMFS-associatedaplastic anemia does not respond to the immunosuppressivetherapy regimens used for acquired aplastic anemia. There-fore time should not be wasted with such treatments assubsequent malignant transformation may result in severelyincreased transplantation-related mortality, most notablyreported in patients with severe congenital neutropeniaand Shwachman-Diamond syndrome. In addition, patientswith Fanconi’s Anemia and dyskeratosis congenita requirereduced intensity conditioning regimens that likely hindertransplantation survival in the setting of malignancy.

It is imperative that BMFS patients are entered intotheir respective registries. These databases enable these rarepatients to be studied in a systematic fashion for bothimmediate and long-term sequelae. Late transplant-relatedeffects may be seen in nearly any organ system, withnotable risks for cardiac, pulmonary, renal [46], endocrine,psychiatric, and fertility dysfunction [47]. Limited data onsurvivors of pediatric HSCT report “good” quality of lifewith median Lansky/Karnofsky performance scores of 90[48]. Furthermore, transplant survivors of nonmalignantdisorders may enjoy a superior health-related quality of liferelative to survivors of malignant disorders [49]. However,given the known increased incidence in transplant-relatedmorbidity as well as nonhematopoietic complications inBMFS patients it will be important for the registries tocontinue their long-term followup.

Acknowledgments

K. Morimoto is funded by the ASH Research Fellow award. K.M. Sakamoto is supported by NIH Grants R01 HL75826 andHL83077. K. M. Sakamoto is a Scholar of the Leukemia andLymphoma Society and has a Developmental HematologyTraining Grant T32 HL086345. This work was funded by theASH Alternative Training Pathway Grant (K. M. Sakamotoand G. Schiller).

6 Bone Marrow Research

References

[1] P. S. Rosenberg, Y. Huang, and B. P. Alter, “Individualized risksof first adverse events in patients with Fanconi anemia,” Blood,vol. 104, no. 2, pp. 350–355, 2004.

[2] D. I. Kutler, A. D. Auerbach, J. Satagopan et al., “Highincidence of head and neck squamous cell carcinoma inpatients with Fanconi anemia,” Archives of Otolaryngology, vol.129, no. 1, pp. 106–112, 2003.

[3] P. S. Rosenberg, M. H. Greene, and B. P. Alter, “Cancerincidence in persons with Fanconi anemia,” Blood, vol. 101,no. 3, pp. 822–826, 2003.

[4] E. Gluckman, A. Devergie, and G. Schaison, “Bone mar-row transplantation in Fanconi anaemia,” British Journal ofHaematology, vol. 45, no. 4, pp. 557–564, 1980.

[5] R. Berger, A. Bernheim, E. Gluckman, and C. Gisselbrecht,“In vitro effect of cyclophosphamide metabolites on chro-mosomes of Fanconi anaemia patients,” British Journal ofHaematology, vol. 45, no. 4, pp. 565–568, 1980.

[6] E. Gluckman, A. Devergie, and J. Dutreix, “Radiosensitivity inFanconi anaemia: application to the conditioning regimen forbone marrow transplantation,” British Journal of Haematology,vol. 54, no. 3, pp. 431–440, 1983.

[7] A. Farzin, S. M. Davies, F. O. Smith et al., “Matched siblingdonor haematopoietic stem cell transplantation in Fanconianaemia: an update of the Cincinnati Children’s experience,”British Journal of Haematology, vol. 136, no. 4, pp. 633–640,2007.

[8] J. E. Wagner, M. Eapen, M. L. MacMillan et al., “Unrelateddonor bone marrow transplantation for the treatment ofFanconi anemia,” Blood, vol. 109, no. 5, pp. 2256–2262, 2007.

[9] E. Gluckman, V. Rocha, I. Ionescu et al., “Results of unrelatedcord blood transplant in fanconi anemia patients: risk factoranalysis for engraftment and survival,” Biology of Blood andMarrow Transplantation, vol. 13, no. 9, pp. 1073–1082, 2007.

[10] R. Pasquini, J. Carreras, M. C. Pasquini et al., “HLA-matchedsibling hematopoietic stem cell transplantation for fanconianemia: comparison of irradiation and nonirradiation con-taining conditioning regimens,” Biology of Blood and MarrowTransplantation, vol. 14, no. 10, pp. 1141–1147, 2008.

[11] F. Locatelli, M. Zecca, A. Pession et al., “The outcomeof children with Fanconi anemia given hematopoietic stemcell transplantation and the influence of fludarabine in theconditioning regimen: a report from the Italian PediatricGroup,” Haematologica, vol. 92, no. 10, pp. 1381–1388, 2007.

[12] M. L. MacMillan et al., “Alternate donor Hct for Fanconianemia (Fa): results of a total body irradiation (Tbi) dose de-escalation study,” Biology of Blood and Marrow Transplanta-tion, vol. 15, no. 2, pp. 3–4, 2009.

[13] A. J. Janov, T. Leong, D. G. Nathan, and E. C. Guinan,“Diamond-Blackfan anemia: natural history and sequelae oftreatment,” Medicine, vol. 75, no. 2, pp. 77–87, 1996.

[14] J. M. Lipton, N. Federman, Y. Khabbaze et al., “Osteogenicsarcoma associated with Diamond-Blackfan anemia: a reportfrom the Diamond-Blackfan anemia registry,” Journal ofPediatric Hematology/Oncology, vol. 23, no. 1, pp. 39–44, 2001.

[15] J. M. Lipton, M. Kudisch, R. Gross, and D. G. Nathan,“Defective erythroid progenitor differentiation system incongenital hypoplastic (Diamond-Blackfan) anemia,” Blood,vol. 67, no. 4, pp. 962–968, 1986.

[16] B. E. Glader and K. Backer, “Elevated red cell adenosinedeaminase activity: a marker of disordered erythropiesis inDiamond-Blackfan anaemia and other haematologic diseases,”

British Journal of Haematology, vol. 68, no. 2, pp. 165–168,1988.

[17] A. Vlachos, S. Ball, N. Dahl et al., “Diagnosing and treatingDiamond Blackfan anaemia: results of an international clinicalconsensus conference,” British Journal of Haematology, vol.142, no. 6, pp. 859–876, 2008.

[18] H. Mugishima, S. Ohga, A. Ohara, S. Kojima, K. Fujisawa, andI. Tsukimoto, “Hematopoietic stem cell transplantation forDiamond-Blackfan anemia: a report from the Aplastic AnemiaCommittee of the Japanese Society of Pediatric Hematology,”Pediatric Transplantation, vol. 11, no. 6, pp. 601–607, 2007.

[19] C. S. August, E. King, and J. H. Githens, “Establishmentof erythropoiesis following bone marrow transplantationin a patient with congenital hypoplastic anemia (DiamondBlackfan syndrome),” Blood, vol. 48, no. 4, pp. 491–498, 1976.

[20] A. Vlachos, N. Federman, C. Reyes-Haley, J. Abramson,and J. M. Lipton, “Hematopoietic stem cell transplantationfor Diamond Blackfan anemia: a report from the DiamondBlackfan Anemia Registry,” Bone Marrow Transplantation, vol.27, no. 4, pp. 381–386, 2001.

[21] V. Roy, W. S. Perez, M. Eapen et al., “Bone marrow transplan-tation for diamond-blackfan anemia,” Biology of Blood andMarrow Transplantation, vol. 11, no. 8, pp. 600–608, 2005.

[22] A. Vlachos and E. Muir, “How I treat Diamond Blackfananemia,” Blood. In press.

[23] M. Kirwan and I. Dokal, “Dyskeratosis congenita, stem cellsand telomeres,” Biochimica et Biophysica Acta, vol. 1792, no. 4,pp. 371–379, 2009.

[24] B. P. Alter, N. Giri, S. A. Savage, and P. S. Rosenberg, “Cancerin dyskeratosis congenita,” Blood, vol. 113, no. 26, pp. 6549–6557, 2009.

[25] B. P. Alter, N. Giri, S. A. Savage et al., “Malignancies andsurvival patterns in the National Cancer Institute inheritedbone marrow failure syndromes cohort study,” British Journalof Haematology, vol. 150, no. 2, pp. 179–188, 2010.

[26] M. Yabe, H. Yabe, K. Hattori et al., “Fatal interstitialpulmonary disease in a patient with dyskeratosis congenitaafter allogeneic bone marrow transplantation,” Bone MarrowTransplantation, vol. 19, no. 4, pp. 389–392, 1997.

[27] B. Nobili, G. Rossi, P. De Stefano et al., “Successful umbilicalcord blood transplantation in a child with dyskeratosiscongenita after a fludarabine-based reduced-intensity condi-tioning regimen,” British Journal of Haematology, vol. 119, no.2, pp. 573–574, 2002.

[28] T. Gungor, S. Corbacioglu, R. Storb, and R. A. Seger,“Nonmyeloablative allogeneic hematopoietic stem cell trans-plantation for treatment of Dyskeratosis congenita,” BoneMarrow Transplantation, vol. 31, no. 5, pp. 407–410, 2003.

[29] A. C. Dietz et al., “Disease-specific hematopoietic celltransplantation: nonmyeloablative conditioning regimen fordyskeratosis congenita,” Bone Marrow Transplant. In press.

[30] K. Devriendt, A. S. Kim, G. Mathijs et al., “Constitutively acti-vating mutation in WASP causes X-linked severe congenitalneutropenia,” Nature Genetics, vol. 27, no. 3, pp. 313–317,2001.

[31] P. S. Rosenberg, B. P. Alter, A. A. Bolyard et al., “The incidenceof leukemia and mortality from sepsis in patients with severecongenital neutropenia receiving long-term G-CSF therapy,”Blood, vol. 107, no. 12, pp. 4628–4635, 2006.

[32] P. S. Rosenberg, C. Zeidler, A. A. Bolyard et al., “Stable long-term risk of leukaemia in patients with severe congenitalneutropenia maintained on G-CSF therapy: short report,”British Journal of Haematology, vol. 150, no. 2, pp. 196–199,2010.

Bone Marrow Research 7

[33] M. H. Freedman, M. A. Bonilla, C. Fier et al., “Myelodysplasiasyndrome and acute myeloid leukemia in patients withcongenital neutropenia receiving G-CSF therapy,” Blood, vol.96, no. 2, pp. 429–436, 2000.

[34] C. Zeidler, K. Welte, Y. Barak et al., “Stem cell transplantationin patients with severe congenital neutropenia without evi-dence of leukemic transformation,” Blood, vol. 95, no. 4, pp.1195–1198, 2000.

[35] C. Ferry, M. Ouachee, T. Leblanc et al., “Hematopoieticstem cell transplantation in severe congenital neutropenia:experience of the French SCN register,” Bone Marrow Trans-plantation, vol. 35, no. 1, pp. 45–50, 2005.

[36] K. Oshima, R. Hanada, R. Kobayashi et al., “Hematopoieticstem cell transplantation in patients with severe congenitalneutropenia: an analysis of 18 Japanese cases,” PediatricTransplantation, vol. 14, no. 5, pp. 657–663, 2010.

[37] S. W. Choi, L. A. Boxer, M. A. Pulsipher et al., “Stem celltransplantation in patients with severe congenital neutropeniawith evidence of leukemic transformation,” Bone MarrowTransplantation, vol. 35, no. 5, pp. 473–477, 2005.

[38] L. Burroughs, A. Woolfrey, and A. Shimamura, “Shwachman-Diamond syndrome: a review of the clinical presentation,molecular pathogenesis,” Hematology/Oncology Clinics ofNorth America, vol. 23, no. 2, pp. 233–248, 2009.

[39] Y. Dror and M. H. Freedman, “Shwachman-Diamond syn-drome: an inherited preleukemic bone marrow failure dis-order with aberrant hematopoietic progenitors and faultymarrow microenvironment,” Blood, vol. 94, no. 9, pp. 3048–3054, 1999.

[40] J. Donadieu, T. Leblanc, B. B. Meunier et al., “Analysisof risk factors for myelodysplasias, leukemias and deathfrom infection among patients with congenital neutropenia.Experience of the French Severe Chronic Neutropenia StudyGroup,” Haematologica, vol. 90, no. 1, pp. 45–53, 2005.

[41] J. Donadieu, G. Michel, E. Merlin et al., “Hematopoietic stemcell transplantation for Shwachman-Diamond syndrome:experience of the French neutropenia registry,” Bone MarrowTransplantation, vol. 36, no. 9, pp. 787–792, 2005.

[42] J. Fleitz, S. Rumelhart, F. Goldman et al., “Successful allo-geneic hematopoietic stem cell transplantation (HSCT) forShwachman-Diamond syndrome,” Bone Marrow Transplanta-tion, vol. 29, no. 1, pp. 75–79, 2002.

[43] R. Vibhakar, M. Radhi, S. Rumelhart, D. Tatman, and F.Goldman, “Successful unrelated umbilical cord blood trans-plantation in children with Shwachman-Diamond syndrome,”Bone Marrow Transplantation, vol. 36, no. 10, pp. 855–861,2005.

[44] M. Sauer, C. Zeidler, B. Meissner et al., “Substitution ofcyclophosphamide and busulfan by fludarabine, treosulfanand melphalan in a preparative regimen for children and ado-lescents with Shwachman-Diamond syndrome,” Bone MarrowTransplantation, vol. 39, no. 3, pp. 143–147, 2007.

[45] D. Bhatla, S. M. Davies, S. Shenoy et al., “Reduced-intensityconditioning is effective and safe for transplantation ofpatients with Shwachman-Diamond syndrome,” Bone MarrowTransplantation, vol. 42, no. 3, pp. 159–165, 2008.

[46] A. Tichelli, A. Rovo, and A. Gratwohl, “Late pulmonary, car-diovascular, and renal complications after hematopoietic stemcell transplantation and recommended screening practices,”Hematology, pp. 125–133, 2008.

[47] Y. R. Abou-Mourad, B. C. Lau, M. J. Barnett et al., “Long-term outcome after allo-SCT: close follow-up on a largecohort treated with myeloablative regimens,” Bone MarrowTransplantation, vol. 45, no. 2, pp. 295–302, 2010.

[48] U. Forinder, C. Lof, and J. Winiarski, “Quality of life andhealth in children following allogeneic SCT,” Bone MarrowTransplantation, vol. 36, no. 2, pp. 171–176, 2005.

[49] C. M. Lof, U. Forinder, and J. Winiarski, “Risk factors for lowerhealth-related QoL after allogeneic stem cell transplantationin children,” Pediatric Transplantation, vol. 11, no. 2, pp. 145–151, 2007.

Hindawi Publishing CorporationBone Marrow ResearchVolume 2011, Article ID 207326, 14 pagesdoi:10.1155/2011/207326

Review Article

Bone Marrow Stem Cell Derived ParacrineFactors for Regenerative Medicine:Current Perspectives and Therapeutic Potential

Tom J. Burdon,1 Arghya Paul,1 Nicolas Noiseux,2 Satya Prakash,1

and Dominique Shum-Tim3, 4

1 Biomedical Technology and Cell Therapy Research Laboratory, Department of Biomedical Engineering, Faculty of Medicine,Duff Medical Building, 3775 University Street, McGill University, QC, Canada H3A 2B4

2 Department de Chirurgie Cardiaque, Centre Hospitalier de l’Universite de Montreal, Pavillon Hotel-Dieu,Universite de Montreal, Montreal, QC, Canada H3C 3J7

3 Divisions of Cardiac Surgery and Surgical Research, The Montreal General Hospital, McGill University Health Center,QC, Canada H3G 1A4

4 The Royal Victoria Hospital, Suite S8.73.B, 687 Pine Avenue West, Montreal, QC, Canada H3A 1A1

Correspondence should be addressed to Dominique Shum-Tim, [email protected]

Received 31 August 2010; Accepted 12 October 2010

Academic Editor: Helen A. Papadaki

Copyright © 2011 Tom J. Burdon et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

During the past several years, there has been intense research in the field of bone marrow-derived stem cell (BMSC) therapy tofacilitate its translation into clinical setting. Although a lot has been accomplished, plenty of challenges lie ahead. Furthermore,there is a growing body of evidence showing that administration of BMSC-derived conditioned media (BMSC-CM) canrecapitulate the beneficial effects observed after stem cell therapy. BMSCs produce a wide range of cytokines and chemokines thathave, until now, shown extensive therapeutic potential. These paracrine mechanisms could be as diverse as stimulating receptor-mediated survival pathways, inducing stem cell homing and differentiation or regulating the anti-inflammatory effects in woundedareas. The current review reflects the rapid shift of interest from BMSC to BMSC-CM to alleviate many logistical and technicalissues regarding cell therapy and evaluates its future potential as an effective regenerative therapy.

1. Introduction

The objective of stem cell regenerative therapy is to treatdamaged organ tissues by avoiding the processes of cell deathand/or inadvertent remodeled Tissue [1]. Great optimismhas resulted from bone marrow derived stem cell (BMSC)research ever since it showed to contribute significantly tothe reestablishment of some functionality in injured organs[2, 3]. The mechanisms by which stem cells function andreverse the effects of cell death include differentiation, cellfusion, and secretion of cytokines or paracrine effects [1, 4–6]. More specifically, studies injecting BMSCs have shownto improve functionality of ischemic tissue by promot-ing neovascularization, inhibition of apoptosis and anti-inflammation, better localization and homing of therapeutic

cells, and stimulation of endogenous cells differentiationand proliferation [7–10]. Although a lot of research hasbeen focused on the ability of stem cells to differentiatewithin the injured areas, more recent research suggests othermechanisms may be more therapeutically relevant. It will beargued that understanding paracrine mechanisms, mediatedby stem cells, is essential if stem cell regenerative therapy isever to reach clinical importance.

Indeed, understanding the therapeutic effects of regener-ative therapy using BMSCs becomes more relevant when welook at the paracrine factors, which are secreted by BMSCs.For example, the frequency of stem cell engraftment andthe number of newly generated cardiomyocytes or vascularcells are too insignificant to represent the remarkablecardiac functional improvement attributed to fusion or

2 Bone Marrow Research

differentiation alone [11]. In addition, in vivo transplantedcells are exposed to local immune cells and soluble media-tors, which influence the cells behavior in an unpredictablemanner in the microenvironment. Thus, it is necessary tofurther understand the potential benefits of maximizingthe paracrine effects for regenerative therapy. This reviewwill take an in-depth look at specific mechanisms regulatedby these factors and potential therapeutic applications ofBMSC-CM and paracrine factors secreted by BMSCs.

BMSCs include many populations of progenitor cells:hematopoietic stem cells (HSC), mesenchymal stem cells orstromal cells (MSC), side population cells, and multipotentadult progenitor cells [12]. BMSCs can be aspirated, and theentire mononuclear cell fraction containing a heterogeneousmix of progenitor and inflammatory cells is obtainedthrough density-gradient centrifugation using Ficoll. MSCs,which are commonly used in the lab, are present at aconcentration several folds lower than their hematopoieticcounterparts, representing approximately 0.01% of the totalnucleated marrow cell population. They are separated fromother cells in culture by their preferential attachment toplastic surfaces [13–16]. MSCs do not express hematopoieticor endothelial cell surface markers. MSCs are expandablein culture without losing their differentiation potentialand constitute an unlimited pool of transplantable cells.They are multipotent and can differentiate into multi-ple lineages, including fibroblasts, osteoblasts, chondrob-lasts, and adipocytes [17–23]. Differentiation of MSCs tocardiomyocyte-like cells has been observed in vitro underspecific conditions and in vivo after injection into themyocardium [24–27].

2. Emerging Role of BMSCs for Cell andTissue Regeneration Therapy

MSCs are particularly suitable for cell therapy because of easyisolation, high expansion potential giving unlimited pool oftransplantable cells, low immunogenicity, amenability to exvivo genetic modification, and multipotency [24, 28, 29].Although MSCs undergo lineage-specific differentiation togenerate bone, fat, and cartilage, they have been reported totransdifferentiate into defined ectodermal and endodermaltissues [30]. Furthermore, MSCs are available for autologoustherapies, can bypass immune rejection, and are inherentlymigratory. Differentiation of MSCs into cells expressingcardiomyocytes markers has been obtained in vitro and invivo [26, 27, 29, 31–36]. They are also known to secrete avariety of biologically active factors and promote collateralblood flow development through paracrine mechanisms[37–44]. Moreover, bone marrow stromal cells are capableof differentiation, regeneration of infarcted myocardium,induction of myogenesis, and promotion of angiogenesis.These cells can potentially differentiate into cardiomyocytesin vivo and even express functional adrenergic and mus-carinic receptors [45, 46]. Moreover, conditioned mediumcollected from MSC (MSC-CM) promotes in vitro prolifer-ation and migration of endothelial cells and vascular smoothmuscle cells, and enhances blood flow recovery of ischemic

hindlimb [37, 43, 44]. Following exposure to hypoxia andserum starvation, MSCs are stimulated to secrete severalgrowth factors and cytokines [47]. Noiseux et al. have shownthat injection of MSC-CM either directly into infarcted heartby intramyocardial or intraperitoneal injections improvemyocardial function and repair [48–56]. The mechanism bywhich MSCs may exhort beneficial effects is debated but iscertainly synergistic [57].

Although the mechanisms and the therapeutic benefitshave not yet been fully elucidated, general consensus statesthat adult stem cells, particularly BMSCs, are very safe.This is based on their usage in human studies, which haveshown that they can be safely cultured in vivo with little riskof malignant transformations [6]. Pitfalls to be consideredwhen working with MSCs are that transplantation of MSCsalone can generate local immune responses and disrupthomeostasis within tissue milieu by causing the release ofinflammatory mediators such as cytokines [32]. Moreover,the therapeutic ability of BMSCs to suppress immuneresponses may indirectly promote metastatic tumour pro-liferation and growth. In addition, cytogenic abnormalities,such as tumour differentiation, have been observed inMSCs which have undergone extensive culturing and long-term passaging [6]. Another area of concern involvingBMSCs was unveiled when a group recently discoveredthat transplanted BMSCs in ischemic rat hearts led tocalcifications/ossifications [58]. This latter group stresses theimportance of long-term studies, which will help determinethe provenance, multipotency, and long-term fate in vivo.However, the same group discovered that use of cytokinesderived from the BMSCs did not lead to these pathologicalmalformations. Therefore, more data describing the long-term effects of administered MSCs and their factors areneeded to be established. In addition, since BMSCs can beobtained directly from adult animal models, there are nopressing ethical issues limiting their usage in clinical trials.

Although there are many benefits attributed to MSCs, itis important to know that the unique properties of MSCs donot guarantee that they will accomplish desired therapeuticoutcomes or that they will evade rejection. This is thereason for the various conflicting results with regards to themechanisms and outcomes of MSCs found in the literature.Hence, current research is highly focused on developing anefficient delivery system that can successfully transplant asufficient number of stem cells at the desired site with higherretention and minimal biological loss or immune response.Some of our more recent work has shown that retention canbe increased by four folds in the heart by using biocompatiblepolymeric alginate microcapsules as a delivery system, whichcan, in turn, help in functional recovery of acutely infarctedheart [59–61].

3. Exploiting the Cell Fusion andDifferentiation Properties of BMSCs:Potentials and Limitations

Since heart disease is reaching pandemic proportions indeveloped and developing countries and the heart is an

Bone Marrow Research 3

20 μm

H9c2 and MSCs

(a)

20 μm

MitoTracker Red

Nanotubular

Network

(b)

Figure 1: Formation of intercellular connections after oxygen glucose deprivation. (a) Nanotubular network formation was observed amongDiO-labeled cardiomyoblasts (green) and DiD-labeled MSCs (red) after 24 hours of coculture. (b) MitoTracker staining (red) revealed activemitochondria in the nanotubular networks [5].

organ of limited regenerative capacity, much research usingstem cell therapy is focused on the regeneration of themyocardium. Therefore, studies focusing on stem cell dif-ferentiation and cell fusion are often looking at the heart.In earlier studies concerning cardiac regeneration, successfultransplantation of neonatal and fetal cardiomyocytes intorats with acute myocardial infarction was observed [62]. Theresults showed improved left ventricular function attributedto the formation of stable intracardiac grafts [63–67]. Theimplanted cells also retained their contractile phenotypeand expressed necessary elements for intercellular electricalcommunication.

Oric et al. demonstrated that bone marrow could bea source of extracardiac source of progenitor cells withthe ability to differentiate into cardiomyocytes and restorecardiac function [9, 68, 69]. Since then, cardiogenic capa-bilities have been confirmed in hematapoietic stem cells,MSCs, circulating endothelial progenitor cells, and residentcardiac progenitor cells [11, 70–74]. Moreover, when theBMSC are transplanted into ischemic heart, they expressthe cardiac specific markers troponin I and cardiac myosin,allowing researchers to easily identify the transformationinto functional cardiomyocytes and shows that histologicaldifference is indeed occurring [71, 75, 76].

Cell Fusion of transplanted stem cells with residentcardiomyocytes has been proven as a feasible mechanismfor differentiation. Fusion of bone marrow-derived cellswith purkinje neurons, hepatocytes and cardiomyocytes wasreported for the first time by Alvarez-Dolado et al. [77].During cell, fusion the cells connect and exchange vitalcell components [4, 78]. Cselenyak and others performedan elaborate in vitro experiment suggesting the presence ofcellular nanotubes which increased the viability of ischemiccardiomyoblasts in the presence of normal MSCs [5]. Itreveals that the viability of H9c2 rat cardiomyoblast cellline depends on the nanotubular connection with ratMSCs because when the nanotubes were blocked with cell

culture inserts, the H9c2 viability decreased significantly.Figure 1 shows cardiomyocytes and human mesenchymalcells communicating through small diameter nanotubes,which allow for the migration of organelles such as themitochondria from MSCs to cardiomyocytes. The graphin Figure 2 illustrates the antiapoptotic effect of MSCs onstressed cardiomyocytes and how the nanotubular networksare crucial to their effects.

In a separate experiment, Noiseux et al. demonstrated,using a model of acute MI in transgenic mice, that trans-planted MSCs could fuse with recipient cardiomyocytes inthe infarcted area [79]. MSCs from wild-type C57BL/6 micewere retrovirally transduced to express green fluorescent pro-tein (GFP) (reporter gene) and Cre recombinase. These MSCwere transplanted into infarcted heart of histocompatibleR26R mice. In these transgenic mice, a loxP-flanked stopsequence was present 5′ of the LacZ expression cassette toprevent transcriptional read through until selective excisionby Cre-mediated recombination from implanted MSC. Thus,the LacZ gene was expressed exclusively after a donor MSCexpressing Cre fused with a recipient cell from the R26Rmice. Consequently, X-gal staining is used to detect cellularfusion events [77, 80]. This useful method, based on Cre/loxrecombination that conditionally turns on the LacZ geneonly in the fused cells, allows experimenters to rapidlyobserve tissue sections covering a large area of the infarctedand MSC-injected heart. The cells that were transplantedwere found within the infarcted myocardium (detection ofMSC by GFP immunoreactivity); early massive engraftmentwas observed at 3 days, but the number of implanted MSCdecreased significantly over time, and by day 28 post-MIvery few cells remained. As early as 3 days following MSCinjection into ischemic heart, they observed cellular fusionwith individual blue cells having typical cardiomyocytesmorphology. Interestingly, the majority of the fusion eventswith LacZ+ cells were detected within the infarct borderzone, in areas with viable cardiomyocytes. Although cell

4 Bone Marrow Research

0102030405060708090

100

Dea

dH

9c2

cells

(%)

C C + MSC C + MSC ins

Figure 2: Cocultivation of cardiomyoblast (H9c2) cells with MSCsdecreased cell death in oxygen glucose-deprived environment.(C): ratio of dead H9c2 cells after oxygen glucose deprivation.(C + MSC): ratio of dead H9c2 cells after co-cultivation with MSCsand under oxygen glucose deprivation (85 ± 8.6 versus 16 ± 3.5,n = 5). (C + MSC ins): ratio of dead H9c2 cells when MSCs wereadded with cell culture inserts (90±5.5, n = 5). Data represent mean± Standard deviation. ∗P < .05 C + MSC versus C and C + MSCversus C + MSC ins [5].

fusion was shown to occur by the appearance of bluecells (fusion of MSC with resident cardiomyocytes) and bythe improved cardiac function, they concluded that fusionwas not responsible because it happened so infrequently.Therefore, they attributed enhanced cardiac function toparacrine effects [79].

Despite hopeful results with experiments looking atfusion and differentiation, the mechanisms by which BMSCswork to help damaged tissue remains unclear even in seem-ingly successful studies. Furthermore, inconsistent resultswhen researching stem cell regenerative capacity have beenlinked to improper labeling of donor cells and inabilityto consistently track them in vivo, making it very difficultto distinguish them from background tissue leading tomisinterpretation [81]. For example, the utilization of theGFP reporter gene is attractive because it is compatible witha variety of imaging techniques; however dead and dying car-diomyocytes have an autofluorescent spectrum that partiallyoverlaps with that of GFP. Therefore, after injury, autoflu-orescence increases due to accumulated lipofuscin, blood-derived pigments, and other intrinsic fluors such as flavinsand reduced nicotinamide adenine dinucleotide (NADH)[82]. Evidence for regeneration includes colocalization ofGFP fluorescence from donor cells, with immunostainingfor cardiomyocyte markers, including sarcomeric actin cells.One major limitation of cell therapy is the low therapeuticefficacy of transdifferentiated cells in the site of injury.The low rate of sustained transplanted engrafted stem cellshas been attributed to cell stress, hostile environment dueto inflammation and hypoxia, insufficient blood supply,and elaboration of inflammatory cytokines resulting fromischemia and cell death [83–86]. In addition, overdiploidDNA fusion events between embryonic stem cells andBMSCs can result in cell fusion with 2 nuclei (2n + 2n), oralternatively nuclear fusion will result in cells with largernuclei 4n chromosomes going up to hexaploid (6n) DNAcontent. This leads to hybrid genotypes with cells failing to

express proper green fluorescent phenotypes adding to theconfusion of tracking differentiated cells [87].

4. Exploiting the Paracrine FactorsReleased from BMSCs

Since research has shown that stem cell differentiation andfusion alone cannot account for enhanced cardiac function,evidence for a nonmyogenic pathway of cardiac repair isbecoming increasingly explored. The debate surroundingfusion and differentiation is of little importance nowbecause the number of reported cardiomyocytes derivedfrom exogenous stem cells is too low to account for theimpressive enhancement of physiological function [62, 88].Thus, it has been proposed that stem cell transplantationpossesses therapeutic effect because of the endogenousrepair mechanisms secreted by the BMSCs that account forfunctional benefits of cell therapy in cases such as cardiacrepair.

MSCs such as BMSCs are known to secrete manybiological factors, and under strenuous hypoxic and serumstarvation, the expression of several of these factors isupregulated. This has prompted the isolation of MSCconditioned medium (MSC-CM) to be concentrated andused therapeutically. Kinnaird et al. reported the presence ofcytokines as: VEGF, FGF-2, IL-6, PIGF, and MCP-1 and thatintramyocardial injection of MSC-CM improved collateralblood flow recovery and limb function and reduced muscleatrophy. Because all these cytokines are known to promoteangiogenesis and vascularization, the conclusion was thatit was not direct cell incorporation but rather paracrinesignaling that may be responsible for the effects of bonemarrow cell therapy in ischemic injury.

It was previously reported that stem cell transplantationinto the ischemic myocardium improved cardiac functionrecovery and repair as early as 72 hours later, which is tooearly to be explained solely by MSC differentiation [55, 56,79]. To see how the MSCs confer protection of the ischemicmyocardium through paracrine mediators and indirecteffects, it was demonstrated that MSC-CM reduced hypoxia-induced apoptosis and triggered spontaneous contraction ofisolated hypoxic adult rat cardiomyocytes in vitro, suggestingthe presence of antiapoptotic and inotropic effects [55].When injected directly into infarcted rat hearts, the MSC-CM limited infarct size as early as 72 hours and improvedventricular function at levels comparable to those observedfollowing MSC transplantation. From this data, it is clearthat the therapeutic effects observed after intramyocardialinjection of MSC in the infarcted hearts are, at least, inpart attributable to the paracrine effects on the ischemicmyocardium. Paracrine factors can also play an importantrole in influencing the adjacent cells and exert their actionsvia diverse mechanisms. Furthermore, it is likely that theparacrine mediators are expressed in a specific temporaland spatial manner that can enhance cell survival andactivate endogenous mechanisms of endogenous repair andregeneration [11]. Figure 3 provides a simplistic overviewof the observed therapeutic potentials that arise from

Bone Marrow Research 5

Cardiomyocyte

Cardiomyocyte

Release of

paracrine factors

Proliferation

Recruitment of resident

stem cells

Increases vascularization

Reduces cardiomyocyte

Increases cardiomyocyteproliferation

Cell fusion

Better perfusion

Increase myocyteNumbers Decreases myocardial dysfunction

Transdifferentiation

Control granulation factors and

scar composition in matrix

Angina

Heart failure

Stem cells

apoptosis

Figure 3: Schematic representation of the hypotheses presenting therapeutic effects of stem cell transplantation for myocardial regeneration.Cell transplantation can improve tissue perfusion and contractile performance by promoting formation of blood vessels and myocyteformation/protection. Central to the beneficial effects of cellular therapy is the paracrine and indirect effects of stem cells with the productionand release of cytokines and growth factors. Depending on the stem cell type and local milieu, the relative contribution of cell incorporation(transdifferentiation and/or fusion) versus paracrine effects may vary [27].

BMSC cardiovascular therapy and illustrates the direct andindirect influences of paracrine factors on overall cardiacfunctionality. Furthermore, Table 1 gives a brief overview ofspecific identified factors secreted by human MSCs detectedby protein array [52].

4.1. Paracrine-Mediated Neovascularization. Blood vesselformation after birth proceeds primarily through 2 mech-anisms: angiogenesis and arteriogenesis. Angiogenesis con-sists of several distinct processes which include sproutingand proliferation of pre-existing capillaries to form newnetworks. This process is tightly regulated by hypoxiathrough a number of proangiogenic factors, including VEGF,FGFs, P1GF, and hepatocyte growth factor (HGF) [89–94].The rapid proliferation of pre-existing collateral arteriescharacterized by arteriogenesis involves remodeling small

arterioles into larger vessels and is triggered in part byan increase in stress in arterioles that run parallel to theoccluded artery. Recruitment of monocytes that differentiateinto macrophages and produce abundant angiogenic growthfactors such as VEGF, NO, MCP-1, and other cytokines,is also essential and ultimately leads to endothelial andsmooth muscle proliferation, migration, vessel remodeling,and extracellular matrix synthesis [44, 94]. Arteriogenesisis the most efficient adaptive mechanism for the survivalof ischemic organs because of its unique capability to con-duct large blood flow. Angiogenesis and arteriogenesis arecomplex processes sharing common mechanisms of action,growth factors, and cytokine dependency. Furthermore,these cytokines act not only in a coordinated time andconcentration-dependent manner, but one cytokine mayinhibit or stimulate the effect of another. The complexity of

6 Bone Marrow Research

Table 1: Detection of secreted factors from human mesenchymal cells under normoxic and hypoxic conditions. Relative concentrationbetween normoxic and hypoxic conditions are expressed from 1–5 with 1 being the lowest concentration and 5 being the highest [52].

Secreted factors Normoxia Hypoxia Biological function

Activin A 3 2 Cell proliferation, differentiation, apoptosis, and immune response

Epiregulin 3 3 Remodeling

Endothelin 4 4 Cytoprotection, cell proliferation

Glypican-3 3 3 Cell proliferation

IGFBP-7 4 5 Cytoprotection, cell migration, and contractility

IL-15 alpha 2 1 Immune response

LRP-1 2 1 Cell migration

LRP-6 4 1 Cell migration

Osteoprotegrin 4 5 Bone development

sFRP-4 3 4 Development, apoptosis

Smad4 3 2 Vessel maturation, cell proliferation

Smad7 2 2 Vessel maturation, cell proliferation

Thrombospondin-1 3 5 Cell migration, apoptosis

TIMP-1 4 5 Cell migration, remodeling

TIMP-2 4 5 Cell migration, remodeling

VEGF 3 4 Cytoprotection, proliferation, migration, and angiogenesis

the process of collateral formation has led to the suggestionthat multifactorial strategies will be necessary to modulatetherapeutic paracrine effects including vessel formation.

Under normal circumstances, myocardial tissue dies andforms scar tissue because the capillary network does notmeet the demands of the hypertrophied cells. The inevitablelack of oxygen and nutrients leads to apoptosis and necrosis.Moreover, inducing angiogenesis requires the coordinatedinteractions between monocytes/macrophages, endothelialcells, smooth muscle cells, and pericytes [92, 94–99]. Humanbone marrow-derived multipotent stem cells allow forangiogenesis by expressing various angiogenic cytokines suchas VEGF, Ang-1, Ang-2, bFGf, HGF, and PDGF-B. Moreimportantly, HGF, VEGF, bFGF, and IGF are known survivalfactors that are expressed by cardiomyocytes [96, 97]. Theseparacrine factors are believed to influence adjacent cellsvia mechanisms including myocardial protection, neovas-cularization, and most the activation of resident cardiacstem cells and/or stimulation of endogenous cardiomyocytereplication.

MSCs express genes encoding a broad spectrum ofcytokines with angiogenic properties of some of which maybe referred to in Table 1 such as VEGF, HGF, FGF, MCP-1, PGF, IL-1 and IL-6, SDF-1, and MMP-9 [37, 44, 47].All of these cytokines can be isolated from the media ofcultured cells and have all shown to have positive effectson experimental blood flow recovery [37, 91]. In addition,MSC-CM promotes in vitro proliferation and migrationof endothelial cells and vascular smooth muscle cells andenhances in vivo collateral blood flow recovery when injectedinto ischemic hindlimb. Another possibility for improvedfunction is that neovascularization by the stem cells is leadingto enhanced blood supply in the peri-infarct region, therebypromoting salvage of stunned, hibernating, or otherwisesusceptible cardiomyocytes [62].

4.2. Paracrine-Mediated Cell Homing/Targeting. Homing sig-nals are extremely important for the efficacy of cell therapybecause it is via these paracrine effects that the preciselocalization of transplanted cells is possible and can beimproved. In the case of cardiac repair, stem cell therapycan be administered via intravenous, intracoronary, trans-myocardial, catheter-based transendocardial with or withoutelectromechanical voltage mapping, and finally transvenousinjection into coronary veins [100–104]. The goal of anycell delivery strategy is to transplant enough cells into thearea of interest to achieve maximum retention in that area.Moreover, the local environment determines the success ofcell delivery since it is the milieu that will influence short-term cell survival, cell properties in regard to adhesion,transmigration through vasculature, and tissue invasion.Therefore, the targeted administration of cells is preferred[105]. Cell homing, transmigration, adhesion, and tissueinvasion are the result of many complex steps. For example,stromal-derived factor-1 (SDF-1) is expressed in the ischemicmyocardium and plays an important role in endogenousstem cell migration, adhesion, homing, and recruitmentfrom bone marrow. SDF-1 is an important chemoattractantfor progenitor cells and thus plays a crucial role in directingcells into areas of infarcted myocardium [75]. Tang andcolleagues administered SDF-1 expressing plasmid intoischemic border zones of murine models injected with BMCs(Lin−c-kit+) [106]. A significant amount of the labeled stemcells were found in the area where the plasmid was injected.Thus, administering homologous cytokines is one way toencourage the proliferation of injected stem cells.

Of all techniques, the intracoronary infusion has beendeemed best for treatment of acute myocardial infarction.However, strategies to augment cell function via betterlocalization/homing is crucial if there is to be any therapeuticbenefit using BMSCs for targeted regenerative therapy.

Bone Marrow Research 7

Additional strategies for better homing of mesenchymal cellsinclude inducing hypoxia for the upregulation of CXCR-4 protein receptor expression at the BMSC cell surface.CXCR-4 are chemokine protein receptors expressed in dyingcells, which can be complexed with SDF-1 protein ligands[107, 108]. The CXCR-4/SDF-1 complex allows for stem cellmigration via chemotaxis where expression of the CXCR-4receptor and the presence of SDF-1 receptor are required toregulate and make cell migration possible. It has been foundthat CXCR-4 expression can be induced by exposure to apool of cytokines such as SCF, IL-6, Flt- ligand, HGF, andIL-3. Furthermore, increasing affinity of the stem cells canalso be done by increasing the concentration of the chemo-attractant SDF-1. Therefore, a depot of angiogenic cytokinesand an increase in SDF-1 ligand can possibly be localized inan area, allowing for efficient stem cell homing [109].

Orlic et al. reported that injection of BMCs (Lin−c-kit+)was able to give rise to new myocardium and improve leftventricle end diastolic pressure by 30% to 40% in mousehearts with infarcts located in the free wall of the leftventricle [1, 110]. In a separate but similar experiment,injections of the BMSC cytokines stem cell factor (SCF) andgranulocyte-colony stimulating factor (G-CSF) increasednumbers of circulating HSCs to a level 250-fold greater thanuntreated myocardium. However, it is not clear whethertissue repair is a result of the homing of BMC to thelesion or if BMC, once mobilized, cells spreads throughthe organism and only those exposed to damaged tissueare triggered to proliferate and differentiate at a faster ratethan the others [9]. Nonetheless, the former hypothesis issupported because of the rapid induction of SCF in manytissues including the myocardium. Therefore, SCF could beresponsible for migration, accumulation, and multiplicationof primitive BMCs in the infarcted zone where they acquirethe appropriate phenotype.

4.3. Paracrine-Mediated Anti-Inflammatory Effects. MSCsare known for having potent anti-inflammatory activities,which are present regardless of their tissue of origin. Theyare said to suppress inflammation through secretion ofanti-inflammatory cytokines such as IL-10, TGF-beta,LIF, soluble HLA-G, and IL-1 receptor antagonists [111].The successful effects of MSC have been demonstrated invivo and have treated animal models in cases of immunemediate/inflammatory pathologies such as MS, colitis,graft versus host, rheumoid arthritis, and ischemic injury[112]. Most recently, Chen et al. have shown that MSCsimplantation after myocardial infarction in a rat modelcaused a significant upregulation of anti-inflammatoryversus the proinflammatory cytokines compared withthe control group without cell therapy [113]. It istherefore conceivable that MSC factors can also workby downregulating the genes, which promote inflammationand therefore are useful against heart failure by suppressingongoing self-perpetuating inflammatory cascades.

As it has been aforementioned, knowing the microen-vironment prior to cell transplantation or conditionedmedia transplantation is crucial to avoid any immune-

mediated rejection. That being said, inclusion of specificcytokine receptor antagonists or inhibitors can prevent theimmune-mediated responses of the host microenvironmenton transplanted cells. Since IL-1α/β, TNF-α, and SDF-1-αare ubiquitous in the cell microenvironment, we must reviewtheir mechanisms. First of all, these are pro-inflammatorymediators, which are primarily synthesized by macrophages,monocytes, and dendritic cells, and are responsible forimmune defense against infection in most tissues [114].Greco and Rameshwar have shown that MSCs express IL-1RI, which is the principal receptor for IL-1α and IL-1β[32]. The researchers found that membrane expression ofthe receptor was maintained throughout transdifferentia-tion of MSCs into neurons. Moreover, IL-1α was foundto alter behavior of undifferentiated MSCs. Stimulationof MSCs with IL-1α caused production of substance Pneurotransmitter by undifferentiated and differentiated cells.Excessive production of substance P neurotransmitter leadsto exacerbated immune responses and untoward effect onMSC differentiation. On the other hand, delivering MSCstogether with inflammatory cytokine antagonist/inhibitormay suppress immunoreactivity allowing for the expectedtherapeutic result [115]. Therefore, knowing the correctinhibitors will assure optimal functioning of transplantedcells or media.

As demonstrated in vitro and in vivo, it is now knownthat BMSCs can be immunosuppressive and escape cyto-toxic lymphocytes [116, 117]. One mechanism by whichimmunosuppressive effects are mediated is by Fas-mediatedapoptosis. Cells expressing FasL specifically protect againstT-cell-mediated cytotoxicity where T cells expressing Fas aresensitive to Fas-mediated apoptosis. One study hypothesizedthat this immunosuppressive action is mediated by FasL-induced killing of activated lymphocytes, thereby, preventinglymphocyte attack. It has been suggested that cell engraft-ment can be improved by the pretreatment of cells with anti-apoptotic or “cytoprotective” genes; some of which can befound in stem cell media (Akt, Bcl2 GSK, IlK, telomerase,and eNOS) [118].

4.4. Paracrine-Mediated Endogenous Cell Stimulation. Acti-vating specific cells through paracrine/endogenous effectsof growth factors released by BMSC is showing promise.The kidney, similar to the heart, has a limited regenerativecapacity. However, recent studies have suggested the presenceof stem cell renal niches, that is, renal papillae in animalsand in the urinary pole of Bowman’s capsule in humans.The importance of these CD24+ and CD133+ stem cellsduring kidney damage has not yet been elucidated, andit has been found that MSCs contribute to renal repair[119]. Similarly to cardiac treatments, when treating thekidney with transplanted MSCs, several studies documentedrenal recovery without differentiation of MSCs into epithelialcells. Therefore, it was also suggested that cell migra-tion facilitated regeneration because of endocrine/paracrineeffects and that it was the kidney stem cells that were re-generating the tubular epithelium [120–125]. Furthermore,the body also possesses local source of cardiac stem cells,

8 Bone Marrow Research

which can differentiate into myocytes and possess growthfactor-receptor systems allowing for the regeneration ofinfarcted myocardium. It is known that HSCs and MSCscan stimulate activation of these cardiac resident stem cellsbecause they can produce HGF and IGF-1 which are notonly anti-apoptotic but can activate endogenous stem cells[126]. Interestingly, these factors are upregulated in responseto inflammatory mediators such as TNF-alpha. Therefore,when responding to inflammatory stimuli, MSCs and HSCscan, in addition, potentially recruit endogenous stem cells[109].

5. Recent Applications ofStem Cell-Conditioned Media inthe Biomedical Fields

5.1. Cardiac Regeneration. Although this is a novel topic,some studies have been done using BMSC-derived con-ditioned media (BMSC-CM). Their preliminary resultssuggested that much more efforts should go into furtherresearching the benefits of conditioned media to betterunderstand the role of individual factors that may opti-mize their effects. Interest in this topic started whenresearchers reported functional improvement of tissue beforethe possibility of attributing it to donor cells. Gnecchiand colleagues found significant functional improvementin the myocardium occurring in less than 72 hours afterintramyocardial injection of Akt-modified MSCs in a ratmodel [56]. The same group later observed that cell fusionevents were occurring and were associated with bettercardiac performance. However, the cell fusion and differen-tiation events were so uncommon that they attributed theenhanced cardiac function to paracrine factors expressedby MSCs [48, 79]. Nguyen et al. are the first to researchthe effects of MSC-CM on the myocardium [52]. Aftersubjecting swine to acute MI by balloon occlusion ofthe left anterior descending coronary artery, the effectsof intracoronary injections of either concentrated MSC-derived growth factors or control medium were observed.MSC-derived factors significantly reduced serum cardiactroponin T (cTnT) that has been validated as a biomarkerthat increases with the severity of injury. The treatmentgroup also showed significant echocardiographic functionalimprovement. Staining analysis also revealed a reductionof the fibrotic area in the infarct zone and incrementof preserved area in the infarcted myocardium. This wasassociated with less apoptosis of cardiomyocytes. The roles ofangiogenic (vascular endothelial growth factor, endothelin,and epiregulin), anti-apoptotic (Galectin-3, Smad-5, sRFP-1, and sRFP-4) and anti-remodeling factors have all beendetected by protein array in the control group.

Recent research has also shown that there is an optimaltime for conditioned media delivery since the paracrineeffects are known to be time dependent. It has beendemonstrated that the prohypertrophic effects on myocytesand proliferation of fibroblasts in rat myocardium weresignificant using media collected after 6 and 9 hours ofsecretion. On the other hand, medium collected at 3 and

24 hours after secretion resulted in no noticeable prohy-pertrophic effects. Consequently, the presence of FGF-2, IL-1, IL-6, TGF-β and TNF-α factors during these optimaltimes suggests that they all take part in hypertrophy andproliferation of the cardiomyocytes [127].

5.2. Renal Regeneration. Similarly to stem cell therapy inmyocardial regeneration, studies in the kidney providedno evidence that marrow stromal cells were incorporatinginto the native renal parenchyma in significant numbers ordirectly contributed to tubular cell repopulation [120]. Biand colleagues studied whether BMSC-CM injections hadany beneficial effects on the renal system. Using in vitromouse proximal tube cells (MPT) cultured (24 hours) withcisplatin with and without BMSC-CM, they showed thatsustained exposure to toxic cisplatin resulted in detachmentand death of many MPTs. The presence of BMSC-CM, how-ever, significantly inhibited cell detachment and inhibited celldeath after 48 h. The same study investigated the mechanismsof renal repair after BMSCs injection. They showed noevidence of renal cell regeneration but rather the endocrineeffects conferred the greatest therapeutic benefit for the cells[125].

5.3. Neurotrophic Actions of BMSC-CM. Bone marrow stro-mal cells also showed great potential with regards toneurological regenerative therapy. BMSCs have shown tocontribute to functional recovery and tissue regenerationin rat models with spinal cord injury, cerebral ischemia,traumatic brain injury, and constructing tissue-engineerednerve grafts [128–130]. Researchers have investigated theparacrine factors released by BMSC-CM using rat dorsal rootganglion cells. They cocultured BMSC-CM with explantsor neurons, and cell proliferation assays showed that neu-rite outgrowth and neuronal cell survival were enhanced.Because they used transwell inserts and BMSC-CM, the cellbehaviors of explants and neurons were induced by solublefactors released by BMSCs into the culture medium. Thefactors contributing to immune cell migration to injurysite, remodeling, and cellular action which all contributeto repairing injured nerves can be attributed to the factorsexerting paracrine factors such as 43 secretory proteins,adhesion molecules, cytokines, and growth factors. Fur-thermore proteomic analysis and ELISA further found thatBMSC-derived soluble factors were involved in activation ofPI3-K/Akt and MAPK signaling pathways in DRG neurons.The activation of PI3-K/Akt and MAPK/Erk1/2 pathwaysis crucial for survival signaling. Therefore, these resultssuggested a new pathway by which cell survival might beattributed to endocrine paracrine mechanisms [131].

6. Future Directions

6.1. Advanced Cell Isolation Methods for Enhancementof Paracrine Activities. Optimizing and maximizing theamount of secreted stem cell paracrine factors is impor-tant to assure its efficient transition in therapeutic andclinical applications. Preparation and isolation of MSC has

Bone Marrow Research 9

traditionally involved plastic adherence isolation. Anothermethod for cell isolation is by immunoselection usingmesenchymal precursor cells (MPC) at higher purity. MSCinvolving plastic adherence (PA) isolation has been shownto be less effective than MPCs immunoselected for stromalprecursor antigen-1 (STRO-1) [132, 133]. Thus STRO-1-MPC displayed greater clonogenecity, proliferative capacity,multilineage differentiation potential, and mRNA expres-sion of mesenchymal stem cell related transcripts. As aresult, STRO-1-MPC exhibited higher gene and proteinexpression and thus greater paracrine activity than PA-MSC in vivo. More importantly, enrichment for STRO-1increases the expression of cardiovascular-relevant cytokinesand enhanced trophic activity. STRO-1-MPC, for example, itexhibited higher expression of CXCL12 and HGF, which arecrucial in endothelial tube formation and cardiac cell pro-liferation. Therefore, immunoselection has great potential inthe field of regenerative therapy.

6.2. Nanobiotechnological Applications. Enhancing effective-ness of paracrine activity has also been attempted usingheparin-binding amphiphile nanoparticles (HBPA) [134].Hypoxic-conditioned medium was derived from MSCswhich led to significantly better haemodynamic performanceat 30 days after infarct in a mouse ischaemia-reperfusionmodel of acute myocardial infarction. Interestingly, enoughMSC-CM-HBPA animals did not vary significantly fromhealthy animals with respect to LV contractility; thus,HBPA nanofibre networks resulted in significantly greatermyocardial functional performance than with HBPA alone orwith conditioned medium alone. The mechanism by whichthe synergistic effects of CM and HBPA functioned waslikely due to the heparin-containing nanofibre gel concen-trating heparin-binding factors, stabilizing factor-receptorcomplexes, prolonging factor activity by suppressive prote-olysis, and increasing retention through the introduction ofmolecular aggregates rather than isolated proteins. The studyalso showed that loading HBPA nanofibre networks withnanogram quantities of growth factors VEGF and bFGF wasenough to recreate the enhanced cardiac function. Howeverthe mechanism by which cardiac function was improved wasnot well described especially because they did not noticeany changes relating to angiogenic factors. Therefore, theypostulated that the HBPA nanofibre in synergy with theconditioned media augmented function via trophic effectswhich has been shown to augment function by modulationof cardiac metabolism. In other words, this new ideaproposes that paracrine factors might help erase ischemicmemory and help the heart utilize the cardiometabolicreserves [135].

7. Conclusion

So far, bone marrow-derived stem cells and their paracrinefactors have shown all the necessary attributes for tissueregeneration, namely, homing, immunosuppression, differ-entiation, angiogenesis, stimulation of endogenous cells,and possible regulation of specific metabolic pathways, only

to name a few. Thus, research into paracrine factors andmechanisms has shown that stem cell therapy is muchmore complicated and greatly enhances the potential andvariety of therapeutic applications. Mediating angiogenicfactors, cell proliferation, and all of the above-mentionedcharacteristics are crucial to stem cell therapy and providesmany new approaches for therapy. Therefore, injecting MSCsinto tissue and observing the extent of differentiation are tobe considered superfluous unless, of course, paracrine effectsare being considered. Results from experiments using onlyconditioned medium showed signs of remarkable promiseand revealed the same regenerative capacities as studies usingcell injection. However the hallmarks of transdifferentiationand cell fusion are not to be disregarded because paracrineactivities will most likely make these processes more efficient.Therefore, research into using conditioned media from stemcells is crucial because it will give us a greater understandinginto the most relevant and important cytokines, theiroptimization, and their mechanisms of action to furthermaximize the application and uses of regenerative stem celltherapy.

Acknowledgments

This work is supported by research grant (to D. Shum-Tim and S. Prakash) from Natural Sciences and EngineeringResearch Council (NSERC), Canada. N. Noiseux is scholarof Fonds de la Recherche en Sante du Quebec (FRSQ) andis supported by the Heart and Stroke Foundation of Quebec,and Department of Surgery, Universite de Montreal. A. Paulacknowledges the financial support from NSERC AlexanderGraham Bell Canada Graduate Scholarship.

References

[1] D. Orlic, J. Kajstura, S. Chimenti, D. M. Bodine, A. Leri,and P. Anversa, “Bone marrow stem cells regenerate infarctedmyocardium,” Pediatric Transplantation, vol. 7, no. 3, pp. 86–88, 2003.

[2] S. Bollini, N. Smart, and P. R. Riley, “Resident cardiacprogenitor cells: at the heart of regeneration,” Journal ofMolecular and Cellular Cardiology. In press.

[3] J. M. Hare and S. V. Chaparro, “Cardiac regeneration andstem cell therapy,” Current Opinion in Organ Transplantation,vol. 13, no. 5, pp. 536–542, 2008.

[4] E. Y. Plotnikov, T. G. Khryapenkova, A. K. Vasileva etal., “Cell-to-cell cross-talk between mesenchymal stem cellsand cardiomyocytes in co-culture,” Journal of Cellular andMolecular Medicine, vol. 12, no. 5, pp. 1622–1631, 2008.

[5] A. Cselenyak, E. Pankotai, E. M. Horvath, L. Kiss, and Z.Lacza, “Mesenchymal stem cells rescue cardiomyoblasts fromcell death in an in vitro ischemia model via direct cell-to-cellconnections,” BMC Cell Biology, vol. 11, article no. 29, 2010.

[6] H. Asanuma, D. R. Meldrum, and K. K. Meldrum, “Ther-apeutic applications of mesenchymal stem cells to repairkidney injury,” Journal of Urology, vol. 184, no. 1, pp. 26–33,2010.

[7] P. R. Crisostomo, M. Wang, C. M. Herring et al., “Gen-der differences in injury induced mesenchymal stem cell

10 Bone Marrow Research

apoptosis and VEGF, TNF, IL-6 expression: role of the 55 kDaTNF receptor (TNFR1),” Journal of Molecular and CellularCardiology, vol. 42, no. 1, pp. 142–149, 2007.

[8] S. M. Nassiri, Z. Khaki, M. Soleimani et al., “The similareffect of transplantation of marrow-derived mesenchymalstem cells with or without prior differentiation induction inexperimental myocardial infarction,” Journal of BiomedicalScience, vol. 14, no. 6, pp. 745–755, 2007.

[9] D. Orlic, J. Kajstura, S. Chimenti et al., “Mobilized bonemarrow cells repair the infarcted heart, improving functionand survival,” Proceedings of the National Academy of Sciencesof the United States of America, vol. 98, no. 18, pp. 10344–10349, 2001.

[10] A. Shabbir, D. Zisa, G. Suzuki, and T. Lee, “Heart failuretherapy mediated by the trophic activities of bone marrowmesenchymal stem cells: a noninvasive therapeutic regimen,”American Journal of Physiology, vol. 296, no. 6, pp. H1888–H1897, 2009.

[11] M. Gnecchi, Z. Zhang, A. Ni, and V. J. Dzau, “Paracrinemechanisms in adult stem cell signaling and therapy,”Circulation Research, vol. 103, no. 11, pp. 1204–1219, 2008.

[12] S. Dimmeler, A. M. Zeiher, and M. D. Schneider, “Unchainmy heart: the scientific foundations of cardiac repair,” Journalof Clinical Investigation, vol. 115, no. 3, pp. 572–583, 2005.

[13] D. C. Colter, R. Class, C. M. DiGirolamo, and D. J. Prockop,“Rapid expansion of recycling stem cells in cultures of plastic-adherent cells from human bone marrow,” Proceedings of theNational Academy of Sciences of the United States of America,vol. 97, no. 7, pp. 3213–3218, 2000.

[14] D. C. Colter, I. Sekiya, and D. J. Prockop, “Identification ofa subpopulation of rapidly self-renewing and multipotentialadult stem cells in colonies of human marrow stromal cells,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 98, no. 14, pp. 7841–7845, 2001.

[15] A. A. Mangi, N. Noiseux, D. Kong et al., “Mesenchymalstem cells modified with Akt prevent remodeling and restoreperformance of infarcted hearts,” Nature Medicine, vol. 9, no.9, pp. 1195–1201, 2003.

[16] L. Da Silva Meirelles and N. B. Nardi, “Murine marrow-derived mesenchymal stem cell: isolation, in vitro expansion,and characterization,” British Journal of Haematology, vol.123, no. 4, pp. 702–711, 2003.

[17] B. Annabi, Y.-T. Lee, S. Turcotte et al., “Hypoxia promotesmurine bone-marrow-derived stromal cell migration andtube formation,” Stem Cells, vol. 21, no. 3, pp. 337–347, 2003.

[18] J. E. Grove, E. Bruscia, and D. S. Krause, “Plasticity of bonemarrow-derived stem cells,” Stem Cells, vol. 22, no. 4, pp.487–500, 2004.

[19] D. S. Krause, “Plasticity of marrow-derived stem cells,” GeneTherapy, vol. 9, no. 11, pp. 754–758, 2002.

[20] J. J. Minguell, A. Erices, and P. Conget, “Mesenchymal stemcells,” Experimental Biology and Medicine, vol. 226, no. 6, pp.507–520, 2001.

[21] A. Peister, J. A. Mellad, B. L. Larson, B. M. Hall, L. F. Gibson,and D. J. Prockop, “Adult stem cells from bone marrow(MSCs) isolated from different strains of inbred mice varyin surface epitopes, rates of proliferation, and differentiationpotential,” Blood, vol. 103, no. 5, pp. 1662–1668, 2004.

[22] M. F. Pittenger, A. M. Mackay, S. C. Beck et al., “Multilineagepotential of adult human mesenchymal stem cells,” Science,vol. 284, no. 5411, pp. 143–147, 1999.

[23] M. F. Pittenger and B. J. Martin, “Mesenchymal stem cells andtheir potential as cardiac therapeutics,” Circulation Research,vol. 95, no. 1, pp. 9–20, 2004.

[24] P. Liu, J.-H. Lin, and B. Zhang, “Differential regulation ofcadherin expression by osteotropic hormones and growthfactors in vitro in human osteoprogenitor cells,” Acta Phar-macologica Sinica, vol. 26, no. 6, pp. 705–713, 2005.

[25] L. Song and R. S. Tuan, “Transdifferentiation potential ofhuman mesenchymal stem cells derived from bone marrow,”FASEB Journal, vol. 18, no. 9, pp. 980–982, 2004.

[26] S. Tomita, R.-K. Li, R. D. Weisel et al., “Autologous trans-plantation of bone marrow cells improves damaged heartfunction,” Circulation, vol. 100, no. 19, pp. 247–256, 1999.

[27] K. C. Wollert and H. Drexler, “Clinical applications of stemcells for the heart,” Circulation Research, vol. 96, no. 2, pp.151–163, 2005.

[28] A. M. Mackay, S. C. Beck, J. M. Murphy, F. P. Barry, C.O. Chichester, and M. F. Pittenger, “Chondrogenic differ-entiation of cultured human mesenchymal stem cells frommarrow,” Tissue Engineering, vol. 4, no. 4, pp. 415–428, 1998.

[29] J. Oswald, S. Boxberger, B. Jørgensen et al., “Mesenchymalstem cells can be differentiated into endothelial cells in vitro,”Stem Cells, vol. 22, no. 3, pp. 377–384, 2004.

[30] D. Chanda, S. Kumar, and S. Ponnazhagan, “Therapeuticpotential of adult bone marrow-derived mesenchymal stemcells in diseases of the skeleton,” Journal of Cellular Biochem-istry, vol. 111, no. 2, pp. 249–257, 2010.

[31] W. Dai, S. L. Hale, B. J. Martin et al., “Allogeneic mes-enchymal stem cell transplantation in postinfarcted ratmyocardium: short- and long-term effects,” Circulation, vol.112, no. 2, pp. 214–223, 2005.

[32] S. J. Greco and P. Rameshwar, “Microenvironmental consid-erations in the application of human mesenchymal stem cellsin regenerative therapies,” Biologics, vol. 2, no. 4, pp. 699–705,2008.

[33] S. Rangappa, C. Fen, E. H. Lee, A. Bongso, and E. S. K. Wei,“Transformation of adult mesenchymal stem cells isolatedfrom the fatty tissue into cardiomyocytes,” Annals of ThoracicSurgery, vol. 75, no. 3, pp. 775–779, 2003.

[34] T. Saito, J.-Q. Kuang, B. Bittira, A. Al-Khaldi, and R. C.-J.Chiu, “Xenotransplant cardiac chimera: immune tolerance ofadult stem cells,” Annals of Thoracic Surgery, vol. 74, no. 1, pp.19–24, 2002.

[35] J. G. Shake, P. J. Gruber, W. A. Baumgartner et al.,“Mesenchymal stem cell implantation in a swine myocardialinfarct model: engraftment and functional effects,” Annals ofThoracic Surgery, vol. 73, no. 6, pp. 1919–1926, 2002.

[36] J.-S. Wang, D. Shum-Tim, J. Galipeau, E. Chedrawy, N.Eliopoulos, and R. C.-J. Chiu, “Marrow stromal cells forcellular cardiomyoplasty: feasibility and potential clinicaladvantages,” Journal of Thoracic and Cardiovascular Surgery,vol. 120, no. 5, pp. 999–1006, 2000.

[37] F. Timmermans, J. De Sutter, and T. C. Gillebert, “Stem cellsfor the heart, are we there yet?” Cardiology, vol. 100, no. 4,pp. 176–185, 2003.

[38] B. E. Strauer, M. Brehm, T. Zeus et al., “Repair of infarctedmyocardium by autologous intracoronary mononuclearbone marrow cell transplantation in humans,” Circulation,vol. 106, no. 15, pp. 1913–1918, 2002.

[39] T.-S. Li, K. Hamano, K. Suzuki, H. Ito, N. Zempo, and M.Matsuzaki, “Improved angiogenic potency by implantationof ex vivo hypoxia prestimulated bone marrow cells in rats,”American Journal of Physiology, vol. 283, no. 2, pp. H468–H473, 2002.

[40] T. Kobayashi, K. Hamano, T.-S. Li et al., “Enhancement ofangiogenesis by the implantation of self bone marrow cells in

Bone Marrow Research 11

a rat ischemic heart model,” Journal of Surgical Research, vol.89, no. 2, pp. 189–195, 2000.

[41] K. Hirata, T.-S. Li, M. Nishida et al., “Autologous bonemarrow cell implantation as therapeutic angiogenesis forischemic hindlimb in diabetic rat model,” American Journalof Physiology, vol. 284, no. 1, pp. H66–H70, 2003.

[42] A. Sacchetti, R. H. Harris, A. Sharon, I. Shpirer, andG. Cotter, “Transendocardial delivery of autologous bonemarrow enhances collateral perfusion and regional functionin pigs with chronic experimental myocardial ischemia,”Journal of the American College of Cardiology, vol. 37, no. 6,pp. 1726–1732, 2001.

[43] T. Kinnaird, E. S. Burnett, M. Shou et al., “Local delivery ofmarrow-derived stromal cells augments collateral perfusionthrough paracrine mechanisms,” Circulation, vol. 109, no. 12,pp. 1543–1549, 2004.

[44] T. Kinnaird, E. Stabile, M. S. Burnett, and S. E. Epstein, “Bonemarrow-derived cells for enhancing collateral development:mechanisms, animal data, and initial clinical experiences,”Circulation Research, vol. 95, no. 4, pp. 354–363, 2004.

[45] D. Hakuno, F. Konishi, Y. Tomita et al., “Bone marrow-derived regenerated cardiomyocytes (CMG cells) expressfunctional adrenergic and muscarinic receptors,” Circulation,vol. 104, no. 17, p. 145, 2001.

[46] S. Makino, K. Fukuda, S. Miyoshi et al., “Cardiomyocytes canbe generated from marrow stromal cells in vitro,” Journal ofClinical Investigation, vol. 103, no. 5, pp. 697–705, 1999.

[47] T. Kinnaird, E. Stabile, M. S. Burnett et al., “Marrow-derivedstromal cells express genes encoding a broad spectrum ofarteriogenic cytokines and promote in-vitro and in-vivoarteriogenesis through paracrine mechanisms,” CirculationResearch, vol. 94, no. 5, pp. 678–685, 2004.

[48] N. Noiseux, A. A. Mangi, M. Gnecchi et al., “Geneticallyengineered mesenchymal stem cells expressing Akt exhibitimproved glucose metabolism, resistance to apoptosis andmarkedly improved performance of infarcted rat hearts,”Circulation, vol. 108, no. 17, p. 5, 2003.

[49] N. Noiseux, A. A. Mangi, D. L. Kong et al., “Geneticallymodified mesenchymal stem cells with Akt reduce apoptosisand normalize performance of infarcted hearts,” MolecularTherapy, vol. 7, no. 5, p. S332, 2003.

[50] N. Noiseux, M. Lopez-Iiasaca, M. Gnecchi et al., “Trans-plantation of bone marrow-derived mesenchymal stem cellsexpressing Akt into infarcted murine heart produces dra-matic improvement in cardiac function despite infrequentcellular fusion,” Circulation, vol. 110, no. 17, p. 68, 2004.

[51] N. Noiseux, M. Gnecchi, M. Lopez-Ilasaca et al., “Mes-enchymal stem cells overexpressing Akt dramatically repairinfarcted myocardium and improve cardiac function despiteinfrequent cellular fusion or differentiation,” Molecular Ther-apy, vol. 14, no. 6, pp. 840–850, 2006.

[52] B. K. Nguyen, S. Maltais, L. P. Perrault et al., “Improvedfunction and myocardial repair of infarcted heart by intra-coronary injection of mesenchymal stem cell-derived growthfactors,” Journal of Cardiovascular Translational Research, vol.3, no. 5, pp. 547–58, 2010.

[53] S. Mansour, D. C. Roy, B. Lemieux, C. Ouellet, L.-M. Stevens,and N. Noiseux, “Stem cell therapy for the broken heart:mini-organ transplantation,” Transplantation Proceedings,vol. 41, no. 8, pp. 3353–3357, 2009.

[54] S. Mansour, D.-C. Roy, V. Bouchard et al., “COMPARE-AMI trial: comparison of intracoronary injection of CD133+ bone marrow stem cells to placebo in patients after acute

myocardial infarction and left ventricular dysfunction: studyrationale and design,” Journal of Cardiovascular TranslationalResearch, vol. 3, no. 2, pp. 153–159, 2010.

[55] M. Gnecchi, H. He, O. D. Liang et al., “Paracrine actionaccounts for marked protection of ischemic heart by Akt-modified mesenchymal stem cells,” Nature Medicine, vol. 11,no. 4, pp. 367–368, 2005.

[56] M. Gnecchi, H. He, N. Noiseux et al., “Evidence supportingparacrine hypothesis for Akt-modified mesenchymal stemcell-mediated cardiac protection and functional improve-ment,” FASEB Journal, vol. 20, no. 6, pp. 661–669, 2006.

[57] T. Asahara, C. Bauters, L. P. Zheng et al., “Synergistic effectof vascular endothelial growth factor and basic fibroblastgrowth factor on angiogenesis in vivo,” Circulation, vol. 92,no. 9, pp. 365–371, 1995.

[58] M. Breitbach, T. Bostani, W. Roell et al., “Potential risksof bone marrow cell transplantation into infarcted hearts,”Blood, vol. 110, no. 4, pp. 1362–1369, 2007.

[59] A. Paul, D. Shum-Tim, and S. Prakash, “Investigation on PEGintegrated alginate-chitosan microcapsules for myocardialtherapy using marrow stem cells genetically modified byrecombinant baculovirus,” Cardiovascular Engineering andTechnology, vol. 1, no. 2, pp. 154–164, 2010.

[60] A. H. Al Kindi, G. Chen, Y. Ge et al., “Microencapsulationto reduce mechanical loss of microspheres: implications inmyocardial cell therapy,” Canadian Journal of Cardiology, vol.23, p. 77C, 2007.

[61] A. Paul, Y. Ge, S. Prakash, and D. Shum-Tim, “Microencap-sulated stem cells for tissue repairing: implications in cell-based myocardial therapy,” Regenerative Medicine, vol. 4, no.5, pp. 733–745, 2009.

[62] M. A. Laflamme, S. Zbinden, S. E. Epstein, and C. E.Murry, “Cell-based therapy for myocardial ischemia andinfarction: pathophysiological mechanisms,” Annual Reviewof Pathology, vol. 2, pp. 307–339, 2007.

[63] Z.-Q. Jia, D. A. G. Mickle, R. D. Weisel et al., “Transplantedcardiomyocytes survive in scar tissue and improve heartfunction,” Transplantation Proceedings, vol. 29, no. 4, pp.2093–2094, 1997.

[64] R.-K. Li, Z.-Q. Jia, R. D. Weisel et al., “Cardiomyocytetransplantation improves heart function,” Annals of ThoracicSurgery, vol. 62, no. 3, pp. 654–661, 1996.

[65] R.-K. Li, D. A. G. Mickle, R. D. Weisel et al., “Naturalhistory of fetal rat cardiomyocytes transplanted into adult ratmyocardial scar tissue,” Circulation, vol. 96, no. 9, pp. 179–187, 1997.

[66] T. Sakai, R.-K. Li, R. D. Weisel et al., “Fetal cell transplanta-tion: a comparison of three cell types,” Journal of Thoracic andCardiovascular Surgery, vol. 118, no. 4, pp. 715–725, 1999.

[67] M. Scorsin, A. Hagege, J.-T. Vilquin et al., “Comparisonof the effects of fetal cardiomyocyte and skeletal myoblasttransplantation on postinfarction left ventricular function,”Journal of Thoracic and Cardiovascular Surgery, vol. 119, no.6, pp. 1169–1175, 2000.

[68] D. Orlic, J. Kajstura, S. Chimenti et al., “Bone marrow cellsregenerate infarcted myocardium,” Nature, vol. 410, no. 6829,pp. 701–705, 2001.

[69] D. Orlic, J. A.N. Kajstura, S. Chimenti, D. M. Bodine, A. Leri,and P. Anversa, “Transplanted adult bone marrow cells repairmyocardial infarcts in mice,” Annals of the New York Academyof Sciences, vol. 938, pp. 221–230, 2001.

[70] C. Badorff, R. P. Brandes, R. Popp et al., “Transdifferentiationof blood-derived human adult endothelial progenitor cells

12 Bone Marrow Research

into functionally active cardiomyocytes,” Circulation, vol.107, no. 7, pp. 1024–1032, 2003.

[71] G. Condorelli, U. Borello, L. De Angelis et al., “Cardiomy-ocytes induce endothelial cells to trans-differentiate intocardiac muscle: implications for myocardium regeneration,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 98, no. 19, pp. 10733–10738, 2001.

[72] H. Oh, S. B. Bradfute, T. D. Gallardo et al., “Cardiac progen-itor cells from adult myocardium: homing, differentiation,and fusion after infarction,” Proceedings of the NationalAcademy of Sciences of the United States of America, vol. 100,no. 21, pp. 12313–12318, 2003.

[73] V. Planat-Benard, C. Menard, M. Andre et al., “Spontaneouscardiomyocyte differentiation from adipose tissue stromacells,” Circulation Research, vol. 94, no. 2, pp. 223–229, 2004.

[74] C. Toma, M. F. Pittenger, K. S. Cahill, B. J. Byrne, and P.D. Kessler, “Human mesenchymal stem cells differentiateto a cardiomyocyte phenotype in the adult murine heart,”Circulation, vol. 105, no. 1, pp. 93–98, 2002.

[75] N. F. Grigoropoulos and A. Mathur, “Stem cells in cardiacrepair,” Current Opinion in Pharmacology, vol. 6, no. 2, pp.169–175, 2006.

[76] J. Kajstura, M. Rota, B. Whang et al., “Bone marrowcells differentiate in cardiac cell lineages after infarctionindependently of cell fusion,” Circulation Research, vol. 96,no. 1, pp. 127–137, 2005.

[77] M. Alvarez-Dolado, R. Pardal, J. M. Garcia-Verdugo etal., “Fusion of bone-marrow-derived cells with Purkinjeneurons, cardiomyocytes and hepatocytes,” Nature, vol. 425,no. 6961, pp. 968–973, 2003.

[78] R. B. Driesen, G. D. Dispersyn, F. K. Verheyen et al., “Partialcell fusion: a newly recognized type of communicationbetween dedifferentiating cardiomyocytes and fibroblasts,”Cardiovascular Research, vol. 68, no. 1, pp. 37–46, 2005.

[79] N. Noiseux, M. Gnecchi, M. Lopez-Ilasaca et al., “Mes-enchymal stem cells overexpressing Akt dramatically repairinfarcted myocardium and improve cardiac function despiteinfrequent cellular fusion or differentiation,” Molecular Ther-apy, vol. 14, no. 6, pp. 840–850, 2006.

[80] J. M. Nygren, S. Jovinge, M. Breitbach et al., “Bone marrow-derived hematopoietic cells generate cardiomyocytes at a lowfrequency through cell fusion, but not transdifferentiation,”Nature Medicine, vol. 10, no. 5, pp. 494–501, 2004.

[81] R. J. Hassink, A. B. De La Riviere, C. L. Mummery, and P.A. Doevendans, “Transplantation of cells for cardiac repair,”Journal of the American College of Cardiology, vol. 41, no. 5,pp. 711–717, 2003.

[82] M. A. Laflamme and C. E. Murry, “Regenerating the heart,”Nature Biotechnology, vol. 23, no. 7, pp. 845–856, 2005.

[83] L. Zeng, Q. Hu, X. Wang et al., “Bioenergetic and func-tional consequences of bone marrow-derived multipotentprogenitor cell transplantation in hearts with postinfarctionleft ventricular remodeling,” Circulation, vol. 115, no. 14, pp.1866–1875, 2007.

[84] E. Martin-Rendon, S. J. Brunskill, C. J. Hyde, S. J. Stanworth,A. Mathur, and S. M. Watt, “Autologous bone marrowstem cells to treat acute myocardial infarction: a systematicreview,” European Heart Journal, vol. 29, no. 15, pp. 1807–1818, 2008.

[85] J. S. Forrester, A. J. White, S. Matsushita, T. Chakravarty, andR. R. Makkar, “New paradigms of myocardial regenerationpost-infarction tissue preservation, cell environment, andpluripotent cell sources,” JACC: Cardiovascular Interventions,vol. 2, no. 1, pp. 1–8, 2009.

[86] B. Doyle, B. J. Kemp, P. Chareonthaitawee et al., “Dynamictracking during intracoronary injection of18F-FDG- labeledprogenitor cell therapy for acute myocardial infarction,”Journal of Nuclear Medicine, vol. 48, no. 10, pp. 1708–1714,2007.

[87] N. Terada, T. Hamazaki, M. Oka et al., “Bone marrowcells adopt the phenotype of other cells by spontaneous cellfusion,” Nature, vol. 416, no. 6880, pp. 542–545, 2002.

[88] R. Liao, O. Pfister, M. Jain, and F. Mouquet, “The bonemarrow-cardiac axis of myocardial regeneration,” Progress inCardiovascular Diseases, vol. 50, no. 1, pp. 18–30, 2007.

[89] I. Buschmann and W. Schaper, “Arteriogenesis versus angio-genesis: two mechanisms of vessel growth,” News in Physio-logical Sciences, vol. 14, no. 3, pp. 121–125, 1999.

[90] I. Buschmann and W. Schaper, “The pathophysiology of thecollateral circulation (arteriogenesis),” Journal of Pathology,vol. 190, no. 3, pp. 338–342, 2000.

[91] I. Buschmann, M. Heil, M. Jost, and W. Schaper, “Influenceof inflammatory cytokines on arteriogenesis,” Microcircula-tion, vol. 10, no. 3-4, pp. 371–379, 2003.

[92] P. Carmeliet, “Mechanisms of angiogenesis and arteriogene-sis,” Nature Medicine, vol. 6, no. 4, pp. 389–395, 2000.

[93] S. E. Epstein, S. Fuchs, Y. F. Zhou, R. Baffour, and R.Kornowski, “Therapeutic interventions for enhancing collat-eral development by administration of growth factors: basicprinciples, early results and potential hazards,” Cardiovascu-lar Research, vol. 49, no. 3, pp. 532–542, 2001.

[94] W. Risau, “Mechanisms of angiogenesis,” Nature, vol. 386,no. 6626, pp. 671–674, 1997.

[95] W. Cheng, J. Kajstura, J. A. Nitahara et al., “Programmedmyocyte cell death affects the viable myocardium afterinfarction in rats,” Experimental Cell Research, vol. 226, no.2, pp. 316–327, 1996.

[96] F. J. Giordano, H.-P. Gerber, S.-P. Williams et al., “A cardiacmyocyte vascular endothelial growth factor paracrine path-way is required to maintain cardiac function,” Proceedingsof the National Academy of Sciences of the United States ofAmerica, vol. 98, no. 10, pp. 5780–5785, 2001.

[97] K. Kitta, R. M. Day, Y. Kim, I. Torregroza, T. Evans, andY. J. Suzuki, “Hepatocyte growth factor induces GATA-4phosphorylation and cell survival in cardiac muscle cells,”Journal of Biological Chemistry, vol. 278, no. 7, pp. 4705–4712, 2003.

[98] N. Lee, T. Thorne, D. W. Losordo, and Y.-S. Yoon, “Repairof ischemic heart disease with novel bone marrow-derivedmultipotent stem cells,” Cell Cycle, vol. 4, no. 7, pp. 861–864,2005.

[99] J. Yamashita, H. Itoh, M. Hirashima et al., “Flk1-positivecells derived from embryonic stem cells serve as vascularprogenitors,” Nature, vol. 408, no. 6808, pp. 92–96, 2000.

[100] L. A. Bokeriia, I. I. Berishvili, and I. I. Sigaev, “Alternativemethods of myocardial revascularization,” Vestnik RossiiskoiAkademii Meditsinskikh Nauk, no. 12, pp. 46–51, 2009.

[101] K. Hamano, M. Nishida, K. Hirata et al., “Local implantationof autologous bone marrow cells for therapeutic angiogenesisin patients with ischemic heart disease—clinical trial andpreliminary results,” Japanese Circulation Journal, vol. 65, no.9, pp. 845–847, 2001.

[102] N. S. Kenyon, M. Willman, N. M. Kenyon et al., “Mesenchy-mal stem cells (MSC) enhance allogeneic islet engraftmentand reverse rejection episodes in cynomolgus monkeys,”American Journal of Transplantation, vol. 98, p. 468, 2009.

[103] J. J. Minguell, F. M. Florenzano, M. R. Ramırez, R. F.Martınez, and G. P. Lasala, “Intracoronary infusion of a

Bone Marrow Research 13

combination of bone marrow-derived stem cells in dogs,”Experimental and Clinical Cardiology, vol. 15, no. 2, pp. 17–20, 2010.

[104] H. C. Quevedo, K. E. Hatzistergos, B. N. Oskouei et al.,“Allogeneic mesenchymal stem cells restore cardiac functionin chronic ischemic cardiomyopathy via trilineage differ-entiating capacity,” Proceedings of the National Academy ofSciences of the United States of America, vol. 106, no. 33, pp.14022–14027, 2009.

[105] B. E. Strauer and R. Kornowski, “Stem cell therapy inperspective,” Circulation, vol. 107, no. 7, pp. 929–934, 2003.

[106] Y. L. Tang, K. Qian, Y. C. Zhang, L. Shen, and M.I. Phillips, “Mobilizing of haematopoietic stem cells toischemic myocardium by plasmid-mediated stromal-cell-derived factor-1α treatment,” Regulatory Peptides, vol. 125,no. 1–3, pp. 1–8, 2005.

[107] S.-C. Hung, R. R. Pochampally, S.-C. Hsu et al., “Short-term exposure of multipotent stromal cells to low oxygenincreases their expression of CX3CR1 and CXCR4 and theirengraftment in vivo,” PLoS ONE, vol. 2, no. 5, article no. e416,2007.

[108] M. Shi, J. Li, L. Liao et al., “Regulation of CXCR4 expressionin human mesenchymal stem cells by cytokine treatment:role in homing efficiency in NOD/SCID mice,” Haematolog-ica, vol. 92, no. 7, pp. 897–904, 2007.

[109] T. E. Ichim, F. Solano, F. Lara et al., “Combination stem celltherapy for heart failure,” International Archives of Medicine,vol. 3, no. 1, p. 5, 2010.

[110] D. Orlic, “Adult bone marrow stem cells regeneratemyocardium in ischemic heart disease,” Annals of the NewYork Academy of Sciences, vol. 996, pp. 152–157, 2003.

[111] J. M. Ryan, F. P. Barry, J. M. Murphy, and B. P. Mahon,“Mesenchymal stem cells avoid allogeneic rejection,” Journalof Inflammation, vol. 2, article no. 8, 2005.

[112] I. K. Ko, B.-G. Kim, A. Awadallah et al., “Targeting improvesMSC treatment of inflammatory bowel disease,” MolecularTherapy, vol. 18, no. 7, pp. 1365–1372, 2010.

[113] G. Chen, M. Nayan, M. Duong et al., “Marrow stromal cellsfor cell-based therapy: the role of antiinflammatory cytokinesin cellular cardiomyoplasty,” Annals of Thoracic Surgery, vol.90, no. 1, pp. 190–197, 2010.

[114] C. A. Dinarello, “The interleukin-1 family: 10 years ofdiscovery,” FASEB Journal, vol. 8, no. 15, pp. 1314–1325,1994.

[115] S. J. Greco, K. Liu, and P. Rameshwar, “Functional similaritiesamong genes regulated by Oct4 in human mesenchymal andembryonic stem cells,” Stem Cells, vol. 25, no. 12, pp. 3143–3154, 2007.

[116] A. Bartholomew, C. Sturgeon, M. Siatskas et al., “Mes-enchymal stem cells suppress lymphocyte proliferation invitro and prolong skin graft survival in vivo,” ExperimentalHematology, vol. 30, no. 1, pp. 42–48, 2002.

[117] J. Mazar, M. Thomas, L. Bezrukov et al., “Cytotoxicity medi-ated by the Fas ligand (FasL)-activated apoptotic pathway instem cells,” Journal of Biological Chemistry, vol. 284, no. 33,pp. 22022–22028, 2009.

[118] F. H. Seeger, A. M. Zeiher, and S. Dimmeler, “Cell-enhancement strategies for the treatment of ischemic heartdisease,” Nature Clinical Practice Cardiovascular Medicine,vol. 4, supplement 1, pp. S110–S113, 2007.

[119] M. Flaquer, P. Romagnani, and J. M. Cruzado, “Growthfactors and renal regeneration,” Nefrologia, vol. 30, no. 4, pp.385–393, 2010.

[120] F. Togel, Z. Hu, K. Weiss, J. Isaac, C. Lange, and C.Westenfelder, “Administered mesenchymal stem cells protectagainst ischemic acute renal failure through differentiation-independent mechanisms,” American Journal of Physiology,vol. 289, no. 1, pp. F31–F42, 2005.

[121] F. Lin, A. Moran, and P. Igarashi, “Intrarenal cells, not bonemarrow-derived cells, are the major source for regenerationin postischemic kidney,” Journal of Clinical Investigation, vol.115, no. 7, pp. 1756–1764, 2005.

[122] J. S. Duffield and J. V. Bonventre, “Kidney tubular epitheliumis restored without replacement with bone marrow-derivedcells during repair after ischemic injury,” Kidney Interna-tional, vol. 68, no. 5, pp. 1956–1961, 2005.

[123] J. S. Duffield, K. M. Park, L.-L. Hsiao et al., “Restorationof tubular epithelial cells during repair of the postischemickidney occurs independently of bone marrow-derived stemcells,” Journal of Clinical Investigation, vol. 115, no. 7, pp.1743–1755, 2005.

[124] M. Broekema, M. C. Harmsen, J. A. Koerts et al., “Determi-nants of tubular bone marrow-derived cell engraftment afterrenal ischemia/reperfusion in rats,” Kidney International, vol.68, no. 6, pp. 2572–2581, 2005.

[125] B. Bi, R. Schmitt, M. Israilova, H. Nishio, and L. G.Cantley, “Stromal cells protect against acute tubular injuryvia an endocrine effect,” Journal of the American Society ofNephrology, vol. 18, no. 9, pp. 2486–2496, 2007.

[126] K. Urbanek, M. Rota, S. Cascapera et al., “Cardiac stem cellspossess growth factor-receptor systems that after activationregenerate the infarcted myocardium, improving ventricularfunction and long-term survival,” Circulation Research, vol.97, no. 7, pp. 663–673, 2005.

[127] R.-X. Xu, X. Chen, J.-H. Chen, Y. Han, and B.-M. Han, “Mes-enchymal stem cells promote cardiomyocyte hypertrophyin vitro through hypoxia-induced paracrine mechanisms,”Clinical and Experimental Pharmacology and Physiology, vol.36, no. 2, pp. 176–180, 2009.

[128] J. Chen, Y. Li, L. Wang, M. Lu, X. Zhang, and M. Chopp,“Therapeutic benefit of intracerebral transplantation of bonemarrow stromal cells after cerebral ischemia in rats,” Journalof the Neurological Sciences, vol. 189, no. 1-2, pp. 49–57, 2001.

[129] Y.-S. Chen, C.-L. Hsieh, C.-C. Tsai et al., “Peripheral nerveregeneration using silicone rubber chambers filled withcollagen, laminin and fibronectin,” Biomaterials, vol. 21, no.15, pp. 1541–1547, 2000.

[130] A. Mahmood, D. Lu, M. Lu et al., “Treatment of traumaticbrain injury in adult rats with intravenous administration ofhuman bone marrow stromal cells,” Neurosurgery, vol. 53, no.3, pp. 697–703, 2003.

[131] Y. Gu, J. Wang, F. Ding, N. Hu, Y. Wang, and X. Gu,“Neurotrophic actions of bone marrow stromal cells onprimary culture of dorsal root ganglion tissues and neurons,”Journal of Molecular Neuroscience, vol. 40, no. 3, pp. 332–341,2010.

[132] P. J. Psaltis, A. C. W. Zannettino, S. G. Worthley, andS. Gronthos, “Concise review: mesenchymal stromal cells:potential for cardiovascular repair,” Stem Cells, vol. 26, no.9, pp. 2201–2210, 2008.

[133] P. J. Psaltis, S. Paton, F. See et al., “Enrichment for STRO-1 expression enhances the cardiovascular paracrine activityof human bone marrow-derived mesenchymal cell popula-tions,” Journal of Cellular Physiology, vol. 223, no. 2, pp. 530–540, 2010.

14 Bone Marrow Research

[134] M. J. Webber, X. Han, S. N. Prasanna Murthy et al., “Cap-turing the stem cell paracrine effect using heparin-presentingnanofibres to treat cardiovascular diseases,” Journal of TissueEngineering and Regenerative Medicine, vol. 4, no. 8, pp. 600–10, 2010.

[135] H. Taegtmeyer, “Cardiac metabolism as a target for thetreatment of heart failure,” Circulation, vol. 110, no. 8, pp.894–896, 2004.

Hindawi Publishing CorporationBone Marrow ResearchVolume 2011, Article ID 252953, 11 pagesdoi:10.1155/2011/252953

Research Article

Differential Secondary Reconstitution ofIn Vivo-Selected Human SCID-Repopulating Cells inNOD/SCID versus NOD/SCID/γ chainnull Mice

Shanbao Cai,1, 2 Haiyan Wang,1, 2 Barbara Bailey,1, 3 Jennifer R. Hartwell,1, 2 Jayne M. Silver,3

Beth E. Juliar,4 Anthony L. Sinn,3 Arthur R. Baluyut,5 and Karen E. Pollok1, 3

1 Section of Pediatric Hematology/Oncology, Department of Pediatrics, Herman B Wells Center for Pediatric Research,Indiana University Simon Cancer Center, The Riley Hospital for Children, 980 West Walnut Street, R3 516, Indianapolis,IN 46202-5525, USA

2 Indiana University Simon Cancer Center (IUSCC), Indiana University School of Medicine (IUSM), Indianapolis, IN 46202, USA3 In Vivo Therapeutics Core, IUSCC, Indianapolis, IN 46202, USA4 Biostatistics and Data Management Core, IUSCC, Indianapolis, IN 46202, USA5 Northside Gastroenterology, St. Vincent Hospital, Indianapolis, IN 46260, USA

Correspondence should be addressed to Karen E. Pollok, [email protected]

Received 2 August 2010; Revised 21 October 2010; Accepted 27 October 2010

Academic Editor: A. Ganser

Copyright © 2011 Shanbao Cai et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Humanized bone-marrow xenograft models that can monitor the long-term impact of gene-therapy strategies will help facilitateevaluation of clinical utility. The ability of the murine bone-marrow microenvironment in NOD/SCID versus NOD/SCID/γchainnull mice to support long-term engraftment of MGMTP140K-transduced human-hematopoietic cells following alkylator-mediated in vivo selection was investigated. Mice were transplanted with MGMTP140K-transduced CD34+ cells and transducedcells selected in vivo. At 4 months after transplantation, levels of human-cell engraftment, and MGMTP140K-transduced cells in thebone marrow of NOD/SCID versus NSG mice varied slightly in vehicle- and drug-treated mice. In secondary transplants, althoughequal numbers of MGMTP140K-transduced human cells were transplanted, engraftment was significantly higher in NOD/SCID/γchainnull mice compared to NOD/SCID mice at 2 months after transplantation. These data indicate that reconstitution ofNOD/SCID/γ chainnull mice with human-hematopoietic cells represents a more promising model in which to test for genotoxicityand efficacy of strategies that focus on manipulation of long-term repopulating cells of human origin.

1. Introduction

Development of humanized hematopoietic xenograft modelsthat can monitor both short- and long-term reconstitutionof human hematopoietic stem and progenitor cells (HSCs)will be a necessity as hematopoietic stem cell gene therapytrials continue to move forward and show promise in theclinic [1–7]. Clearly the use of small and large animal modelshas been an intricate component in paving the way for thesesuccesses [8–24]. Adverse events, however, directly relatedto retroviral-mediated insertional mutagenesis in humantrials, have been reported in some patients enrolled in gene-therapy clinical trials for severe combined immunodeficiency

(SCID)-X1 [25, 26] and chronic granulomatous disease [7,27]. In addition, as preclinical studies in nonhuman pri-mates, dogs, and mice have progressed over the past decade,development of leukemias most likely linked or initiated byvector-mediated insertional mutagenesis have been reported[28–33]. Therefore, more investigations are needed to under-stand retroviral-mediated genome instability in addition tothe design of new vector systems to prevent this adverse eventin human gene-therapy trials. The evaluation of both short-and long-term reconstituting human HSC will be critical inaddressing these issues and is the focus of the current study.

In the past, in vivo experiments designed to study theengraftment and multilineage differentiation of human HSC

2 Bone Marrow Research

were challenging since residual immune reactivity in theanimals could result in loss of transplanted human HSC.Additionally, transplant of human HSC into NOD/SCIDmice, for example, was typically a short-term 3-4 monthassay. With the introduction of new immunodeficient strainssuch as the NOD/SCID/γ chainnull mouse (NSG), long-term analysis of engrafted human hematopoietic cells isnow a reality and represents a significant improvementover the commonly used NOD/SCID mouse for studiesfocused on in vivo analysis of human HSC function [34].The NSG mice are deficient in the Interleukin-2 receptor(IL2R) common γ chain which is an essential componentof multiple cytokine receptors-IL-2, IL-4, IL-7, IL-9, IL-15,and IL-21; this mutation results in a substantial decrease inoverall immune function in this mouse strain. In contrast tothe NOD/SCID mouse, this additional defect in NSG micecauses a significant block in natural killer-cell developmentand makes them less susceptible to lymphoma developmentwhich allows for an increased life span (NSG = 16 monthsversus NOD/SCID = 8–10 months). With the increased lifespan, these mice should be amendable to long-term followupof human HSC function.

Studies using small and large animal models continueto investigate the consequences of genotoxic stress onretrovirally transduced HSC. A powerful tool for in-vivoselection of HSC, a mutant form of O6-methylguanineDNA methyltransferase DNA repair protein (MGMT)-MGMTP140K-has shown promise in nonhuman primates,mice, and humanized xenograft models. Overexpressionof MGMTP140K in HSC, which is resistant to the MGMTinhibitor, O6-benzylguanine (6BG) allows for selection andprotection of the MGMTP140K-transduced HSC followingadministration of 6BG in combination with alkylators suchas BCNU, CCNU, or temozolomide [35]. If MGMTP140K

expression is adequate in the HSC, it should also protectthe HSC from high-dose alkylator therapy required in somecancer treatments and thereby prevent therapy-inducedmyelotoxicity. Generation of HSC that efficiently repair DNAdamage due to chemotherapy may protect patients from life-threatening cytopenias commonly observed following dose-intensified therapy. A case in point, in recent phase II clin-ical trials, patients with nitrosourea-resistant gliomas weresimultaneously treated with 6BG to deplete MGMT in thecancer cells, followed by treatment with the DNA-damagingagents, BCNU or temozolomide [36, 37]. Although lack oftumor progression was transiently observed in some patients,effective dose-escalation therapy could not be achieveddue to severe hematopoietic toxicity. These studies provideclinical proof that strategies protecting HSC during dose-intensified therapy are indeed clearly needed in relapsedpatients requiring high-dose alkylator therapy. In addition,expression of MGMTP140K in HSC can be used as a meansto preferentially select and amplify small populations ofMGMTP140K-transduced cells in the bone marrow. However,long-term impact of this treatment in humanized mousemodels which include secondary transplantation has notbeen investigated. To what extent administration of evenlow doses of chemotherapeutic drugs such as BCNU orTMZ could have on long-term human hematopoiesis is not

well understood and secondary malignancies are always aconcern.

Repetitive low-dose treatment for in vivo selection ofMGMTP140K-transduced cells has been successful in miceand large animal models [8, 16]. Numerous transplantstudies have convincingly proven that long-term repopu-lating murine stem cells could be selected in vivo with6BG/BCNU, 6BG/TMZ, or 6BG/CCNU [9, 11, 14, 17, 18,20, 21, 38–40]. In regards to modeling of this approachwith human HSC, we and others previously demonstratedthat MGMTP140K-transduced SCID-repopulating cells andtheir progeny could be selected in vivo in NOD/SCIDmice [19, 24]. Human HSC derived from umbilical cordblood (UCB) or granulocyte colony-stimulating factor (G-CSF)-mobilized peripheral blood (MPB) that expressedMGMTP140K could be selected in vivo by nonmyeloablativedoses of 6BG and BCNU. Zielske et al. also reportedsimilar results using MGMTP140K-transduced UCB in theNOD/SCID xenograft model [24]. Additionally, our labora-tory went on to investigate the extent to which MGMTP140K-transduced human SCID-repopulating cells and progenycould be protected in vivo by MGMTP140K expression duringdelivery of high-dose alkylator therapy that kills cancer cells.In this study, we compared the outcome of administering alow-dose 6BG/BCNU regimen versus a high-dose regimen inNOD/SCID mice transplanted with MGMTP140K-transducedmobilized peripheral blood CD34+ cells. We found that, atleast in the NOD/SCID xenograft model, when human MPBwere transduced with an oncoretroviral vector that expressesMGMTP140K, only low numbers of human MPB cells wereprotected following delivery of the myeloablative regimenand that these cells were limited to mature lymphoid andmyeloid cells [10]. In all these studies, NOD/SCID mice wereused and analysis of long-term reconstitution in secondaryrecipient mice was not determined.

The objective of our current study was to determineto what degree long-term human SCID-repopulating cellscould be selected in vivo by alkylator therapy and tocompare the levels of selection in primary and secondaryNOD/SCID and NSG mice. Our data demonstrate thathuman hematopoietic cells of multiple lineages were capableof expressing MGMTP140K for at least 4 months in primaryrecipients and in vivo-selected populations while not asrobust as nonselected populations, were able to home andengraft in the bone marrow of secondary recipient NSGmonths for at least an additional 2 months. In contrast tothe NOD/SCID xenograft model, the NSG bone-marrowmicroenvironment appears to allow for optimal reconsti-tution and feasibility of long-term followup of humanhematopoiesis.

2. Materials and Methods

2.1. Animals. Breeding colonies of NOD.Cg-Prkdcscid

(NOD/SCID) and NOD.Cg-Prkdcscid IL2rgtm1W jl/Sz (NSG)mice [34] was established at the Laboratory Animal ResearchCenter and maintained in the In Vivo Therapeutics Coreat the Indiana University Simon Cancer Center (IUSCC)

Bone Marrow Research 3

(Indianapolis, IN). All protocols and were approved by theInstitutional Animal Care and Use Committee at the IndianaUniversity School of Medicine.

2.2. Isolation of Umbilical Cord Blood (UCB) CD34+ Cells.All protocols were approved by Indiana University Schoolof Medicine’s Institutional Review Board (IRB) and St.Vincent Hospital’s IRB (Indianapolis, IN). Samples of UCBwere collected from normal, full-term infants delivered bycesarean section and the CD34+ cells isolated using theCD34 MicroBead kit and VarioMACS Separator (MiltenyiBiotech Inc., Auburn, CA) according to the manufacturer’sinstructions.

2.3. Retrovirus Backbones for Expression of MGMTP140K inCD34+ Cells. The oncoretroviral vector, SF1-MGMTP140K-IRES-EGFP (SF1-P140K) was utilized to coexpressMGMTP140K and EGFP in human CD34+ cells and hasbeen described previously [10]. Retroviral vectors werepseudotyped with the gibbon ape leukemia virus envelope(GALV) using the PG13 packaging cell line (American TypeCulture Collection, Manassas, Va.) [41]. Titers were initiallydetermined on human erythroleukemia (HEL) cells bylimiting-dilution analysis.

2.4. Transduction of UCB CD34+ Cells. The transductionswere done as previously described by our laboratory [10,19, 42]. The starting cell number prior to prestimulationand transduction was 4 × 105 per mouse transplanted.Isolated CD34+ cells were prestimulated at a cell density of5 × 105 cells per ml in Ex Vivo-10 serum-free medium-containing 1% human serum albumin. The medium wassupplemented with Granulocyte-Colony Stimulating Factor(G-CSF), stem cell factor (SCF), and thrombopoietin (TPO)(PeproTech, Rocky Hill, NJ). Each cytokine was used at100 ng/ml for prestimulation. Nontissue culture 10-cm plates(Falcon, Franklin Lakes, NJ), were coated with 2 μg/cm2

Retronectin—(Takara Shuzo, Otsu, Japan) overnight at4◦C. Cells were plated at a concentration of 2 × 105

cells per cm2 for transduction. Cells were infected with a1 : 1 ratio of retrovirus supernatant : complete media withcytokines for 4 hours on 2 consecutive days, with a changeto complete medium-containing cytokines for overnightincubation. After the second round of infection, cells wereallowed to remain overnight on Retronectin-coated plateswith fresh medium and cytokines and then transplanted. Thetransduction efficiency was determined by flow cytometry onthe day of the transplant. For secondary transplants, bone-marrow cells were resuspended at 1 × 106 per ml in ExVivo-10 serum-free medium-containing 1% human serumalbumin, 100 ng/ml SCF, and 100 ng/ml IL-6 (PeproTech) for36–48 hours prior to transplantation.

2.5. Transplantation of NOD/SCID and NSG Mice withHuman CD34+ Cells. Immunodeficient mice were placedon food pellets containing 0.0625% doxycycline for 5–7 days prior to irradiation. Mice were then conditionedwith 300-cGy total-body irradiation using a GammaCell

40 (Nordion International Inc., Ontario Canada) equippedwith two opposing Cesium-137 sources. UCB CD34+ cellswere resuspended in IMDM containing 0.2% endotoxin-freebovine serum albumin and injected into the lateral tail veinof each animal.

2.6.Chemotherapy Administration. O6-benzylguanine (6BG)(Sigma-Aldrich, St Louis, MO) was dissolved in 40%polyethylene glycol-400 (v/v) and 60% saline (v/v). BCNU(Sigma) was dissolved in 10% ethanol (v/v) and 90% normalsaline solution (v/v). BCNU was placed on ice and usedimmediately after reconstitution. One cycle of treatmentconsisted of 20 mg/kg 6BG followed by 5 mg/kg BCNU one-hour later. Two cycles of treatment were delivered one weekapart.

2.7. Analysis of Human Cell Engraftment. Mice were sac-rificed at 8 weeks after 6BG/BCNU injection and single-cell suspensions of the bone marrow (BM) prepared aspreviously described [10, 19, 42]. Human cell engraftmentmeasured by human CD45 staining and the proportionof engraftment in various lineages was determined byimmunostaining and flow cytometric analysis. Aliquots of 1-2× 105 cells/tube were stained with various antibodies for 25minutes at 4◦C in complete medium and washed 1 time inPBS containing 1% FBS. All antibodies were titered and usedat saturating concentrations. The lack of crossreactivity ofhuman-specific antibodies with murine cells was confirmedin every experiment by staining BM from a nontransplantedmouse with each antibody combination. Cells were stainedwith Allophycocyanin (APC)-conjugated anti-human-CD45(anti-HLe-1; Becton Dickinson Immunocytometry, San Jose,CA) alone or in combination with phycoerthyrin (PE)-conjugated anti-human CD33 (anti-Leu-M9; Becton Dick-inson). Identical aliquots were stained with APC-conjugatedanti-human CD34 (clone 581; PharMingen, San Diego, CA)in combination with anti-human CD19-PE (PharMingen) oranti-human CD38-PE (anti-Leu-17; Becton Dickinson). Theforward and right-angle light scatter parameters were used toset the gates for analysis. In experiments where engraftmentof human cells was >5%, ∼20,000–40,000 events werecollected and analyzed. In experiments in which humanengraftment was <5% ∼200,000 events were collected andanalyzed. All samples were acquired and analyzed on aBecton-Dickinson FACSCalibur using CellQuest software(Becton-Dickinson).

2.8. Statistical Analysis. The Mann-Whitney U-test wasutilized to determine statistical significance. Differencesbetween mouse cohorts were considered significant at P <.05 using two-sided tests.

3. Results

3.1. Comparison of Human Hematopoietic Cell Engraftmentin NSG versus NOD/SCID Mice. In order to gain a betterunderstanding of the engraftment capabilities of NSG andNOD/SCID mice, we first evaluated the engraftment of

4 Bone Marrow Research

0

10

20

30

40

50

60

70

80

0 1 2.5 5

NSGNOD/SCID

Hu

man

CD

45+

cell

engr

aftm

ent

(%)

∗∗∗∗

CD34+ cells transplanted per mouse

(×104/mouse)

Figure 1: Comparison of human hematopoietic cell engraftmentin NSG versus NOD/SCID mice. Nonmanipulated umbilical cordblood CD34+ cells (1 × 104–5 × 104 per mouse) were transplantedinto sublethally irradiated (300 cGy) NSG or NOD/SCID mice andhuman cell engraftment in the bone marrow (% human CD45+

cells) analyzed 8 weeks later by flow cytometry. Multi-lineageengraftment was similar in both strains (data not shown). n = 4mice per cohort; ∗NS2 versus NOD/SCID, P < .001.

nontransduced UCB CD34+ cells transplanted into NSGversus NOD/SCID mice. Cohorts of sublethally irradiatedNSG and NOD/SCID mice were transplanted with increasingnumbers of nonmanipulated, freshly isolated UCB CD34+

cells (1.0–5.0 × 104 per mouse). At 2 months after trans-plantation, the bone marrow was harvested and analyzedfor the engraftment of human cells (Figure 1). Humanhematopoietic cells engrafted in all NSG and NOD/SCIDmice but at higher levels in NSG versus NOD/SCID mice.Engraftment levels in the NSG mice was ∼3-fold higher thanlevels observed in the NOD/SCID mice.

3.2. 6BG/BCNU-Mediated In Vivo Selection of MGMTP140K-Transduced UCB CD34+ Cells in Primary Recipient NOD/SCID and NSG Mice. Human umbilical cord blood CD34+

cells were next transduced with a GALV-pseudotypedoncoretroviral vector, SF1-MGMTP140K-IRES-EGFP pre-viously shown by our group to express high levelsof MGMTP140K in NOD/SCID mice transplanted withMGMTP140K-transduced human cells [10]. The transductionefficiency was determined by EGFP expression in bulkCD34+ cells and clonogenic cells and was 65% and 60%EGFP+, respectively, (data not shown). Equal numbers oftransduced cells were transplanted into sublethally irradiated(300 cGy) NSG or NOD/SCID mice (Figure 2). A relativelylarge number of UCB CD34+ cells were transplanted permouse (4 × 105 total CD34+ cells) so sufficient numbersof human cells could be obtained for the secondary trans-plants following the in vivo selection and recovery period.One month after transplantation, mice received 2 cyclesadministered one week apart consisting of 20 mg/kg 6BGfollowed one hour later with 5 mg/kg BCNU. At 4 monthsafter transplantation, the bone marrow was harvested and

2 days

cytokines

UCBCD34+

cells Retronectin

NSG 6BG + BCNU

Onemonth

NSG NOD/SCID

NOD/SCID

Secondary transplantation

+IL-6/SCF for 40 hours

NOD/SCIDNSG

300 cGy 300 cGy

2 cycles oftransduction =

50–75% EGFP+ CD34+ cells

Harvest @4 months post-transplant

Harvest @2 months post-transplant

GALV/SF1-MGMTP140K-IRES-EGFP

Figure 2: Overview of primary and secondary reconstitutionexperiments in NSG versus NOD/SCID mice. Human umbil-ical cord blood CD34+ cells were transduced with a GALV-pseudotyped oncoretroviral vector, SF1-MGMTP140K-IRES-EGFP(MO1 = 5) and transduced cells transplanted into sublethallyirradiated (300 cGy) NSG or NOD/SCID mice. One month aftertransplantation, cells were selected in vivo with 2 cycles of 20 mg/kg6BG followed one hour later with 5 mg/kg BCNU. At 4 monthsafter transplantation, the bone marrow was harvested and analyzedfor the presence of human cells by flow cytometry. Bone marrowcultures were prestimulated in IL-6 and SCF for 48 hours and thenequal numbers of human bone-marrow cells were transplanted intosecondary recipient NSG and NOD/SCID mice. At 2 months aftertransplant, the secondary mice were analyzed for the presence ofhuman cells via flow cytometry.

analyzed for the presence of human cells by flow cytometry.Both mouse strains contained high levels of human-cellengraftment in the bone marrow (Figure 3(a)) and invivo selection of transduced cells with 6BG/BCNU wasachieved in both mouse strains (Figure 3(b)). However, theoverall human engraftment (human CD45+ cells) in vehicle-treated NOD/SCID mice was slightly decreased comparedto engraftment in vehicle-treated NSG mice (Figure 3(a),vehicle-treated NOD/SCID mice-69%± 8% versus vehicle-treated NSG mice-81%± 4). In addition, the variability ofhuman-cell engraftment within the drug-treated NOD/SCIDcohort was more pronounced than that seen in the drug-treated NSG cohort (Figure 3(a), drug-treated NOD/SCIDmice-63%± 32% human CD45+ cells versus drug-treatedNSG mice-76%± 10% human CD45+ cells). In the drug-treated NOD/SCID mice for example, 2/8 exhibited a sig-nificant decrease in human chimerism following 6BG/BCNU

Bone Marrow Research 5

%huCD45+ cells:

Vehicle 6BG/BCNU

NOD/SCID

NSGNOD/SCID

NSG

huC

D45

+ce

lls(%

)

0

10

20

30

40

50

60

70

80

90

100

69± 8 81± 4 63± 32 76± 10

(a)

Vehicle 6BG/BCNU

NOD/SCID

NSGNOD/SCID

NSG

0

10

20

30

40

50

60

70

80

90

100

%huCD45+

EGFP+ cells:

huC

D45

+E

GFP

+ce

lls(%

)

∗ ∗

29± 11 33± 11 82± 14 83± 1

(b)

Figure 3: Comparison of in vivo selected MGMTP140K-transducedhuman cells in NSG and NOD/SCID mice. The starting cell numberwas 4 × 105 cells per transplanted mouse. MGMTP140K-transducedCD34+ cells were transplanted into NS2 and NOD/SCID mice andselected in vivo as outlined in Figure 2. The BM was harvested andanalyzed for the level of total human cell engraftment (%huCD45+

cells) and the level of transduced human cells (%huCD45+EGFP+)by flow cytometry. The BM cellularity was similar between theNSG and NOD/SCID mice and the differences in total humanengraftment in NSG versus NOD/SCID mice were not significant(P > .05). The data presented here are from 1 transplant experiment(n = 3-4 mice in vehicle-treated cohorts and n = 8 mice in6BG/BCNU-treated cohorts). Similar results were obtained in asecond independent transplant experiment.

whereas human-cell engraftment remained more consistentwithin the NSG drug-treated cohort (Figure 3(a)).

While significant in vivo selection of humanMGMTP140K-transduced cells was clearly evident at4 months after transplant regardless of the mouse strain,these data indicate that the microenvironment of the NSGmouse which is devoid of murine NK-cell activity may bemore amendable for the in vivo selection and amplificationof transduced human cells. A detailed flow cytometricanalysis was also performed to see if any specific cell lineagewas preferentially affected by the in vivo selection in themouse strains. While the variability in the percentages

huC

D34

+ce

lls(%

)

0

2

4

6

8

10

12

14

16NSGNOD/SCID

Vehicle6BG/

BCNUVehicle

6BG/BCNU

(a)

huC

D34

+E

GFP

+ce

lls(%

)

NSGNOD/SCID

0

20

40

60

80

100

Vehicle6BG/

BCNUVehicle

6BG/BCNU

∗∗

(b)

Figure 4: Analysis of human CD34+ cells in the bone marrow ofprimary recipient vehicle- and 6BG/BCNU-treated mice. (a) Thepercentage of total human CD34+ cells and (b) CD34+ humancells that are EGFP+ (huCD34+EGFP+) was determined by flowcytometry at 8 weeks after drug treatment. ∗P < .001, vehicle versus6BG/BCNU treated.

of human CD34+ cells was larger in the drug-treatedNOD/SCID mice than the NSG mice, it did not reachstatistical significance (Figure 4(a)). Consistent with theoverall in vivo selection data presented in Figure 3, therewas a significant increase in the in vivo-selected CD34+

cells in vehicle- versus drug-treated regardless of the mousestrain used indicating a selection and some expansionof the human EGFP+, CD34+ cells in the bone marrow(Figure 4(b)). The outcome of 6BG/BCNU treatment onmultilineage differentiation typically found in the bonemarrow of NOD/SCID and NSG mice reconstituted withhuman cells was next evaluated (Figure 5). In vehicle-treatedprimary recipient NSG and NOD/SCID mice, similarpercentages of total and transduced CD34+ progenitor,human B-lymphoid, and myeloid cells were present in thehuman bone-marrow grafts. Human bone-marrow graftsin vehicle-treated NSG and NOD/SCID mice consisted of10–15% CD34+ progenitor, 65–80% B-lymphoid, and 10–20% myeloid cells, and 20–30% of these lineages expressedEGFP (data not shown). In 6BG/BCNU-treated NSG andNOD/SCID mice, a significant selection of human EGFP+

cells was evident in CD34+ progenitor, CD19+ B-lymphoid

6 Bone Marrow Research

and human CD33+ myeloid cells with at least 88% ofthese cell lineages expressing EGFP+ at 8 weeks after drugtreatment (Figure 5).

3.3. Secondary Reconstitution of NOD/SCID and NSG Micewith MGMTP140K-Transduced UCB CD34+ Cells Derived fromNOD/SCID and NSG Mice. Peled et al. originally demon-strated that secondary reconstitution of immunodeficientmice is improved by exposure to IL-6 and SCF prior toretransplantation [43]. The primary mechanism for theenhancement in repopulating ability is increased expressionof the CXCR-4 receptor previously shown to play a keyrole in the homing of transplanted cells. In agreement withthese observations, we found that IL-6/SCF prestimulation ofhuman bone-marrow cultures derived from the transplantedNOD/SCID or NSG mice led to an increase or at leastmaintenance of CXCR-4 expression on the surface of themajority of the human cells (data not shown). Followingthe 48-hour prestimulation period, 10 × 106 human CD45+

cells (vehicle or drug-treated) which contained similarpercentages of CD34+ cells were transplanted into sublethallyirradiated NOD/SCID or NSG mice as outlined in Figure 2.At 2 months after transplantation, the bone marrow of thesecondary recipient mice was analyzed for the percentageof total human CD45+ cells (Figure 6(a)) and EGFP+

human CD45+ cells (Figure 6(b)). Even though equivalentnumbers of transduced human cells from primary recipientNOD/SCID and NSG mice were transplanted into secondaryrecipient mice, striking and statistically significant differ-ences in the level of engraftment between the NOD/SCIDand the NSG mice was observed (Figure 6(a)). Low levelsof engraftment were observed in secondary NOD/SCIDmice receiving human cells from vehicle-treated primaryNOD/SCID mice and ranged from 2%–4% human CD45+

cells with 25–45% of these human hematopoietic cellsremaining EGFP+. In addition, in NOD/SCID secondaryrecipient mice transplanted with bone marrow from primaryrecipient NOD/SCID mice previously drug treated, only 1of 4 mice showed detectable levels of human cells (1.5%human CD45+ cells). In comparison to the NOD/SCIDtransplants, different levels of human-cell engraftment werefound in the secondary transplant experiments using theNSG mice. Human-cell engraftment was detected in allsecondary recipient NSG mice (Figure 6), although engraft-ment levels were different from that previously observedin the primary transplanted NSG mice. Engraftment ofNSG mice receiving humanized bone marrow from vehicle-treated mice was 55%–70% human CD45+ with 15–30% ofthe cells transduced. Engraftment of secondary NSG micereceiving humanized bone marrow from drug-treated miceranged from 4%–18% human CD45+ with EGFP+ cellsranging from 15%–95%. While a significant increase in thepercentage of EGFP+CD34+ cells was observed in the drug-treated versus the vehicle-treated secondary recipient mice,significant differences were noted in the overall engraftmentlevels of the CD34+ cells in the secondary recipient NSGmice (Figure 7). A significant decrease in the percentageof CD34+ cells in the grafts derived from the drug-treated

mice compared to grafts derived from vehicle-treated micewas evident and is consistent with decreased long-termSCID-repopulating ability in the secondary recipient NSGmice that received bone marrow from drug-treated pri-mary NSG recipients. The impact of 6BG/BCNU treatmenton multilineage differentiation of human cells was nextevaluated (Figure 8) and a significant selection of EGFP+

human cells was evident in multiple cell lineages in theNSG mice. Increased percentages of EGFP+ cells were foundin populations of human hematopoietic CD34+ progenitor,CD19+ B-lymphoid, and CD33+ myeloid cells derived fromthe bone marrow of drug-treated NSG mice indicatingthat secondary reconstituting SCID-repopulating cells werecapable of multilineage differentiation.

4. Discussion and Conclusions

Maintenance of genome stability in hematopoietic stemand progenitor cells (HSC) following ex vivo manipulationand transplantation will be essential for normal blood-celldevelopment. The objective of this study was to develop ahumanized bone-marrow xenograft model to follow long-term reconstitution of genetically modified human HSC. Wedetermined if populations of in vivo-selected MGMTP140K-transduced human HSC residing in the bone marrowstill contained sufficient SCID-repopulating cell activity toreconstitute secondary recipient animals. In this initial study,our focus was to define the basic parameters of the modelso that adequate engraftment of HSC in secondary recipientscould be achieved. The MGMTP140K-transduced human HSCunderwent in vivo selection in the primary recipient miceone-month after transplant and then the mice were allowedto recover out to 4-months after transplant. We found thatwhile both the NOD/SCID and NSG models adequatelysupported human HSC reconstitution and in vivo-selectionin the primary recipient mice, NSG mice were far superior tothe NOD/SCID model for secondary reconstitution. It willnow be possible to extend the post-transplant time frame,and studies are currently in progress in which geneticallymodified human HSC are being followed for at least a yearafter transplantation.

With basic parameters of the model defined, we arecontinuing to develop this valuable xenograft model. Studiesare now in progress to test lenti- and foamy-viral vectorsystems to determine if increased numbers of long-termSCID-repopulating cells can be transduced. In addition,the consistency of long-term transgene expression andalso analysis of retroviral-insertion sites in primary andsecondary transplanted animals in the absence and presenceof in vivo selection will be evaluated in these long-termstudies.

Since Natural-Killer cell precursor cells require expres-sion of the common γ chain protein for development intomature NK cells in vivo, lack of this protein will precludetheir development. The rejection of human hematopoieticcells by murine NK cells in NOD/SCID mice is mostlikely the primary reason for decreased engraftment ofthe human CD34+ cells in the NOD/SCID mice but not

Bone Marrow Research 7

+6BG/BCNU

NOD/SCID

NSG

98%

Progenitors

100

101

102

103

104

100 101 102 103 104

EGFP

91%

B-lymphoid

100

101

102

103

104

100 101 102 103 104

EGFP

Myeloid

94%

100

101

102

103

104

100 101 102 103 104

EGFP

98%

100

101

102

103

104

100 101 102 103 104

EGFP

88%

100

101

102

103

104

100 101 102 103 104

EGFP

90%

100

101

102

103

104

100 101 102 103 104

EGFP

CD

A34

AP

C

CD

19P

E

CD

33P

E

CD

A34

AP

C

CD

19P

E

CD

33P

EFigure 5: Representative example of human multilineage engraftment in 6BG/BCNU-treated primary recipient mice. The bone marrow washarvested at 8-weeks after drug treatment, and the level of total and transduced human progenitors (CD34+), B-lymphoid (CD19+), andmyeloid (CD33+) cells were determined via flow cytometry. The level of transduced cells was similar between NOD/SCID and NSG mice.The percentage in the upper right-hand quadrant represents the percentage of the human progenitor, B-lymphoid, or myeloid cells that areEGFP+.

in the NSG mice. Others have shown previously that theinjection of anti-CD122 inhibits NK activity in NOD/SCIDmice and improves engraftment in the bone marrow [34,44–48]. In order to begin to understand the mechanisticdifferences seen in the engraftment levels of the two mousestrains, we initially looked at the “engraftability” of non-manipulated CD34+ cells in primary recipient NSG versusNOD/SCID mice. Differences in the engraftment of theSCID-repopulating cells in primary NSG versus NOD/SCIDrecipients mice were apparent when low numbers of humanCD34+ cells were transplanted. Similar engraftgment resultsin primary recipient NSG versus NOD/SCID mice have beenrecently reported by McDermott et al. [44]. It is importantto note that our data also demonstrated that when largernumbers of transduced CD34+ cells were initially trans-planted into primary recipients NOD/SCID or NOD/SCID/γchainnull mice, differences in overall human engraftmentwere evident but were not statistically significant. However,once the chimeric bone-marrow cells were transplantedinto secondary recipient animals, engraftment differences inSCID-repopulating activity between NSG and NOD/SCIDwere statistically significant.

Bone-marrow cellularity was similar in the NSG andNOD/SCID mouse cohorts indicating that the absolutenumbers of EGFP+ cells in the bone marrow in the vehicle-and treated-mice in the primary NOD/SCID and NSG micewere also similar. Flow cytometric analyses indicated that themean fluorescence intensities for EGFP were also similar inthe EGFP+ populations found in both strains of mice (datanot shown). We have previously shown that EGFP expressiondoes correlate with MGMTP140K expression [10]. Therefore,potential differences in transgene expression most likely donot account for the engraftment differences seen between theNOD/SCID and NSG mice.

While the major cellular difference noted in the originalpaper by Shultz et al. was the lack of murine NK-cell devel-opment [34], we cannot rule out that the function of othercell types in the bone-marrow microenvironment could bemodulated by lack of the common γ chain expression andthat this could affect the homing, survival, expansion, and/ordifferentiation of the human hematopoietic cells. Differentialengraftment differences observed in our study are consistentwith the hypothesis that the lack of mature murine NKcells in the NSG mice allows for more optimal engraftment,

8 Bone Marrow Research

huC

D45

+ce

lls(%

)

NODSCID NSG NOD

SCID NSG

0

10

20

30

40

50

60

70

80

Vehicle-treatedprimary mice

6BG/BCNU-treatedprimary mice

Bone marrowsource:

(a)

NODSCID NSG NOD

SCID NSG

Vehicle-treatedprimary mice

6BG/BCNU-treatedprimary mice

Bone marrowsource:

0

10

20

30

40

50

60

70

80

90

100

huC

D45

+E

GFP

+ce

lls(%

)

(b)

Figure 6: Analysis of human cell engraftment in secondary recipi-ent NOD/SCID and NSG mice. (a) The percentage of total humancells (%huCD45+) in the bone marrow and (b) the percentage of thehuCD45+ cells that are EGFP+ (%huCD45+EGFP+) were analyzedat 8 weeks after transplant into secondary recipient NOD/SCIDand NSG mice as outlined in Figure 2. ∗P < .001, NSG versusNOD/SCID. The date presented are from 1 transplant experiment(n = 3 secondary recipient mice transplanted with BM fromvehicle-treated cohorts and n = 4–8 secondary mice transplantedwith BM from 6BG/BCNU-treated cohorts. Similar results wereobtained in a second independent transplant experiment.

survival, and differentiation of hematopoietic cell subsetssince murine NK cells are not present to destroy the humangrafts in the NSG mice.

Our data indicate that while we were able to detecthuman HSC engraftment in secondary NSG mice trans-planted with bone marrow from drug-treated primary mice(4%–18% human CD45+), it did not approach the levelsof reconstitution seen in secondary NSG mice transplantedwith bone marrow from vehicle-treated mice (55%–70%

0

2

4

6

8

10

huC

D34

+ce

lls(%

)

Vehicle 6BG/BCNU

(a)

0

20

40

60

80

100

huC

D34

+E

GFP

+ce

lls(%

)

Vehicle 6BG/BCNU

∗ ∗

(b)

Figure 7: Analysis of human CD34+ cells in secondary recipientNSG mice transplanted with BM from vehicle- or 6BG/BCNU-treated mice. (a) The percentage of total human CD34 cells(%huCD34+) (b) the percentage of the human CD34+ cells thatare EGFP+ (%huCD34+EGFP+) in the bone marrow were analyzedat 8 weeks after transplant into secondary recipient NSG mice byflow cytometry. The source of the bone-marrow cells that weretransplanted into the secondary recipient mice were derived fromvehicle- or 6BG/BCNU-treated mice primary NSG mice. ∗P < .001,total CD34+ cells-vehicle versus 6BG/BCNU treated; ∗P < .05,transduced CD34+ cells-vehicle versus 6BG/BCNU treated.

human CD45+). Furthermore, the percentage of EGFP+ cellsin secondary NSG mice transplanted with bone marrowfrom drug-treated primary mice cells varied from 15%to 95%. Transgene expression was more uniform in thedrug-treated primary recipient mice (Figure 2(b)). It is notclear at this time whether this decrease in the percent oftransduced cells has to do with silencing of the vector and/ordecreased SCID-repopulating cell activity in the transducedpopulation. No 6BG/BCNU treatment was administered inthe secondary recipient mice and future experiments willalso address the outcome of administering in vivo selectivepressure in the secondary recipient mice. Transplantation ofex vivo manipulated HSC into NOD/SCID/γ chainnull micerepresents a feasible model in which to test and validatenovel strategies that focus on therapeutic manipulation oflong-term repopulating cells from human stem cell sources.Umbilical-cord blood cells were utilized as the transplantsource in this study. It will be important to define thelong-term SCID-repopulating cell frequency in human bonemarrow and G-CSF-mobilized peripheral blood in the NSG

Bone Marrow Research 9

100

101

102

103

104

100 101 102 103 104

EGFP

100

101

102

103

104

100 101 102 103 104

EGFP

100

101

102

103

104

100 101 102 103 104

EGFP

100

101

102

103

104

100 101 102 103 104

EGFP

100

101

102

103

104

100 101 102 103 104

EGFP

100

101

102

103

104

100 101 102 103 104

EGFP

100

101

102

103

104

100 101 102 103 104

EGFP

100

101

102

103

104

100 101 102 103 104

EGFP

Total Progenitors B-lymphoid Myeloid

Vehicle-treated BM

6BG/BCNU-treated BM

Source ofprimary BM:

Reconstitution in secondary recipients

CD

45A

PC

CD

34A

PC

CD

19P

E

CD

33P

E

CD

45A

PC

CD

34A

PC

CD

19P

E

CD

33P

E

35% 25% 46% 23%

55% 78% 70% 75%

Figure 8: Representative example of multilineage engraftment in secondary recipient NSG mice transplanted with BM from vehicle- or6BG/BCNU-treated mice. The percentage of total human (CD45+) and transduced human progenitors (CD34+), B-lymphoid (CD19+), andmyeloid (CD33+) cells in the bone marrow were determined via flow cytometry at 8 weeks after transplant in secondary recipient NSG mice.The percentage in the upper right-hand quadrant represents the percentage of the human cells that are EGFP+ in the progenitor, B-lymphoid,or myeloid cell subsets. The source of the bone-marrow cells that were transplanted into the secondary recipient mice were derived fromvehicle- or 6BG/BCNU-treated mice primary NSG mice.

mice since these sources will be used in human gene-therapy trials, and typically have lower primary SCID-repopulating cell frequencies than UCB [49], and can bemore refractive to transduction than UCB-derived HSC [42].These data indicate that long-lived SCID-repopulating cellscan be transduced with oncoretroviral vectors and selectedin vivo and offers a promising and unique model in which toassess potential genotoxicity and ultimately optimize in vivoselection strategies.

Grant Support

Hope Street Kids Foundation (KEP); the Riley Children’sFoundation (SC, HW, and KEP); RO1 CA138798 (KEP).

Acknowledgments

The authors wish to thank Drs. Scott Goebel and HelmutHanenberg for their support and helpful discussions. Theyalso thank Mary Murray for her support and critical readingof this paper. They thank the nursing staff at the St. VincentHospital for donation of their time for collection of umbilicalcord blood products. They also acknowledge the continuingsupport of the Riley Childrens’ Foundation and the In VivoTherapeutics Core at the Indiana University Simon CancerCenter.

References

[1] A. Aiuti, S. Slavin, M. Aker et al., “Correction of ADA-SCIDby stem cell gene therapy combined with nonmyeloablativeconditioning,” Science, vol. 296, no. 5577, pp. 2410–2413,2002.

[2] A. Aiuti, S. Vai, A. Mortellaro et al., “Immune reconstitutionin ADA-SCID after PBL gene therapy and discontinuation ofenzyme replacement ,” Nature Medicine, vol. 8, no. 5, pp. 423–425, 2002.

[3] M. Cavazzana-Calvo, S. Hacein-Bey, G. de Saint-Basile, F.Le Deist, and A. Fischer, “Gene therapy of severe combinedimmunodeficiencies,” Transfusion Clinique et Biologique, vol.7, pp. 259–260, 2000.

[4] M. Cavazzana-Calvo, S. Hacein-Bey, G. De Saint Basile et al.,“Gene therapy of human severe combined immunodeficiency(SCID)-X1 disease,” Science, vol. 288, no. 5466, pp. 669–672,2000.

[5] J. Chinen, J. Davis, S. S. De Ravin et al., “Gene therapyimproves immune function in preadolescents with X-linkedsevere combined immunodeficiency,” Blood, vol. 110, no. 1,pp. 67–73, 2007.

[6] D. B. Kohn and F. Candotti, “Gene therapy fulfilling itspromise,” The New England Journal of Medicine, vol. 360, no.5, pp. 518–521, 2009.

[7] M. G. Ott, M. Schmidt, K. Schwarzwaelder et al., “Correctionof X-linked chronic granulomatous disease by gene therapy,augmented by insertional activation of MDS1-EVI1, PRDM16or SETBP1,” Nature Medicine, vol. 12, no. 4, pp. 401–409,2006.

10 Bone Marrow Research

[8] B. C. Beard, G. D. Trobridge, C. Ironside, J. S. McCune, J. E.Adair, and H.-P. Kiem, “Efficient and stable MGMT-mediatedselection of long-term repopulating stem cells in nonhumanprimates,” Journal of Clinical Investigation, vol. 120, no. 7, pp.2345–2354, 2010.

[9] S. Cai, A. Ernstberger, H. Wang et al., “In vivo selection ofhematopoietic stem cells transduced at a low multiplicity-of-infection with a foamy viral MGMTP140K vector,” Experi-mental Hematology, vol. 36, no. 3, pp. 283–292, 2008.

[10] S. Cai, J. R. Hartwell, R. J. Cooper et al., “In vivo effects ofmyeloablative alkylator therapy on survival and differentia-tion of MGMTP140K-transduced human G-CSF-mobilizedperipheral blood cells,” Molecular Therapy, vol. 13, no. 5, pp.1016–1026, 2006.

[11] B. M. Davis, L. Humeau, and B. Dropulic, “In vivo selectionfor human and murine hematopoietic cells transduced witha therapeutic MGMT lentiviral vector that inhibits HIVreplication,” Molecular Therapy, vol. 9, no. 2, pp. 160–172,2004.

[12] C. E. Dunbar and A. Larochelle, “Gene therapy activates EVI1,destabilizes chromosomes,” Nature Medicine, vol. 16, no. 2, pp.163–165, 2010.

[13] Y.-J. Kim, Y.-S. Kim, A. Larochelle et al., “Sustained high-levelpolyclonal hematopoietic marking and transgene expression4 years after autologous transplantation of rhesus macaqueswith SIV lentiviral vector-transduced CD34+ cells,” Blood, vol.113, no. 22, pp. 5434–5443, 2009.

[14] E. L. Kreklau, K. E. Pollok, B. J. Bailey et al., “Hematopoieticexpression of O6-methylguanine DNA methyltransferase-P140K allows intensive treatment of human glioma xenograftswith combination O6-benzylguanine and 1,3-bis-(2-chloroethyl)-1-nitrosourea,” Molecular Cancer Therapeutics,vol. 2, no. 12, pp. 1321–1329, 2003.

[15] A. Larochelle, U. Choi, Y. Shou et al., “In vivo selection ofhematopoietic progenitor cells and temozolomide dose inten-sification in rhesus macaques through lentiviral transductionwith a drug resistance gene,” Journal of Clinical Investigation,vol. 119, no. 7, pp. 1952–1963, 2009.

[16] T. Neff, B. C. Beard, L. J. Peterson, P. Anandakumar, J. Thomp-son, and H.-P. Kiem, “Polyclonal chemoprotection againsttemozolomide in a large-animal model of drug resistance genetherapy,” Blood, vol. 105, no. 3, pp. 997–1002, 2005.

[17] D. A. Persons, E. R. Allay, N. Sawai et al., “Successful treatmentof murine β-thalassemia using in vivo selection of geneticallymodified, drug-resistant hematopoietic stem cells,” Blood, vol.102, no. 2, pp. 506–513, 2003.

[18] D. A. Persons, J. A. Allay, A. Bonifacino et al., “Transientin vivo selection of transduced peripheral blood cells usingantifolate drug selection in rhesus macaques that receivedtransplants with hematopoietic stem cells expressing dihydro-folate reductase vectors,” Blood, vol. 103, no. 3, pp. 796–803,2004.

[19] K. E. Pollok, J. R. Hartwell, A. Braber et al., “In vivo selectionof human hematopoietic cells in a xenograft model usingcombined pharmacologic and genetic manipulations,” HumanGene Therapy, vol. 14, no. 18, pp. 1703–1714, 2003.

[20] S. Ragg, M. Xu-Welliver, J. Bailey et al., “Direct reversal ofDNA damage mutant methyltransferase protein protects miceagainst dose-intensified chemotherapy and leads to in vivoselection of hematopoietic stem cells,” Cancer Research, vol. 60,no. 18, pp. 5187–5195, 2000.

[21] N. Sawai, D. A. Persons, S. Zhou, T. Lu, and B. P. Sorrentino,“Reduction in hematopoietic stem cell numbers with in vivodrug selection can be partially abrogated by HOXB4 gene

expression,” Molecular Therapy, vol. 8, no. 3, pp. 376–384,2003.

[22] S. Sellers, T. J. Gomes, A. Larochelle et al., “Ex vivo expansionof retrovirally transduced primate CD34+ cells results inoverrepresentation of clones with MDS1/evi1 insertion sites inthe myeloid lineage after transplantation,” Molecular Therapy,vol. 18, no. 9, pp. 1633–1639, 2010.

[23] J. Xie, A. Larochelle, I. Maric, M. Faulhaber, R. E. Donahue,and C. E. Dunbar, “Repetitive busulfan administration afterhematopoietic stem cell gene therapy associated with adominant HDAC7 clone in a nonhuman primate,” HumanGene Therapy, vol. 21, no. 6, pp. 695–703, 2010.

[24] S. P. Zielske, J. S. Reese, K. T. Lingas, J. R. Donze, and S.L. Gerson, “In vivo selection of MGMT(P140K) lentivirus-transduced human NOD/SCID repopulating cells withoutpretransplant irradiation conditioning,” Journal of ClinicalInvestigation, vol. 112, no. 10, pp. 1561–1570, 2003.

[25] S. Hacein-Bey-Abina, C. Von Kalle, M. Schmidt et al., “Aserious adverse event after successful gene therapy for X-linked severe combined immunodeficiency,” The New EnglandJournal of Medicine, vol. 348, no. 3, pp. 255–256, 2003.

[26] S. Hacein-Bey-Abina, C. Von Kalle, M. Schmidt et al., “LMO2-associated clonal T cell proliferation in two patients after genetherapy for SCID-X1,” Science, vol. 302, no. 5644, pp. 415–419,2003.

[27] J.-Y. Metais and C. E. Dunbar, “The MDS1-EVI1 genecomplex as a retrovirus integration site: impact on behaviorof hematopoietic cells and implications for gene therapy,”Molecular Therapy, vol. 16, no. 3, pp. 439–449, 2008.

[28] X. Li, M. M. Le Beau, S. Ciccone et al., “Ex vivo culture ofFancc−/− stem/progenitor cells predisposes cells to undergoapoptosis, and surviving stem/progenitor cells display cyto-genetic abnormalities and an increased risk of malignancy,”Blood, vol. 105, no. 9, pp. 3465–3471, 2005.

[29] Z. Li, J. Dullmann, B. Schiedlmeier et al., “Murine leukemiainduced by retroviral gene marking,” Science, vol. 296, no.5567, p. 497, 2002.

[30] U. Modlich, O. S. Kustikova, M. Schmidt et al., “Leukemiasfollowing retroviral transfer of multidrug resistance 1 (MDR1)are driven by combinatorial insertional mutagenesis,” Blood,vol. 105, no. 11, pp. 4235–4246, 2005.

[31] A. W. Nienhuis, C. E. Dunbar, and B. P. Sorrentino, “Genotoxi-city of retroviral integration in hematopoietic cells,” MolecularTherapy, vol. 13, no. 6, pp. 1031–1049, 2006.

[32] R. Seggewiss, S. Pittaluga, R. L. Adler et al., “Acute myeloidleukemia is associated with retroviral gene transfer tohematopoietic progenitor cells in a rhesus macaque,” Blood,vol. 107, no. 10, pp. 3865–3867, 2006.

[33] X.-B. Zhang, B. C. Beard, G. D. Trobridge et al., “Highincidence of leukemia in large animals after stem cell genetherapy with a HOXB4-expressing retroviral vector,” Journalof Clinical Investigation, vol. 118, no. 4, pp. 1502–1510, 2008.

[34] L. D. Shultz, B. L. Lyons, L. M. Burzenski et al., “Humanlymphoid and myeloid cell development in NOD/LtSz-scidIL2Rγnull mice engrafted with mobilized human hemopoieticstem cells,” Journal of Immunology, vol. 174, no. 10, pp. 6477–6489, 2005.

[35] S. L. Gerson, “Clinical relevance of MGMT in the treatment ofcancer,” Journal of Clinical Oncology, vol. 20, no. 9, pp. 2388–2399, 2002.

[36] J. A. Quinn, S. X. Jiang, D. A. Reardon et al., “Phase IItrial of temozolomide plus O6-benzylguanine in adults withrecurrent, temozolomide-resistant malignant glioma,” Journalof Clinical Oncology, vol. 27, no. 8, pp. 1262–1267, 2009.

Bone Marrow Research 11

[37] J. A. Quinn, J. Pluda, M. E. Dolan et al., “Phase II trial of car-mustine plus O6-benzylguanine for patients with nitrosourea-resistant recurrent or progressive malignant glioma,” Journalof Clinical Oncology, vol. 20, no. 9, pp. 2277–2283, 2002.

[38] M. Jansen, U. R. Sorg, S. Ragg et al., “Hematoprotection andenrichment of transduced cells in vivo after gene transfer ofMGMTP140k into hematopoietic stem cells,” Cancer GeneTherapy, vol. 9, no. 9, pp. 737–746, 2002.

[39] M. D. Milsom, L. B. Woolford, G. P. Margison, R. K.Humphries, and L. J. Fairbairn, “Enhanced in vivo selectionof bone marrow cells by retroviral-mediated coexpressionof mutant O6-methylguanine-DNA-methyltransferase andHOXB4,” Molecular Therapy, vol. 10, no. 5, pp. 862–873, 2004.

[40] N. Sawai, S. Zhou, E. F. Vanin, P. Houghton, T. P. Brent, and B.P. Sorrentino, “Protection and in vivo selection of hematopoi-etic stem cells using temozolomide, O6-benzylguanine, andan alkyltransferase-expressing retroviral vector,” MolecularTherapy, vol. 3, no. 1, pp. 78–87, 2001.

[41] A. D. Miller, J. V. Garcia, N. von Suhr, C. M. Lynch, C. Wilson,and M. V. Eiden, “Construction and properties of retroviruspackaging cells based on gibbon ape leukemia virus,” Journalof Virology, vol. 65, no. 5, pp. 2220–2224, 1991.

[42] K. E. Pollok, J. C. M. Van der Loo, R. J. Cooper et al., “Dif-ferential transduction efficiency of SCID-repopulating cellsderived from umbilical cord blood and granulocyte colony-stimulating factor-mobilized peripheral blood,” Human GeneTherapy, vol. 12, no. 17, pp. 2095–2108, 2001.

[43] A. Peled, I. Petit, O. Kollet et al., “Dependence of humanstem cell engraftment and repopulation of NOD/SCID miceon CXCR4,” Science, vol. 283, no. 5403, pp. 845–848, 1999.

[44] S. P. McDermott, K. Eppert, E. R. Lechman, M. Doedens, andJ. E. Dick, “Comparison of human cord blood engraftmentbetween immunocompromised mouse strains,” Blood, vol.116, no. 2, pp. 193–200, 2010.

[45] J. L. McKenzie, O. I. Gan, M. Doedens, and J. E. Dick, “Humanshort-term repopulating stem cells are efficiently detectedfollowing intrafemoral transplantation into NOD/SCID recip-ients depleted of CD122+ cells,” Blood, vol. 106, no. 4, pp.1259–1261, 2005.

[46] L. D. Shultz, S. J. Banuelos, J. Leif et al., “Regulation ofhuman short-term repopulating cell (STRC) engraftmentin NOD/SCID mice by host CD122+ cells,” ExperimentalHematology, vol. 31, no. 6, pp. 551–558, 2003.

[47] L. D. Shultz, P. A. Schweitzer, S. W. Christianson et al., “Mul-tiple defects in innate and adaptive immunologic function inNOD/LtSz- scid mice,” Journal of Immunology, vol. 154, no. 1,pp. 180–191, 1995.

[48] T. Tanaka, F. Kitamura, Y. Nagasaka, K. Kuida, H. Suwa,and M. Miyasaka, “Selective long-term elimination of naturalkiller cells in vivo by an anti-interleukin 2 receptor βchain monoclonal antibody in mice,” Journal of ExperimentalMedicine, vol. 178, no. 3, pp. 1103–1107, 1993.

[49] A. Larochelle, J. Vormoor, H. Hanenberg et al., “Identificationof primitive human hematopoietic cells capable of repopulat-ing NOD/SCID mouse bone marrow: implications for genetherapy,” Nature Medicine, vol. 2, no. 12, pp. 1329–1337, 1996.

Hindawi Publishing CorporationBone Marrow ResearchVolume 2011, Article ID 385124, 6 pagesdoi:10.1155/2011/385124

Clinical Study

One-Year Safety Analysis of the COMPARE-AMI Trial:Comparison of Intracoronary Injection of CD133+ Bone MarrowStem Cells to Placebo in Patients after Acute MyocardialInfarction and Left Ventricular Dysfunction

Samer Mansour,1, 2 Denis-Claude Roy,3 Vincent Bouchard,4 Louis Mathieu Stevens,2, 5

Francois Gobeil,1 Alain Rivard,1, 2 Guy Leclerc,1 Francois Reeves,1, 2 and Nicolas Noiseux2, 5

1 Division de Cardiologie, Departement de Medecine, Centre Hospitalier de l’Universite de Montreal (CHUM), 3840, Rue Saint Urbain,Montreal, Quebec, Canada H2W 1T8

2 Axe Cardio-Metabolique, Centre de Recherche du CHUM (CRCHUM), Montreal, Quebec, Canada H2W 1T73 Departement d’Hematologie, Hopital Maisonneuve-Rosemont (HMR), Montreal, Quebec, Canada H1T 2M44 Faculte de Medicine, Universite de Montreal, Montreal, Quebec, Canada H3C 3J75 Division de Chirurgie Cardiaque, Departement de Chirurgie, Centre Hospitalier de l’Universite de Montreal (CHUM), Montreal,Quebec, Canada H2W 1T8

Correspondence should be addressed to Samer Mansour, [email protected] andNicolas Noiseux, [email protected]

Received 22 August 2010; Accepted 12 December 2010

Academic Editor: Mark R. Litzow

Copyright © 2011 Samer Mansour et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Bone marrow stem cell therapy has emerged as a promising approach to improve healing of the infarcted myocardium. Despiteinitial excitement, recent clinical trials using non-homogenous stem cells preparations showed variable and mixed results. SelectedCD133+ hematopoietic stem cells are candidate cells with high potential. Herein, we report the one-year safety analysis on the initial20 patients enrolled in the COMPARE-AMI trial, the first double-blind randomized controlled trial comparing the safety, efficacy,and functional effect of intracoronary injection of selected CD133+ cells to placebo following acute myocardial infarction withpersistent left ventricular dysfunction. At one year, there is no protocol-related complication to report such as death, myocardialinfarction, stroke, or sustained ventricular arrhythmia. In addition, the left ventricular ejection fraction significantly improved atfour months as compared to baseline and remained significantly higher at one year. These data indicate that in the setting of theCOMPARE-AMI trial, the intracoronary injection of selected CD133+ stem cells is secure and feasible in patients with left ventricledysfunction following acute myocardial infarction.

1. Introduction

Despite improvements in survival rate after myocardialinfarction (MI) with recent medical and surgical advances[1], the reduced heart function attributed to irreversible lossof viable cardiomyocytes remains a major clinical problem[2]. Intracoronary (IC) injection of progenitor cells emergedas a valuable therapeutical approach to improve healingof the ischemic heart. Despite the use of various experi-mental protocols with mixed cells showing mixed results

[3], four recent meta-analyses of bone marrow stem cells(BMSCs) transplantation in the setting of acute MI provedthe feasibility and safety and demonstrated a slight butpositive functional effect [4–7]. However, there is currentlyuncertainty as to which of the BMSC population is mostpotent in stimulating angiogenesis and cardiac repair. Thehematopoietic CD133+ cells possess high engraftment, mul-tipotent, and angiogenic capacity and appear to be valuablefor cardiac repair in experimental myocardial infarction [8–11].

2 Bone Marrow Research

Table 1: Demographic and clinical characteristics. BMI indicatesbody mass index, CAD coronary artery disease, PCI percutaneouscoronary intervention, CK creatinine kinase, and CK-MB creatinekinase-muscle and brain.

Age (yrs) 52, 2± 8.9

Male (%) 90

BMI 28± 5.7

Diabetes (%) 15

Hypertension (%) 35

Hyperlipidemia (%) 20

Smoking (%) 70

Family history of CAD (%) 15

Door to balloon (Min) 123.6± 97.4

PCI to injection (days) 6.4± 2.2

Killip class I: 95; II: 5

Maximal ST elevation (mm) 11.2± 5.5

Q waves on admission (%) 55

Peak troponin T (UI/mL) 10.48± 8.34

Peak CK (UI/mL) 2744± 2193

Peak CK-MB (UI/mL) 341± 260.7

In a phase I study, Bartunek et al. tested the IC injectionof CD133+ cells in patients following acute MI and demon-strated a significant improvement of the left ventricular(LV) function and perfusion as compared to controls [12].However, they noted in the treated group an increase ofcoronary events with greater in-stent proliferation and largerluminal loss in the nonstented segments of the infarct-related(IR) artery [13]. Of note, these data were obtained fromretrospective analysis and they lack randomized controls.Hence, we designed the COMPARE-AMI trial, the first ran-domized, double-blind, placebo-controlled phase II singlecenter study comparing the IC injection of CD133+ cells toplacebo on the cardiac repair in patients with acute MI andpersistent LV dysfunction [14, 15]. Herein, we report theone-year safety analysis on the first randomized patients.

2. Methods

2.1. Study Design and Objectives. COMPARE-AMI study is asingle-center, randomized, prospective, double-arm, double-blind, placebo-controlled phase II clinical trial investigatingthe feasibility, safety, and efficacy of the IC administration ofselected autologous CD133+-enriched hematopoietic BMSCversus placebo in patients following acute MI with persistentdysfunctional myocardium despite successful catheter reper-fusion. The flow chart of the trial is detailed in Figure 1.The primary end point of the study is a composite ofsafety and efficacy end points aiming to determine thechange in the coronary atherosclerotic burden progressionproximal and distal to the stented segment of the culpritartery at 4 months as compared to baseline, and the changein the left ventricle ejection fraction (LVEF) at 4 monthsrelative to baseline. The secondary end points include

the occurrence of a major adverse cardiac event (MACE)defined as death, MI, stroke, and target vessel revascularisa-tion (TVR) or the occurrence of major arrhythmias definedas sustained ventricular arrhythmia or survived suddendeath. Patients are followed for 2 years, and the data willbe unblinded 12 months after the randomization of the lastpatient.

2.2. Study Population and Protocol. In brief, patients between30 and 75 years are eligible if they presented an acute STelevation MI successfully reperfused by means of coronarystent implantation and demonstrated a substantial persis-tent LV dysfunction defined by a LVEF ≤50% but ≥25%on echocardiography obtained within 48 hours after thesuccessful reperfusion therapy. Written informed consentis obtained within 7 days after the onset of acute MI ifthe patient is free of any exclusion criteria such as knownprevious MI, cardiogenic shock, chronic cardiomyopathy,liver disease, renal failure, concomitant disease with a lifeexpectancy of less than 1 year, alcohol or drug dependency,contraindication for bone marrow (BM) aspiration, bloodtransfusion in the previous 24 hours, hematopoietic disease,chronic inflammatory disease, malignancy, stroke in theprevious 3 months, or transient ischemic attack in theprevious 24 hours. The study protocol was approved byHealth Canada (CTA 113925 and ITA 123840) and byinstitutional ethics review board.

Until September 2009, 20 patients were successfullyenrolled in the trial. A total of 18 patients completed the 4-month followup and 13 patients reached the 1-year followup.All patients underwent BM aspiration 3 to 7 days afterreceiving reperfusion therapy for acute MI. An averagevolume of 100 mL of the BM was obtained and transferredimmediately to the cell processing laboratory where patientswere randomly assigned to receive placebo medium (normalsaline containing 10% autologous plasma) or autologousCD133+ cells. CD133+ progenitor cells for this trial areimmunomagnetically selected using the CliniMACS� sys-tem (Miltenyi Biotec GmbH) according to manufacturer’sprotocols. Within 12 hours of BM harvest, the study suspen-sion is sent back to the cardiac catheterisation laboratory forIC administration in the IR artery using intermittent ballooninflation as previously described in [16]. In summary, theinfusate is delivered via an over-the-wire percutaneous trans-luminal coronary angioplasty catheter over 3 min of ballooninflation within the previously placed intracoronary stent.Three minutes of reperfusion are incorporated after eachcycle of cell infusion. A total of 4 cycles are given. Patientsare routinely discharged 48 hours following the infusionprocedure with the usual postacute MI care medications.

Myocardial functional evaluation is done with echocar-diography, cardiac magnetic resonance imaging (MRI),left ventricular angiography, echocardiography, technetium99 m sestamibi single photon emission computed tomog-raphy, and positron emission tomography with [17] flu-orodeoxyglucose at baseline and 4 months of followup.Cardiac MRI and echocardiography are repeated at 1 year offollowup.

Bone Marrow Research 3

PCI

LVEF <50%Randomization

Day 3–7

Acute MI

IC placebo(n = 20)

Follow-upevaluation

Follow-upevaluation

<7 days 4 months 1 year

IC CD133+

(n = 20)

evaluation:Baseline

(i) Echocardiography

(ii) LV angiography

(iii) Coronarography

(iv) IVUS/FFR

(v) PET + MIBI

(vi) MRI

(vii) Stress echo

(viii) Holter ECG

(i) Echocardiography

(ii) LV angiography

(iii) Coronarography

(iv) IVUS/FFR

(v) PET + MIBI

(vi) MRI

(vii) Stress echo

(viii) Holter ECG

(i) Echocardiography

(ii) MRI

(iii) Holter ECG

Figure 1: Flow chart of the COMAPRE-AMI trial. MI indicates myocardial infarction, PCI percutaneous coronary intervention, LVEF leftventricular ejection fraction, IC intracoronary, LV left ventricle, IVUS, intravascular ultrasound, FFR fractional flow reserve, PET positronemission tomography, MIBI technetium 99 m sestamibi, MRI magnetic resonance imaging, and ECG electrocardiogram.

Table 2: Biological markers after IC injection, no change over time (All P = NS). WBC indicates white blood cells count, CRP C-reactiveprotein, CK creatinine kinase, and CK-MB creatine kinase-muscle and brain.

Baseline (end of the IC injection) 30 min 60 min 6 hrs Day 1

WBC 8.12± 1.6 8.65± 1.96 8.92± 2.31 7.74± 1.94 7.29± 1.69

CRP 11.09± 12.16 10.15± 8.56 12.15± 12.8 13.18± 14.24 13.08± 13.96

Troponin T 0.83± 1.02 0.88± 1.07 0.91± 1.13 0.8 ± 0.95 0.66± 0.84

CK 167.4± 95.6 176.5± 99 204± 135.9 171.5± 85.3 160.4± 81.97

CK-MB 46.8± 22.2 51.33± 28.17 55.56± 27.58 72± 40.28 60.7± 39.90

Table 3: In-hospital, 4- and 12-month reported adverse events.SVT indicates sustained ventricular tachycardia, ICD implantablecardioverter-defibrillator, BMS bare metal stent DES drug-elutingstent, BM bone marrow, and IC intracoronary.

In-hospital 4 months 12 months

Mortality (n) 0 0 0

Re-infarct (n) 0 0 0

Stroke (n) 0 0 0

SVT/ICD (n) 0 0 0

Heart failure (n) 2 1 1

Stent thrombosis (n) 0 0 0

In-BMS restenosis/occlusion 0 3 0

In-DES restenosis/occlusion 0 0 0

Bleeding/transfusion (n) 2/0 0/0 0/0

BM harvest related (n) 0 NA NA

IC Injection related (n) 0 NA NA

2.3. Safety Evaluation. To determine the safety of the ICinjection of CD133+ cell suspension, standard laboratorytests including white blood cells (WBC) count, C-reactive

protein (CRP), creatine kinase (CK), and creatine kinase-muscle and brain (CK-MB) type and troponin T measure-ments were performed before, 30 and 60 minutes, 6 hours,and 24 hours after the IC injections. In addition, all patientswere continuously monitored until discharge and underwent24-hour ECG monitoring at baseline, 1, 4, and 12 monthsafter the procedure. Furthermore, a quantitative coronaryangiography, pressure-derived fractional flow reserve (FFR),and an IC ultrasonography were performed in the IR arteryand a remote artery at baseline and 4-month followup.

2.4. Statistical Analysis. Baseline demographic and clinicalvariables were collected for all patients. Invasive and non-invasive analyses were performed by operators blinded toall clinical and other functional data. Data are presentedas mean ± standard deviation or median and interquartilerange when appropriate. Continuous variables were ana-lyzed using Student’s t-test. Fully parameterized longitudinalmixed effect models were built for echocardiographic mea-surement, including a coefficient for each point in time atbaseline, 4 and 12 months (the MIXED procedure in SASsoftware, version 9.1; SAS Institute, Cary, NC). These models

4 Bone Marrow Research

30

40

50

60

70

LEV

F(%

)

Baseline 4 months 12 months

Figure 2: Left ventricular ejection fraction improvement assessedby echocardiography for each randomized patient at baseline beforetreatment (n = 20), 4 months (n = 18) and 12 months of followup(n = 13). Cardiac function was significantly improved at 4 and 12months (P < .001). LVEF indicates left ventricular ejection fraction.

account for the correlation between repeated measurementsin the same patient. For all analyses, P < .05 were consideredstatistically significant.

3. Results

3.1. Clinical Characteristics. Herein, we are presenting theone-year safety analysis of the first twenty patients random-ized and treated in the COMPARE-AMI trial. The demo-graphic and clinical characteristics are showed in Table 1.The mean age was 52.2 ± 8.9 years with a predominanceof males (90%) and the culprit lesion was located on theleft anterior descending artery in 90%. A complete occlusionwas noted in the proximal segment of the IR artery in 65%of cases. Maximum troponin T and CKMB were 10.48 ±8.34 Ug/L and 341 ± 260.7 U/L, respectively, suggesting largeinfarcts. All patients underwent a successfully reperfusionwith stenting using a drug-eluted stent (DES) in 75%or a bare metal stent (BMS) in 25% of cases. Adjuvantmedical therapy included aspirin, clopidogrel, β-blockers,angiotensin-converting enzyme inhibitors, and statins.

3.2. Safety Analysis. BM harvests were well tolerated by allpatients without any related complication. For an average of10× 106 CD133+ cells, we obtained a mean recovery of morethan 95% when comparing before and after cell separation.Moreover, we obtained a mean purity higher than 65% forselected CD133+ cells. Cell purity is validated using flowcytometric analysis using standard techniques. Cell viabilityis assessed by trypan bleu exclusion, which is always greaterthan 95%. The IC injection was performed 6.4 ± 2.2 daysafter stenting. Two patients had a small hematoma at thearterial access site (one radial and one femoral) that did notrequire any specific treatment. As compared to baseline, nosignificant differences in the WBC count, CRP, CK, CK-MB,and troponin T levels were found at 24 hours (Table 2). Inaddition, in-hospital continuous ECG monitoring did notreveal any episode of sustained ventricular arrhythmia, andonly two patients had at least one episode of nonsustained

ventricular tachycardia (NSVT). The serial 24-hour holterECG monitoring performed at 1, 4, and 12 months revealeda small increase in the number of patients who had at leastone run of a NSVT (five patients at 4 and 12 months ascompared to two patients at baseline). In addition, despitea low number of isolated ventricular premature beats (VPB)at baseline of 6.5 [1.8, 14.8] and after treatment of 4.5[3.0, 10.5], we observed an important increase at 1-monthfollowup with 30.0 [13.0, 129.5]. Number of VPBs decreasedthereafter at 4 months with a median of 9.0 [3.0, 117.0]. Untilone-year followup, no spontaneous sustained ventriculararrhythmias or survived sudden death were noted and nopatient required an implantable cardioverter-defibrillator(ICD).

Furthermore, up to 12-month followup no MACE, angi-na, or stent thrombosis were reported (Table 3). During theinitial hospitalization and before randomization, 2 patientshad an episode of heart failure that required diuretics. Ofnote, one patient stopped for one week his medication andwas hospitalized 3 months after his randomization for heartfailure with a temporary drop of his LVEF down to 15%.Finally, at 4-month followup, restenting was necessary in 3patients to treat in-BMS restenosis. No in-DES restenosis orde novo lesions in the IR artery were observed. Moreover,despite the significant lower pressure-derived FFR in theculprit artery as compared to the nonculprit artery atbaseline:, 0.88 ± 0.05 versus 0.96 ± 0.04, P < .001, nosignificant modification was noted in the delta FFR at 4-month followup as compared to baseline in the culprit versusnonculprit artery (−3.7%±5.4 versus−1.1%±4.6 resp., P =.148) suggesting the absence of local or diffuse acceleration ofcoronary atherosclerosis.

3.3. LV Function Assessment after Randomization. As shownin Figure 2, echocardiographic assessment of the LVEFdemonstrated a significant improvement at 4-month fol-lowup (n = 18) as compared to baseline (51.1%± 2.5 versus41.2%±1.1 resp., P < .001). This improvement was sustainedat 12-month followup (52.3%±2.0, P < .001 versus baseline,n = 13).

4. Discussion

In the setting of acute MI, several studies [4–7] have shown afunctional benefit of IC infusion of BMSC compared with thestandard treatment alone. Probably, the simplest approach tomyocardial cell therapy in the clinical setting is the transferof BM mononuclear cells into the myocardium. However,using unmodified marrow or unselected mononuclear cellscan prevent the delivery of the “ideal” cell for myocardialregeneration that could be available in a small number orlost during the preparation process. In addition, few of theunselected mononuclear cells could formally meet the stemcell criteria and the vast majority are leukocytes that mayprimarily induce local inflammation, rendering the actualstem cell effects insignificant. Selected CD133+ BM cellpopulation contains a small proportion of clonogenic cells,which have a very high potential to induce neoangiogenesis

Bone Marrow Research 5

[18]. Furthermore, there is accumulating evidence that theCD133+/CD34− subpopulation includes multipotent stemcells with a significant potential for differentiation intomesenchymal and other nonhematopoietic lineages.

Therefore, Bartunek et al. conducted a phase I trial [12]testing the IC injection of purified CD133+ BMSC into theacutely infracted myocardium after successful stenting of theculprit artery. They showed a significant improvement in theLV function in parallel with an increase in the myocardialperfusion and viability suggesting a potent angiogenic effectof the injected cells. However, the following safety concernsin the cell-treated group were reported: first a small butsignificant rise in the CRP, second ventricular arrhythmiaswere observed in 4 treated patients, and third a higher rate ofin-BMS restenosis and de novo lesion was found. Moreover,they raised the possibility of a “Janus-like” effect where theyreported in a retrospective analysis a significant reduction ofthe luminal diameters and epicardial conductance of the IRartery, as assessed from the pressure-derived FFR [13] mainlyin early compared to late IC CD133+ cells administration[17] which seems to be consistent with a higher risk foratherosclerosis progression. Yet, these data were obtainedfrom retrospective analysis; they lack proper randomizedcontrols and systematic use of IC ultrasound imaging to trackchanges in the vascular wall.

Accordingly, we initiated the COMPARE-AMI trial,a randomized, double-blind, controlled study testing thesafety, feasibility, and functional effect of the IC injectionof CD133+ cells as compared to placebo in relatively high-risk patients with acute MI and persistent LV dysfunction.The safety and functional effect will be addressed usingtime frames that are consistent with clinical applicabilityand emerging safety profile. To date twenty patients weresuccessfully randomized and treated, eighteen completed 4months of followup, and thirteen of them reached one yearof followup. Herein, we are presenting our first safety analysiswhere we did not observe any major adverse events such asdeath, stroke, MI, angina, heart failure, or stent thrombosiswithin the first 12 months following randomization. Onepatient stopped his medication for one week and washospitalized for heart failure. In parallel with other trialresults, [12, 16] we did not report any complication relatedto the BM harvest or the technique of IC infusions. Cardiacenzymes levels remained unchanged during the initial 24hours following injection. Furthermore, and in contrary towhat was reported in the Bartunek trial [12], neither theWBC counts nor the CRP levels increased. Additionally,no sustained ventricular arrhythmias were recorded andnone of the randomized patients required an implantablecardioverter-defibrillator. However, the serial 24-hour ECGmonitoring revealed a temporary significant increase in thenumber of isolated VPB at 4 months that spontaneouslydecreased at 12 months of followup suggesting a transientelectrical irritability that could be related to the ischemicburden in the first months after the MI. Of note, only 5patients had at least one run of NSVT until one year offollowup. On the other hand, all randomized patient hada coronary angiogram control at 4 months after the ICinjection; no in-DES restenosis was reported; however three

over five patients treated with a BMS had a significant butasymptomatic restenosis requiring restenting with a DESbecause of a documented significant ischemia in the targetterritory. Interestingly, no de novo lesion was reported at 4-month followup, and no significant difference was found inthe delta FFR as compared to baseline in the culprit versusnonculprit artery suggesting the absence of any accelerationof atherosclerosis. Finally, the 4-month echocardiographyassessment showed a significant increase in the LVEF supe-rior to what we observed in the placebo groups of therandomized trials [4–7] suggesting a potential effect on thecardiac recovery in the CD 133+ injected patients.

In conclusion, the proposed experimental protocol inwhich IC autologous CD133+ cells are injected after acuteMI is safe, without any serious adverse events up to oneyear of followup. In addition, a potential beneficial effecton the LV recovery of the enrolled patients is observed.Final results will be available 1 year after the last patient hasbeen randomized; meanwhile the presented one-year safetyresults justify the continuation of the COMPARE-AMI trial.This randomized clinical study should help to support thetherapeutical potential of CD133+ cells for the treatment ofthe infracted myocardium.

Acknowledgments

The authors appreciate assistance of Drs. B. Lemieux andC. Ouellette, in performing the bone marrow collection.They are thankful to R. Duclos, RN and C. Lemay RN formanagement of the study. The authors received financialsupport from fonds de la recherche en sante du Quebec,Miltenyi Biotec, Inc., and Boston Scientific in Canada.

References

[1] K. A. Jackson, S. M. Majka, H. Wang et al., “Regeneration ofischemic cardiac muscle and vascular endothelium by adultstem cells,” Journal of Clinical Investigation, vol. 107, no. 11,pp. 1395–1402, 2001.

[2] F. Timmermans, J. de Sutter, and T. C. Gillebert, “Stem cellsfor the heart, are we there yet?” Cardiology, vol. 100, no. 4, pp.176–185, 2003.

[3] A. Rosenzweig, “Cardiac cell therapy—mixed results frommixed cells,” The New England Journal of Medicine, vol. 355,no. 12, pp. 1274–1277, 2006.

[4] M. Hristov, N. Heussen, A. Schober, and C. Weber, “Intra-coronary infusion of autologous bone marrow cells and leftventricular function after acute myocardial infarction: a meta-analysis,” Journal of Cellular and Molecular Medicine, vol. 10,no. 3, pp. 727–733, 2006.

[5] M. J. Lipinski, G. G. L. Biondi-Zoccai, A. Abbate et al., “Impactof intracoronary cell therapy on left ventricular function inthe setting of acute myocardial infarction. A collaborativesystematic review and meta-analysis of controlled clinicaltrials,” Journal of the American College of Cardiology, vol. 50,no. 18, pp. 1761–1767, 2007.

[6] E. Martin-Rendon, S. J. Brunskill, C. J. Hyde, S. J. Stanworth,A. Mathur, and S. M. Watt, “Autologous bone marrow stemcells to treat acute myocardial infarction: a systematic review,”European Heart Journal, vol. 29, no. 15, pp. 1807–1818, 2008.

6 Bone Marrow Research

[7] A. Abdel-Latif, R. Bolli, I. M. Tleyjeh et al., “Adult bonemarrow-derived cells for cardiac repair: a systematic reviewand meta-analysis,” Archives of Internal Medicine, vol. 167, no.10, pp. 989–997, 2007.

[8] M. Bhatia, “AC133 expression in human stem cells,” Leukemia,vol. 15, no. 11, pp. 1685–1688, 2001.

[9] E. A. de Wynter, D. Buck, C. Hart et al., “CD34+AC133+

cells isolated from cord blood are highly enriched in long-term culture-initiating cells, NOD/SCID-repopulating cellsand dendritic cell progenitors,” Stem Cells, vol. 16, no. 6, pp.387–396, 1998.

[10] S. Kuci, J. T. Wessels, H. J. Buhring et al., “Identification of anovel class of human adherent CD34− stem cells that give riseto SCID-repopulating cells,” Blood, vol. 101, no. 3, pp. 869–876, 2003.

[11] N. Quirici, D. Soligo, L. Caneva, F. Servida, P. Bossolasco, andG. L. Deliliers, “Differentiation and expansion of endothelialcells from human bone marrow CD133+ cells,” British Journalof Haematology, vol. 115, no. 1, pp. 186–194, 2001.

[12] J. Bartunek, M. Vanderheyden, B. Vandekerckhove et al.,“Intracoronary injection of CD133− positive enriched bonemarrow progenitor cells promotes cardiac recovery afterrecent myocardial infarction: feasibility and safety,” Circula-tion, vol. 112, no. 9, pp. I178–I183, 2005.

[13] S. Mansour, M. Vanderheyden, B. de Bruyne et al., “Intra-coronary delivery of hematopoietic bone marrow stem cellsand luminal loss of the infarct-related artery in patients withrecent myocardial infarction,” Journal of the American Collegeof Cardiology, vol. 47, no. 8, pp. 1727–1730, 2006.

[14] S. Mansour, D. C. Roy, V. Bouchard et al., “COMPARE-AMItrial: comparison of intracoronary injection of CD133+ bonemarrow stem cells to placebo in patients after acute myocardialinfarction and left ventricular dysfunction: study rationale anddesign,” Journal of Cardiovascular Translational Research, vol.3, no. 2, pp. 153–159, 2010.

[15] S. Mansour, D. C. Roy, B. Lemieux, C. Ouellet, L. M. Stevens,and N. Noiseux, “Stem cell therapy for the broken heart: mini-organ transplantation,” Transplantation Proceedings, vol. 41,no. 8, pp. 3353–3357, 2009.

[16] B. E. Strauer, M. Brehm, T. Zeus et al., “Repair of infarctedmyocardium by autologous intracoronary mononuclear bonemarrow cell transplantation in humans,” Circulation, vol. 106,no. 15, pp. 1913–1918, 2002.

[17] M. Vanderheyden, S. Vercauteren, S. Mansour et al., “Time-dependent effects on coronary remodeling and epicardial con-ductance after intracoronary injection of enriched hematopoi-etic bone marrow stem cells in patients with previousmyocardial infarction,” Cell Transplantation, vol. 16, no. 9, pp.919–925, 2007.

[18] M. Peichev, A. J. Naiyer, D. Pereira et al., “Expression ofVEGFR-2 and AC133 by circulating human CD34+ cellsidentifies a population of functional endothelial precursors,”Blood, vol. 95, no. 3, pp. 952–958, 2000.