Review article

Beyond Hox: the role of ParaHox genes in normal and malignant hematopoiesisVijay P. S. Rawat,1 R. Keith Humphries,2 and Christian Buske1

1The Institute of Experimental Cancer Research, Comprehensive Cancer Center and University Hospital Ulm, Ulm, Germany; and 2The Terry Fox Laboratory,British Columbia Cancer Agency, Vancouver, BC

During the past decade it was recognizedthat homeobox gene families such as theclustered Hox genes play pivotal rolesboth in normal and malignant hematopoi-esis. More recently, similar roles havealso become apparent for members of theParaHox gene cluster, evolutionarilyclosely related to the Hox gene cluster.This is in particular found for the caudal-type homeobox genes (Cdx) genes,known to act as upstream regulators of

Hox genes. The CDX gene family memberCDX2 belongs to the most frequent aber-rantly expressed proto-oncogenes in hu-man acute leukemias and is highly leuke-mogenic in experimental models.Correlative studies indicate that CDX2functions as master regulator of per-turbed HOX gene expression in humanacute myeloid leukemia, locating thisParaHox gene at a central position forinitiating and maintaining HOX gene dys-

regulation as a driving leukemogenicforce. There are still few data about poten-tial upstream regulators initiating aber-rant CDX2 expression in human leuke-mias or about critical downstream targetsof CDX2 in leukemic cells. Characterizingthis network will hopefully open the wayto therapeutic approaches that target de-regulated ParaHox genes in human leuke-mia. (Blood. 2012;120(3):519-527)

Introduction

The last decade has seen large advances in our understanding of theregulatory network in prenatal and adult hematopoiesis. As onepivotal class of regulatory factors, the highly conserved family ofhomeobox genes have been identified as playing major roles notonly in normal hematopoiesis but also in leukemogenesis.1 Thehomeobox was first identified in the 1980s as a sequence motifshared among Drosophila homeotic genes (the HOM-C complex)that play key roles in embryonic differentiation along the anterior-posterior axis. The existence of homeotic genes was alreadyproposed in the early 1900s, based on observations of Drosophilamutants with changes in the body structure.2 The homeobox isnow known to be present in many genes in virtually alleukaryotic species.3 It is estimated that the human genomecontains � 200 homeobox genes.4 Unlike the HOM-C/HOX genes,which are organized in gene clusters, most homeobox genes aredispersed throughout the genome.5,6 Homeobox-containing genesconstitute a gene family characterized by a highly conserved183-nucleotide sequence encoding a 61-aa domain, the homeodo-main (HD). These HDs are structurally related to the helix-turn-helix motif of prokaryotic DNA-binding proteins and have sequence-specific DNA binding activity.7 Homeobox genes are divided into2 classes. Class I includes clustered genes (HOX) that comprises39 members, and the class II divergent homeobox genes aredispersed through the genome and include smaller families such asthe MSX, ParaHox (CDX), PAX, DLX, and Engrailed (EN) groups.Although individual members of homeobox families often sharelittle sequence similarity other than the homeobox, some familieshave additional conserved sequence motifs that contribute to theirdistinct functional properties.5,8

A key role of homeobox genes for normal and malignant stemand progenitor cell behavior was first shown for Hox genes. Initialhints that these genes are essential for regulating HSCs andprogenitor cells came from data showing that Hox genes of the

A and B cluster are highly expressed in normal murine and humanHSCs and progenitor cells, but silenced in their more differenti-ated progeny.9,10 Studies in experimental models have convinc-ingly proven the functional relevance of Hox genes for normalhematopoietic behavior in mouse and man; for example,Hoxa9�/� BM cells manifest a severe defect at the level of HSCsand committed progenitors11,12; however, aberrant expression ofHox/HOX genes such as Hoxa9 or Hoxa10 can induce acutemyeloid leukemia (AML) in mice, whereas constitutive expressionof Hoxb4 increases the self-renewal capacity of murine HSCswithout initiating leukemic transformation.13-15 Similar data wereobtained for the human system with the use of retroviral genetransfer techniques and the NOD/SCID xenograft model.16,17 Earlydata indicated aberrant expression of HOX genes in patients withacute leukemia,18 and this was confirmed later by microarrayanalyses in large cohorts of patients with AML, thereby associatingaberrant expression of HOX genes such as Hoxa9 with clinicaloutcome.19-22 The rich body of data on the role of Hox genes innormal and malignant hematopoiesis has been reviewed in detailbefore.1,7,8,23 In the past years, however, it was shown that otherhomeobox genes, not belonging to the clustered Hox genes, have acrucial role in regulating early hematopoietic cells and facilitatingleukemic transformation. One example is the so-called “three-amino-acid loop extension” superfamily, forming an “atypicalclass” of homeobox genes that show low sequence identity withother classes of homeobox genes.24 Most prominent members ofthis family are Meis1 and Pbx1. Both can directly interact withHox genes and with each other, thereby forming trimeric com-plexes, which alter Hox–DNA binding properties and in the case ofMeis1 accelerate Hox-induced leukemogenesis.25 This reviewfocuses on another homeobox gene family, the ParaHox cluster,which was shown to be involved in normal as well as malignanthematopoiesis.6,24

Submitted February 16, 2012; accepted April 23, 2012. Prepublished online asBlood First Edition paper, April 30, 2012; DOI 10.1182/blood-2012-02-385898.

© 2012 by The American Society of Hematology

519BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

The ParaHox cluster genes: emerging keyplayers in normal and malignant hematopoiesis

Genomic organization and molecular characteristics

The ParaHox gene cluster was discovered in 1998 by Brooke et al,who reported that Gsx (genomic screened homeobox), Xlox(Xenopus laevis homeobox), and Cdx (Caudal-type homeobox)genes form their own cluster in the cephalochordate am-phioxus.26 This cluster was termed the ParaHox cluster becauseit showed homology to the different paralogue groups of theHox cluster. In mice and man homologous genes cluster onchromosome 5 and 13q12, respectively; these comprise thegenes Gsh1/GSH1, Xlox/XLOX (also known as Pdx1/PDX1),and CDX2.27,28 In contrast to chordates, mammals have addi-tional ParaHox genes, namely Gsh2/GSH2, Cdx1/CDX1, andCdx4/CDX4, which are spread over 3 different chromosomes inthe human genome (Table 1; Figure 1A).

The ParaHox and Hox clusters are thought to originate froman ancient hypothetical ProtoHox cluster by duplication beforethe split of protostomes and deuterostomes, characterizing theParaHox cluster as one of the evolutionary early homeoboxgene clusters.6,42,43 This also explains the homologies of theParaHox genes to paralogue groups of the Hox cluster. Theparalogue groups of the Hox cluster subdivide the Hox genesinto 4 groups: the anterior group (paralogue group [PGs] 1-2),group 3, central group (PGs 4-8), and posterior group (PGs9-13).6,43 The Gsx gene is more closely related to Hox anteriorgroup genes than to other ParaHox genes; Xlox is more similarto Hox group 3, and Cdx is more similar to the Hox posteriorgroup genes.

Structure–functional analyses have characterized the molecularcharacteristics of ParaHox proteins in detail (Table 1; Figure 1B).As for homeobox genes, all the ParaHox genes are characterized bya DNA-binding HD and in part by a N-terminal transactivation

domain. However, homology between the different familymembers of the ParaHox genes is relatively low and proteinsdisplay individual features. Thus, the ParaHox gene PDX1contains an antennapedia-type hexapeptide motif (PFPWMK),which preferentially binds the DNA motif 5�-[CT]TAAT[TG]-3�. PDX1 is phosphorylated by HIPK2 on Ser-268 on glucoseaccumulation, which mediates shifting of the protein fromnuclear periphery to nucleoplasm.44 GSH1 contains an addi-tional Snail/Gfi domain essential for transcriptional repression.On the basis of bioinformatics analysis it is postulated that GSHgenes bind to DNA as monomers or as homodimers and/orheterodimers, in a sequence-specific manner. CDX1, in contrastto other CDX members, does not contain the highly conservedpentapeptide domain for binding of the three-amino-acid loopextension cofactor Pbx1.

Lessons from knockout mice: initial clues to function ofParaHox genes

Knockout mice have been an important tool for our understandingof ParaHox gene function and their essential role in body develop-ment (Table 1). Gsh1 homozygous mutant mice are phenotypicallynormal at birth but fail to grow properly after 7-10 days. Homozy-gous mutants exhibit extreme dwarfism, sexual infantilism, andsignificant perinatal mortality with total white blood cell count5-fold decreased in mutant mice. Both lymphoid and myeloid cellswere similarly reduced, whereas macrophages were not affected.29

Gsh2 knockout mice display severe structural alterations in thebrain, accompanied by apnea and reduced blood oxygen levels,leading to death within the first day after birth.31 Pdx-1 was shownto be crucial for the development of the pancreas and anteriorduodenum.32 Cdx1-null mutant mice are viable and fertile, but theydisplay anterior homeotic transformations of vertebrae accompa-nied by posterior shifts of Hox gene expression in the somaticmesoderm.34 Cdx2-null mutant embryos die between 3.5 and5.5 days after coitum, and heterozygotes have tail abnormalitiesand exhibit anterior homeotic transformations that involve the

Table 1. Characteristics of ParaHox genes

Humangene

Proteinsize, kDa

Homology betweenman and mouse, %*

Chromosomaltranslocation

Aberrant expression inleukemia Transforming potential Phenotype of knockout mice

GSH1 27 98.01 — Unknown Not tested Phenotypically normal at birth, but

75% of mice die at 4 wks of age;

marked growth retardation,

infertile, and low WBC count29

GSH2 32 89.14 CHIC2-ETV630 AML (case reports)30 Transforms NIH 3T3 cells

in vitro30

Born healthy, but death within

24 hours after birth; homozygous

mutants have apnea and severe

brain alterations31

PDX1 30 87.99 — Unknown Not tested Death at 1 dpp; apancreatic32

CDX1 28 85.28 — Not expressed33 Not tested Viable and fertile, anterior homeotic

transformations; posterior shifts

of Hox gene expression in the

somitic mesoderm34

CDX2 33 93.89 ETV6-CDX235 AML, ALL, CML in blast

crises33,35,36

Causes AML in mice37 Peri-implantation embryonic

lethality; heterozygous mutants

show growth retardation and

anterior transformation38

CDX4 30 82.98 — AML39 Overexpression induces AML

in mice39,40

Viable, only a mild anterior

transformation with low

penetrance41

WBC indicates white blood cell; CML, chronic myeloid leukemia; and —, not described.*At amino acid level.

520 RAWAT et al BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

cervical and upper thoracic vertebrae, ribs, and midgut endoderm,underlining the crucial role of Cdx genes for anterior-posteriorregional identity. Ninety percent of Cdx2 heterozygote mutantmice develop multiple intestinal adenomatous polyps, particu-larly in the proximal colon, suggesting that Cdx2 mutationsmight be the primary event in the genesis of some intestinaltumors.38,45,46 In contrast to Cdx2, Cdx4-null murine embryosare born healthy and do not show any major structural orhistologic abnormalities.41

Cdx2 and Cdx4 are key regulators of embryonic hematopoiesis

Over the past several years it has become evident that ParaHoxCdx family members are critical for ordered proliferation anddifferentiation of embryonic hematopoietic cells.47 This wasparticularly found for the Cdx genes Cdx4 and Cdx2. Evidencefor a key role of Cdx genes in embryonic hematopoiesis initiallycame from studies in zebrafish; homozygous mutations in cdx4caused a severe defect in embryonic hematopoiesis with aprofound reduction of hemoglobin-positive erythroid cells thatresulted in severe anemia.48 These blood defects were accompa-nied by deregulated or reduced expression of hoxb4, hoxb5a,hoxb6b, hoxb7a, hoxb8a, hoxb8b, and hoxa9a. Of note, theblood deficiency in zebrafish embryos could be rescued byoverexpression of hoxb7a or hoxa9a but not by scl, indicatingthat the hematopoietic defect is at least partly caused byperturbation of distinct Hox genes.48 The same group found thatmorpholino-mediated knockdown of cdx1a in a cdx4-mutantbackground caused a severe perturbation of Hox gene expres-

sion and a complete failure to specify blood in zebrafish.Furthermore, these Cdx-deficient embryos displayed a signifi-cant reduction in expression of key hematopoietic genes such asscl, runx1, and gata1.49 This hematopoietic defect could beefficiently rescued by hoxa9 overexpression, but not by hoxb7.49

In mouse, ectopic expression of Cdx4 enhances hematopoi-etic mesoderm formation and further promotes blood progenitorspecification and hematopoietic engraftment of murine embry-onic stem cells (ESCs) in adult mice.50 In addition, Cdx4overexpression rescues defective blood progenitor formation inmouse ESCs deficient in Mll, a Hox regulator involved indefinitive hematopoiesis.51 Surprisingly, Cdx4 knockout mice donot show any major hematopoietic defects.40 Indeed, in bothgermline and conditional Cdx4 knockout models, there is noapparent defect in HSCs and progenitor cells as well as matureblood cells. This indicates that Cdx4 as a single factor is notessential for normal adult hematopoietic stem cell functions inmice and that Cdx4 function can be compensated at least partlyby other Cdx genes.40 One candidate for this is Cdx2; Cdx2-deficient ESCs show impaired production of multipotentialblood progenitor colonies in vitro consistent with an essentialrole of Cdx2 in early embryonic hematopoiesis.52 By injectingCdx2�/� or Cdx2�/� ESCs into normal lacZ� blastocysts,Wang et al could show an intrinsic requirement for Cdx2 duringprimitive embryonic hematopoiesis in vivo.52 According togenome-wide expression analysis they reported that Cdx2deficiency dramatically perturbs expression profiles of Hoxgenes and genes involved in signal transduction, cell growth and

GSH PDX CDX

1 1

2

Chr 13q12 (Chr 5 82cM; 41cM for Gsh1)

Chr 5q31-33 (Chr 18 30cM)

Chr 4q12-13 (Chr 5 82cM)

Ch: X13.2 (Chr X 42.5 cM)

2

1

4

A

B

Figure 1. Characteristics of ParaHox genes. (A) Chro-mosomal location of human ParaHox genes. Chromo-somal locations of murine ParaHox genes are given inbrackets. The direction of transcription is indicated by anarrow. (B) Protein structure of ParaHox genes; shown isthe protein structure depicting the N-terminal transactiva-tion domain (blue), the Pbx-interacting motif (PIM; brown),the HD (green), and validated phosphorylation sites(orange).

PARAHOX GENES IN NORMAL AND MALIGNANT HEMATOPOIESIS 521BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

proliferation, and hematopoiesis. Expression of several compo-nents of the canonical signal transduction pathways such as theJak/Stat and sonic hedgehog pathways were altered in Cdx2�/�

embryonic bodies (EBs). Expression of several hematopoiesis-specific genes, including Scl, Gata1, Runx1, and �-H1 globin,were markedly decreased in Cdx2-defecient EBs,52 supportingthe notion that Cdx2 is a key player during hematopoieticdevelopment.

In contrast to Cdx4 and Cdx2, loss of Cdx1 gene functiondoes not affect hematopoietic development as shown by homozy-gous Cdx1 mutant mice,34 but also by studies in zebrafish,whereby cdx1a had no effect on rostral blood island hematopoi-esis and the intermediate cell mass development.49 The distincteffects of the different members of the Cdx gene family waselegantly shown in a study,52,53 which ectopically expressedCdx1, Cdx2, and Cdx4 in murine ESCs; whereas Cdx1 and Cdx4both promoted the hematopoietic potential of the EB-derivedCD41�c-kit� population, Cdx2 grossly reduced EB-derivedhematopoietic potential. Expression of all the different Cdxgenes increased Hox gene expression with the notable differ-ence that Cdx2 induced changes in both the anterior andposterior HoxA and HoxB genes, whereas Cdx1 and Cdx4 onlyaltered posterior Hox gene expression.53

Taken together, these data indicate that Cdx genes play essentialand both distinct and overlapping roles in embryonic hematopoiesis.

ParaHox genes in malignant hematopoiesis

ParaHox genes have emerged as pivotal players in leukemogenesiswith a rich body of evidence to link aberrant expression of theParaHox gene Cdx2 and other ParaHox genes with acute leukemia(Table 1; Figure 2).

GSH2

The Gsh2 gene was previously characterized as a brain-specificParaHox gene important in forebrain and hindbrain formation of themouse. Peter Marynen’s group (Cools et al30) has shown that GSH2 isnot expressed in adult human normal hematopoietic cells. However, thegroup has reported ectopic expression of GSH2 in 4 patients with AMLcarrying the translocation t(4;12)(q11-q12;p13), which itself generatedno detectable fusion gene. GSH2 is located at 4q11-q12 and thus locatedclosely to the respective breakpoint, which implies that the translocationinduces ectopic expression of GSH2. Indeed the researchers coulddemonstrate that aberrant expression of GSH2 caused focus formationin the NIH3T3 assay as a surrogate marker for the oncogenic potential ofthis ParaHox gene.30

Cdx1

In normal adult hematopoietic tissue, expression of this homeoboxgene was not reported so far, and there are no data to suggest

Normal Hematopoiesis

21444

Stem cell

TransformedProgenitors

Stem cell

Leukemia initiating cells

Transformedstem cells

CDCDXX22

CDCDCDXX22

Acute Myeloid LeukemiaHOX

CDX4

CDX2

CDX120 10080604000

% of positive patients

Cdx/CDXMalignant Hematopoiesisp

HOX expression

CDX4 expression

CDX2 No expression

CDX1 No expression

Stem cell Progenitor Oligo lineageprecursor

Maturecells

Cdx/CDX

214444

A B

Figure 2. Cdx expression in normal and malignant hematopoiesis. (A) Normal hematopoiesis.Cdx4 and HOX A and B cluster genes are expressed in normal stem andprogenitor cells, whereas Cdx2 and Cdx1 remain silenced. (B) Malignant hematopoiesis. Ectopic expression of CDX2 at the level of stem or progenitor cells induces increasedexpression of Hox A and B cluster genes and leukemic transformation of hematopoietic cells. The proportion of patients with AML, which ectopically express CDX2 andoverexpress HOX genes, is indicated. CDX4 is expressed only in a minority of patients, and CDX1 expression is not detectable in patients with leukemia.

522 RAWAT et al BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

involvement of Cdx1 in malignant hematopoiesis. However, in theadult Cdx1 is known to be involved in the intestinal differentiationand the development of intestinal metaplasias.54-56

CDX2

Aberrant expression of CDX2 is a frequent event in patients withAML. Several studies have reported that CDX2 is normally notexpressed in adult murine and human normal hematopoiesis(Figure 2). Ectopic expression of CDX2 in AML was reported byChase et al in 1999 in a patient with AML M1 carrying thechromosomal translocation t(12;13)(p13;q12), creating a novelfusion of ETV6 and CDX2.35 In addition, aberrant expression ofCDX2 was found in a case of blast crisis of chronic myeloidleukemia (CML-BC) with no detectable chromosomal abnormal-ity.35 The fusion partner ETV6 is an important regulator of HSCsurvival and frequently affected by translocations.57,58 Later, weand others have reported that CDX2 is aberrantly expressed in� 80% of human AML patients, characterizing CDX2 as one of themost frequent aberrantly expressed proto-oncogenes in humanleukemia (Figure 2). Aberrant expression of CDX2 was found in alltypes of cytogenetic subgroups. Scholl et al reported the highestexpression of CDX2 in patients with t(9;11) followed by normalkaryotype AML (CN-AML) and t(15;17)–positive patients,36 simi-lar to our results with highest expression of CDX2 in CN-AMLfollowed by t(9;11)– and t(15;17)–positive cases.33 Aberrant expres-sion of CDX2 was not only more common in patients withCN-AML than in patients with abnormal karyotype (89% vs 64%,respectively), but the expression level of CDX2 was significantlyand � 15-fold higher in patients with CN-AML as well. Further-more, the level of CDX2 expression correlated with aberrant HOXgene expression in human AML.33 The analyses did not show anycorrelation of the NPM1 mutational status with the frequency andlevel of aberrant CDX2 expression. Aberrant CDX2 expression isalso detectable in most of the human AML cell lines. Themechanisms responsible for disruption of the normal CDX2silencing in human AML are unknown and are not because ofpromoter mutation, gene amplification, or abnormal promotermethylation.36 In contrast to CDX2, all CN-AML cases werenegative for expression of the other members of the CDX genefamily CDX1 and CDX4.33

CDX2 is aberrantly expressed in most patients with acutelymphoblastic leukemia (ALL). Ectopic expression of CDX2 isnot restricted to patients with AML; we showed that CDX2 isaberrantly expressed in most adult patients with ALL, with 81%overall positivity and a higher overall expression level thandetectable in patients with CN-AML. The expression of CDX2showed high variation between the different ALL subgroups(classified according to the criteria of the European Group for theImmunologic Characterization of Leukemias59) with highest me-dian expression in pre-T-ALL, followed by c-ALL and pro-B-ALL.In contrast, patients with B-ALL/Burkitt lymphoma and thymicT-ALL exhibited only low CDX2 expression. In addition, thenumber of patients aberrantly expressing CDX2 differed consider-ably between ALL subtypes, with 100% positivity in patients withpro-B-ALL, c-ALL, and Ph� ALL versus only 40% positivity inB-ALL/Burkitt lymphoma and � 70% in thymic and pre-T-ALL.60

The expression level of CDX2 in patients with ALL significantlycorrelated with treatment outcome (high expression levels wereassociated with inferior overall survival) of the analyzed patients.Although that study was limited by a low number of patients,CDX2 remained an independent risk factor even after adjusting forthe risk factors of age and presence of molecular markers by

bivariate Cox regression.60 These data were confirmed for pediatricALL whereby most patients also showed aberrant expression ofCDX2. Also in this study, aberrant expression of CDX2 signifi-cantly correlated to an inferior disease outcome.61 Another recentindependent study reported CDX2 expression in 61% of patientswith de novo adult ALL and decrease of CDX2 expression inpatients with complete remission and increase in relapse.62

Similar to AML the mechanism of aberrant CDX2 expression inALL is not understood. We have not identified any difference inthe methylation level of the CDX2 promoter in patients withCDX2-positive and -negative ALL.60 In contrast to AML,whereby aberrant CDX2 expression correlated with HOX genederegulation, there was only a low correlation between theexpression of HOX genes with the expression of CDX2, in linewith reports that HOX gene deregulation is less common in ALLthan in AML.63,64 Taken together, these data link aberrantexpression of CDX2 to myeloid as well as lymphoid leukemia.So far it is unknown whether the stage of differentiation of thecell, which is initially hit by aberrant CDX2 expression,determines the phenotype of the leukemia. Experiments testingwhether aberrant expression of CDX2 in lymphoid progenitorcells generates ALL in mice are lacking so far.

Cdx2 is highly leukemogenic in experimental models. In amurine BM transplantation model it was shown that ectopicexpression of the ParaHox gene CDX2, but not expression of thefusion gene ETV6/CDX2, which was described in a patient withtranslocation t(12;13)(p13;q12)–positive AML,35 was highlyleukemogenic.37 This was the first direct evidence that theParaHox gene Cdx2 is highly leukemogenic and induces aggres-sive AML when aberrantly expressed in hematopoietic progeni-tor cells. Retrovirally enforced expression of Cdx2 in murinehematopoietic progenitors resulted in increased expression ofleukemogenic Hox genes such as Hoxa5, Hoxa7, Hoxa9,Hoxa10, as well as Hoxb3, Hoxb6, and Hoxb8.33 These data arein agreement with previous findings that Cdx2 can deregulateHox gene expression in embryonic as well as adult nonhemato-poietic tissues.45,65 The ability to up-regulate Hox gene expres-sion and to induce AML clearly depended the N-terminaltransactivation domain of Cdx2, known to harbor serine phos-phorylation sites, which affect the transcriptional activity of thegene.33,66 Interestingly, the ETV6-CDX2 fusion gene also lacksthe N-terminal transactivation domain of CDX2, which wouldexplain the obvious discrepancy of the oncogenic potentialbetween Cdx2 and the ETV6-CDX2 fusion.

Aberrant Cdx2 expression also led to an altered expression of anumber of genes involved in lymphopoiesis such as Lef1, Tcf3, orId3. These genes are frequently deregulated in lymphoid leuke-mia60,67,68 and are able in the case of Lef1 to induce ALL in micethat received a transplant.33

Most data on the leukemogenicity of Cdx2 come frommurine models. However, at least at the level of human leukemiacell lines it could be shown that leukemic growth depends onaberrant CDX2 expression; thus, shRNA-mediated depletion ofCDX2 in the human pre-B cell line Nalm-6 and several AML celllines significantly impaired leukemic cell proliferation andclonogenicity.56,60

CDX4 is less leukemogenic than Cdx2. It has been shown thatoverexpression of Cdx4 confers enhanced replating and prolifera-tive potential to transduced murine hematopoietic cells. Micetransplanted with Cdx4-overexpressing BM developed a myelopro-liferative disorder, which progressed to AML over time.39 In thismodel aberrant expression of Cdx4 had to collaborate with the

PARAHOX GENES IN NORMAL AND MALIGNANT HEMATOPOIESIS 523BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

homeobox gene Meis1 to induce AML in all experimentalanimals,39 thus clearly differing from Cdx2, which caused rapiddisease in all mice that received a transplant.37 In line with this,the long latency of leukemia development in primary animalsthat received a transplant suggested that aberrant expression ofCdx4 alone is not sufficient to induce leukemia but ratherinduces a preleukemic stage. Loss of Cdx4 significantly pro-longed the latency of disease onset in a mouse model ofmixed-lineage leukemia-AF9–induced AML and altered thephenotype of the resultant disease with increased expression ofB- and T-lymphoid markers.40 The mechanism of Cdx4-inducedleukemogenesis is not completely understood. However, re-cently it has been shown that CDX4 activates transcription ofthe Hox gene HOXA10 in myeloid cells and binds to a ciselement of its promoter. Vice versa HOXA10 is able to activateCDX4 expression, pointing to a positive feedback loop betweenCDX4 and a HOX gene, which itself is linked to myeloidleukemogenesis.17,69,70 It was also found that Cdx4 collaborateswith Menin to activate Hoxa9 expression by binding with Meninat the same regulatory region at the Hoxa9 locus.71 In summary,these data categorize CDX4 as an important collaborativepartner in Hox-associated leukemogenesis.

In patients with AML CDX4 is expressed aberrantly in � 25%of all cases (Figure 2). CDX4 expression, however, did notcorrelate to a certain leukemia subtype, and its expression did notcorrelate to HOX gene deregulation.39 This finding suggests thatthe ability of CDX4 to deregulate HOX gene expression is of minorrelevance for its role in human leukemia.

Mechanisms of ParaHox-induced leukemogenesis:speculations and future directions

Gene expression profiling and functional data clearly support acritical involvement of CDX2 in human leukemogenesis. However,molecular mechanisms of Cdx2-induced leukemogenesis are poorlyunderstood. One obvious possibility is that CDX2 exerts itsleukemogenic potential by deregulating expression of leukemo-genic genes of the HOXA and HOXB cluster, known to bedysregulated in 65% of all AML cases, in particular in normalkaryotype (CN-AML) and mixed-lineage leukemia–rearrangedAML33,36 (Figure 2). This would classify the CDX-HOX axis asone of the main driving forces in human AML. However, it couldbe clearly shown that aberrant Cdx2 expression, besides Hoxderegulation, also led to an altered expression of a number of genesinvolved in embryonic development, lymphopoiesis, Wnt pathway,and genes involved in stem cell function. These data support thehypothesis that Cdx2 might deregulate multiple pathways inleukemia, which would also explain the differences between Cdx2and Cdx4 in their leukemogenic potential.

So far strategies to target Cdx2-induced transformation remainelusive. One possibility would be to identify signaling pathwaysthat are able to be affected by drugs on which Cdx2 activity isdependent. Interestingly, it was shown in a colon cancer cell linethat the transcriptional activity of Cdx2 depends on phosphoryla-tion of the N-terminal S60 site by ERK kinases. We have shownthat the N-terminal activation domain of Cdx2, at which the S60site is localized, is essential for Cdx2-induced leukemogenesisand perturbation of Hox gene expression.33 Recent studiesdocumented that Erk1/2 and their upstream effectors Mek1/2 areconstitutively activated in primary human AML cells.72-74

Elevated p-ERK levels were found in 83.3% of the patients withAML.75 Because aberrant expression of CDX2 is detected in80% of all AML cases, regulation of Cdx2 could be indirectly ordirectly linked with MEK/ERK signaling72,76 (Figure 3A-B).

Therefore, it will be interesting to test whether post-transcriptional modification of CDX2 is crucial for its leukemo-genic properties and whether MAPK inhibitors are able to blockCdx2-induced leukemogenesis.

Another possibility to block Cdx2-induced leukemogenesiswould be to inhibit mechanisms that induce and maintain aberrantCdx2 expression. However, mechanisms that lead to aberrantexpression of CDX2 in leukemogenesis are still a mystery. Asreported by Scholl et al, the methylation levels of the promoterregion of CDX2 in patients with AML did not correspond to itsexpression levels.36 In addition, treatment of CDX2-negative celllines with demethylating agents did not lead to an induction ofCDX2 expression,77 suggesting that CDX2 expression in leukemiais independent of promoter hypomethylation as well as of themethylation level of cis-regulatory elements. Amplification of theCDX2 locus was detected only in a small percentage of patientswith complex karyotype, however, not in AML patients withnormal karyotype, who are characterized by high CDX2 expressionlevels.36,78 Another possibility, leading to aberrant transcription ofCDX2 could be deregulation of trans-acting pathways, whichtarget cis-regulatory elements in the 5� region of the CDX2 gene assuggested by Hinoi et al.77 However, it was reported that CDX2expression is monoallelic in most of the analyzed patients withleukemia,36 which indicates that the altered CDX2 expression is notprimarily because of differences in the expression of trans-actingfactors but rather because of altered cis-regulatory elements,which affect only one allele. Mutations were not detected in thepromoter region of CDX2. However, this does not exclude stillunknown mutations in cis-regulatory elements of the CDX2gene, which cause aberrant expression. It would be interesting,therefore, to expand the sequence analysis of the CDX2 genebeyond its coding and promoter region and to conduct a morecomprehensive sequence analysis. One target of genetic altera-tions, for example, could be negative regulators, which Wangand Shashikant postulated to be present in an 11.4-kb sequenceflanking Cdx2.79 Disruption of these elements in hematopoieticcells could lead to an aberrant expression of the normallysilenced CDX2 as well as mutations within 2 enhancer regions inthe first intron of CDX2, described to induce ectopic expression ofCdx2 in the murine embryo.79 Interestingly, several binding sitesfor fibroblast growth factor (FGF) and Wnt signaling wereidentified in these regions. Mutations therefore could possiblyaffect the binding of factors such as FGF4, which previouslywas shown to target Cdx280 (Figure 3). Furthermore, it isprobable that Cdx2 not only regulates but also is targeted by Wntsignaling as already demonstrated for other ParaHox genes suchas Cdx1 and Cdx4.81,82 Another pathway that was shown toregulate Cdx2 expression is Ras/MAPK signaling,76 a pathwaythat links extracellular stimuli among others to proliferation,differentiation, and cell survival. Because both the Wnt andRas/MAPK pathways are frequently deregulated in leukemia,this deregulation in combination with genetic alterations in thecis-regulatory elements of Cdx2 could possibly affect thetranscription of the gene (Figure 3).

Conclusion

In summary, there is a great body of evidence that ParaHox genesplay a pivotal role in normal hematopoiesis but also in leukemogen-esis. These data add another homeobox gene family to the growingnumber of homeobox genes known to be critically involved inorchestrating normal hematopoiesis and promoting leukemogen-esis. Thus, the story “homeobox genes in normal and malignant

524 RAWAT et al BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

hematopoiesis” is not at its end but is still growing rapidly since thelast review on this topic in this journal nearly 10 years ago.1 Inthose 10 years dysregulation of homeobox genes has been recog-nized as one of the molecular hallmarks of human AML. We knowtoday that homeobox genes are key players for determining“stemness” and are essential for initiating and maintaining self-renewal programs on which leukemic stem cell function depends.83

ParaHox genes such as Cdx2 are particularly powerful to transformhematopoietic progenitors into leukemic stem cells, at least partlybecause it stands upstream of Hox genes and is able to perturbexpression of several Hox genes at the same time. It is intriguing tospeculate that targeting dysfunctional ParaHox genes such asCDX2 in human leukemia might reverse HOX gene dysfunction ona larger scale, thereby depriving the leukemic stem cell the

LRP5/6

Frizzled

Wnt

-Catenin

-Catenin

Hox/HOX

Cdx/CDX

AB

221144

FGFR

L1

SEF

FRS2Ras/Raf

MAPK

FGFR

AB

221144

FGF

Hox/HOX

Cdx/CDX

Wnt Signalling FGF Signalling

?

?? Model based on embryonic studies

??

??

stimulates blocks

Cytoplasm

Nucleus

TCF/LEF

LRP5/6

Frizzled

Wnt

-Catenin

-Catenin

Hox/HOX

Cdx/CDX

AB

211444

FGFR

L1

SEF

FRS2Ras/Raf

MAPK

FGFR

AB

211444

FGF

Hox/HOX

Cdx/CDX

Cytoplasm

Nucleus

Wnt Signalling FGF Signalling

Mutation

??

?? Model based on embryonic studies

??

stimulates blocks

??

Normal hematopoiesis A

B Malignant hematopoiesis (hypothetical model)

Figure 3. Hypothetical model of Cdx gene regulation in normal and malignant hematopoiesis. (A) Normal hematopoiesis, based on results from studies inembryogenesis, the canonical Wnt pathway regulates the expression of CDX4 by the Lef1/Tcf complex in hematopoiesis. The FGF-MAPK pathway is not constitutively active instem and progenitors cells. (B) Malignant hematopoiesis, either deregulation of the WNT pathway or FGF signaling induces aberrant CDX2 expression in human leukemia.

PARAHOX GENES IN NORMAL AND MALIGNANT HEMATOPOIESIS 525BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

molecular “environment” it needs to propagate leukemic growth. Itis indeed unfortunate that we know so little about the mechanismsthat induce aberrant CDX2 expression in patients with AML,because this would open possibilities to develop strategies tore-silence CDX2 expression in leukemic cells. Another hope is toidentify signaling cascades that are essential for CDX2-inducedtransformation. Many questions are still open, and clarification ofthe role of other members of the ParaHox gene cluster such as GSH2 forleukemogenesis was just initiated. We can be confident that the nextyears will generate exciting data about the role of ParaHox genes fornormal tissue homeostasis and will improve our understanding of howto correct ParaHox dysfunction in human leukemias.

AcknowledgmentsThe authors apologize to authors whose work could not be citedbecause of space constraints. They thank all members of their

institutes for vivid discussions on this topic. They also thank SilviaThoene, Shiva Bamezai, and Tamoghna Mandal for assisting in thepreparation of figures.

This work is supported by grants from the Else Kroner-Fresenius-Foundation, the Deutsche Forschungsgemeinschaft, and the Bundesmin-isterium fur Bildung und Forschung (V.P.S.R. and C.B.).

Authorship

Contribution: V.P.S.R., R.K.H, and C.B. contributed to the writingof the manuscript.

Conflict-of-interest disclosure: The authors declare no compet-ing financial interests.

Correspondence: Christian Buske, Institute of ExperimentalCancer Research, University of Ulm, Albert-Einstein-Allee 11,89081 Ulm, Germany; e-mail: christian.buske@uni-ulm.de.

References

1. Lawrence HJ, Largman C. Homeobox genes innormal hematopoiesis and leukemia. Blood.1992;80(10):2445-2453.

2. Bridges CB. Current maps of the location of themutant genes of Drosophila melanogaster. ProcNatl Acad Sci U S A. 1921;7(4):127-132.

3. Stein S, Fritsch R, Lemaire L, Kessel M. Check-list: vertebrate homeobox genes. Mech Dev.1996;55(1):91-108.

4. Tupler R, Perini G, Green MR. Expressing thehuman genome. Nature. 2001;409(6822):832-833.

5. Abate-Shen C. Deregulated homeobox gene ex-pression in cancer: cause or consequence? NatRev Cancer. 2002;2(10):777-785.

6. Garcia-Fernandez J. The genesis and evolutionof homeobox gene clusters. Nat Rev Genet.2005;6(12):881-892.

7. Argiropoulos B, Humphries RK. Hox genes in he-matopoiesis and leukemogenesis. Oncogene.2007;26(47):6766-6776.

8. Owens BM, Hawley RG. HOX and non-HOX ho-meobox genes in leukemic hematopoiesis. StemCells. 2002;20(5):364-379.

9. Sauvageau G, Lansdorp PM, Eaves CJ, et al.Differential expression of homeobox genes infunctionally distinct CD34� subpopulations ofhuman bone marrow cells. Proc Natl Acad SciU S A. 1994;91(25):12223-12227.

10. Pineault N, Helgason CD, Lawrence HJ,Humphries RK. Differential expression of Hox,Meis1, and Pbx1 genes in primitive cells through-out murine hematopoietic ontogeny. Exp Hema-tol. 2002;30(1):49-57.

11. Lawrence HJ, Christensen J, Fong S, et al. Lossof expression of the Hoxa-9 homeobox gene im-pairs the proliferation and repopulating ability ofhematopoietic stem cells. Blood. 2005;106(12):3988-3994.

12. Lawrence HJ, Helgason CD, Sauvageau G, et al.Mice bearing a targeted interruption of the ho-meobox gene HOXA9 have defects in myeloid,erythroid, and lymphoid hematopoiesis. Blood.1997;89(6):1922-1930.

13. Thorsteinsdottir U, Sauvageau G, Humphries RK.Hox homeobox genes as regulators of normaland leukemic hematopoiesis. Hematol Oncol ClinNorth Am. 1997;11(6):1221-1237.

14. Thorsteinsdottir U, Kroon E, Jerome L, Blasi F,Sauvageau G. Defining roles for HOX and MEIS1genes in induction of acute myeloid leukemia.Mol Cell Biol. 2001;21(1):224-234.

15. Antonchuk J, Sauvageau G, Humphries RK.HOXB4-induced expansion of adult hematopoi-etic stem cells ex vivo. Cell. 2002;109(1):39-45.

16. Buske C, Feuring-Buske M, Abramovich C, et al.Deregulated expression of HOXB4 enhances theprimitive growth activity of human hematopoieticcells. Blood. 2002;100(3):862-868.

17. Buske C, Feuring-Buske M, Antonchuk J, et al.Overexpression of HOXA10 perturbs human lym-phomyelopoiesis in vitro and in vivo. Blood. 2001;97(8):2286-2292.

18. Lawrence HJ, Sauvageau G, Ahmadi N, et al.Stage- and lineage-specific expression of theHOXA10 homeobox gene in normal and leukemichematopoietic cells. Exp Hematol. 1995;23(11):1160-1166.

19. Haferlach C, Mecucci C, Schnittger S, et al. AMLwith mutated NPM1 carrying a normal or aberrantkaryotype show overlapping biologic, pathologic,immunophenotypic, and prognostic features.Blood. 2009;114(14):3024-3032.

20. Golub TR, Slonim DK, Tamayo P, et al. Molecularclassification of cancer: class discovery and classprediction by gene expression monitoring. Sci-ence. 1999;286(5439):531-537.

21. Rice KL, Licht JD. HOX deregulation in acute my-eloid leukemia. J Clin Invest. 2007;117(4):865-868.

22. Krivtsov AV, Armstrong SA. MLL translocations,histone modifications and leukaemia stem-celldevelopment. Nat Rev Cancer. 2007;7(11):823-833.

23. Lawrence HJ, Sauvageau G, Humphries RK,Largman C. The role of HOX homeobox genes innormal and leukemic hematopoiesis. Stem Cells.1996;14(3):281-291.

24. Banerjee-Basu S, Baxevanis AD. Molecular evo-lution of the homeodomain family of transcriptionfactors. Nucleic Acids Res. 2001;29(15):3258-3269.

25. Argiropoulos B, Yung E, Humphries RK. Unravel-ing the crucial roles of Meis1 in leukemogenesisand normal hematopoiesis. Genes Dev. 2007;21(22):2845-2849.

26. Brooke NM, Garcia-Fernandez J, Holland PW.The ParaHox gene cluster is an evolutionary sis-ter of the Hox gene cluster. Nature. 1998;392(6679):920-922.

27. Ferrier DE, Dewar K, CookA, Chang JL, Hill-ForceA,Amemiya C. The chordate ParaHox cluster. CurrBiol. 2005;15(20):R820-822.

28. Cook CE, Jimenez E, Akam M, Salo E. The Hoxgene complement of acoel flatworms, a basal bi-laterian clade. Evol Dev. 2004;6(3):154-163.

29. Li H, Zeitler PS, Valerius MT, Small K, Potter SS.Gsh-1, an orphan Hox gene, is required for nor-mal pituitary development. EMBO J. 1996;15(4):714-724.

30. Cools J, Mentens N, Odero MD, et al. Evidencefor position effects as a variant ETV6-mediatedleukemogenic mechanism in myeloid leukemiaswith a t(4;12)(q11-q12;p13) or t(5;12)(q31;p13).Blood. 2002;99(5):1776-1784.

31. Szucsik JC, Witte DP, Li H, Pixley SK, Small KM,Potter SS. Altered forebrain and hindbrain devel-opment in mice mutant for the Gsh-2 homeoboxgene. Dev Biol. 1997;191(2):230-242.

32. Offield MF, Jetton TL, Labosky PA, et al. PDX-1 isrequired for pancreatic outgrowth and differentia-tion of the rostral duodenum. Development. 1996;122(3):983-995.

33. Rawat VP, Thoene S, Naidu VM, et al. Overex-pression of CDX2 perturbs HOX gene expressionin murine progenitors depending on its N-terminaldomain and is closely correlated with deregulatedHOX gene expression in human acute myeloidleukemia. Blood. 2008;111(1):309-319.

34. Subramanian V, Meyer BI, Gruss P. Disruption ofthe murine homeobox gene Cdx1 affects axialskeletal identities by altering the mesodermal ex-pression domains of Hox genes. Cell. 1995;83(4):641-653.

35. Chase A, Reiter A, Burci L, et al. Fusion of ETV6to the caudal-related homeobox gene CDX2 inacute myeloid leukemia with the t(12;13)(p13;q12). Blood. 1999;93(3):1025-1031.

36. Scholl C, Bansal D, Dohner K, et al. The homeo-box gene CDX2 is aberrantly expressed in mostcases of acute myeloid leukemia and promotesleukemogenesis. J Clin Invest. 2007;117(4):1037-1048.

37. Rawat VP, Cusan M, Deshpande A, et al. Ectopicexpression of the homeobox gene Cdx2 is thetransforming event in a mouse model of t(12;13)(p13;q12) acute myeloid leukemia. Proc NatlAcad Sci U S A. 2004;101(3):817-822.

38. Chawengsaksophak K, James R, Hammond VE,Kontgen F, Beck F. Homeosis and intestinal tu-mours in Cdx2 mutant mice. Nature. 1997;386(6620):84-87.

39. Bansal D, Scholl C, Frohling S, et al. Cdx4 dys-regulates Hox gene expression and generatesacute myeloid leukemia alone and in cooperationwith Meis1a in a murine model. Proc Natl AcadSci U S A. 2006;103(45):16924-16929.

40. Koo S, Huntly BJ, Wang Y, et al. Cdx4 is dispens-able for murine adult hematopoietic stem cells butpromotes MLL-AF9-mediated leukemogenesis.Haematologica. 2010;95(10):1642-1650.

41. van Nes J, de Graaff W, Lebrin F, Gerhard M,Beck F, Deschamps J. The Cdx4 mutation affectsaxial development and reveals an essential roleof Cdx genes in the ontogenesis of the placental

526 RAWAT et al BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

labyrinth in mice. Development. 2006;133(3):419-428.

42. Furlong RF, Mulley JF. ParaHox cluster evolu-tion–hagfish and beyond. Zoolog Sci. 2008;25(10):955-959.

43. Garcia-Fernandez J. Hox, ParaHox, ProtoHox:facts and guesses. Heredity (Edinb). 2005;94(2):145-152.

44. An R, da Silva Xavier G, Semplici F, et al. Pancre-atic and duodenal homeobox 1 (PDX1) phosphor-ylation at serine-269 is HIPK2-dependent andaffects PDX1 subnuclear localization. BiochemBiophys Res Commun. 2010;399(2):155-161.

45. Chawengsaksophak K, de Graaff W, Rossant J,Deschamps J, Beck F. Cdx2 is essential for axialelongation in mouse development. Proc NatlAcad Sci U S A. 2004;101(20):7641-7645.

46. Tamai Y, Nakajima R, Ishikawa T, Takaku K,Seldin MF, Taketo MM. Colonic hamartoma de-velopment by anomalous duplication in Cdx2knockout mice. Cancer Res. 1999;59(12):2965-2970.

47. McGinnis W, Krumlauf R. Homeobox genes andaxial patterning. Cell. 1992;68(2):283-302.

48. Davidson AJ, Ernst P, Wang Y, et al. cdx4 mu-tants fail to specify blood progenitors and can berescued by multiple hox genes. Nature. 2003;425(6955):300-306.

49. Davidson AJ, Zon LI. The caudal-related homeo-box genes cdx1a and cdx4 act redundantly toregulate hox gene expression and the formationof putative hematopoietic stem cells during ze-brafish embryogenesis. Dev Biol. 2006;292(2):506-518.

50. Wang Y, Yates F, Naveiras O, Ernst P, Daley GQ.Embryonic stem cell-derived hematopoietic stemcells. Proc Natl Acad Sci U S A. 2005;102(52):19081-19086.

51. Ernst P, Mabon M, Davidson AJ, Zon LI,Korsmeyer SJ. An Mll-dependent Hox programdrives hematopoietic progenitor expansion. CurrBiol. 2004;14(22):2063-2069.

52. Wang Y, Yabuuchi A, McKinney-Freeman S, et al.Cdx gene deficiency compromises embryonichematopoiesis in the mouse. Proc Natl Acad SciU S A. 2008;105(22):7756-7761.

53. McKinney-Freeman SL, Lengerke C, Jang IH,et al. Modulation of murine embryonic stem cell-derived CD41�c-kit� hematopoietic progenitorsby ectopic expression of Cdx genes. Blood. 2008;111(10):4944-4953.

54. Wong NA, Wilding J, Bartlett S, et al. CDX1 is animportant molecular mediator of Barrett’s meta-plasia. Proc Natl Acad Sci U S A. 2005;102(21):7565-7570.

55. Bonhomme C, Calon A, Martin E, et al. Cdx1, adispensable homeobox gene for gut developmentwith limited effect in intestinal cancer. Oncogene.2008;27(32):4497-4502.

56. Silberg DG, Swain GP, Suh ER, Traber PG. Cdx1and cdx2 expression during intestinal develop-ment. Gastroenterology. 2000;119(4):961-971.

57. Hock H, Meade E, Medeiros S, et al. Tel/Etv6 isan essential and selective regulator of adult he-matopoietic stem cell survival. Genes Dev. 2004;18(19):2336-2341.

58. Bohlander SK. ETV6: a versatile player in leuke-mogenesis. Semin Cancer Biol. 2005;15(3):162-174.

59. Bene MC, Castoldi G, Knapp W, et al. Proposalsfor the immunological classification of acute leu-kemias. European Group for the ImmunologicalCharacterization of Leukemias (EGIL). Leukemia.1995;9(10):1783-1786.

60. Thoene S, Rawat VP, Heilmeier B, et al. The ho-meobox gene CDX2 is aberrantly expressed andassociated with an inferior prognosis in patientswith acute lymphoblastic leukemia. Leukemia.2009;23(4):649-655.

61. Riedt T, Ebinger M, Salih HR, et al. Aberrant ex-pression of the homeobox gene CDX2 in pediat-ric acute lymphoblastic leukemia. Blood. 2009;113(17):4049-4051.

62. Lu HY, Xia HL, Chen XW, Zhu LX, Wang QY,Cheng X. [Expression and clinical significance ofcaudal type homeobox transcription factor-2 inadult acute lymphocytic leukemia] (in Chinese).Zhongguo Shi Yan Xue Ye Xue Za Zhi. 2011;19(2):298-302.

63. Armstrong SA, Hsieh JJ, Korsmeyer SJ. Genomicapproaches to the pathogenesis and treatment ofacute lymphoblastic leukemias. Curr Opin Hema-tol. 2002;9(4):339-344.

64. Quentmeier H, Dirks WG, Macleod RA, Reinhardt J,Zaborski M, Drexler HG. Expression of HOX genesin acute leukemia cell lines with and without MLLtranslocations. Leuk Lymphoma. 2004;45(3):567-574.

65. Liu T, Zhang X, So CK, et al. Regulation of Cdx2expression by promoter methylation, and effectsof Cdx2 transfection on morphology and geneexpression of human esophageal epithelial cells.Carcinogenesis. 2007;28(2):488-496.

66. Rings EH, Boudreau F, Taylor JK, Moffett J, Suh ER,Traber PG. Phosphorylation of the serine 60 residuewithin the Cdx2 activation domain mediates its trans-activation capacity. Gastroenterology. 2001;121(6):1437-1450.

67. Mullighan CG, Goorha S, Radtke I, et al.Genome-wide analysis of genetic alterations inacute lymphoblastic leukaemia. Nature. 2007;446(7137):758-764.

68. Ferrando AA, Look AT. Gene expression profilingin T-cell acute lymphoblastic leukemia. SeminHematol. 2003;40(4):274-280.

69. Bei L, Huang W, Wang H, Shah C, Horvath E,Eklund E. HoxA10 activates CDX4 transcriptionand Cdx4 activates HOXA10 transcription in my-

eloid cells. J Biol Chem. 2011;286(21):19047-19064.

70. Thorsteinsdottir U, Sauvageau G, Hough MR,et al. Overexpression of HOXA10 in murine he-matopoietic cells perturbs both myeloid and lym-phoid differentiation and leads to acute myeloidleukemia. Mol Cell Biol. 1997;17(1):495-505.

71. Yan J, Chen YX, Desmond A, et al. Cdx4 andmenin co-regulate Hoxa9 expression in hemato-poietic cells. PLoS One. 2006;1:e47.

72. Milella M, Kornblau SM, Estrov Z, et al. Thera-peutic targeting of the MEK/MAPK signal trans-duction module in acute myeloid leukemia. J ClinInvest. 2001;108(6):851-859.

73. Wu J, Wong WW, Khosravi F, Minden MD, Penn LZ.Blocking the Raf/MEK/ERK pathway sensitizesacute myelogenous leukemia cells to lovastatin-induced apoptosis. Cancer Res. 2004;64(18):6461-6468.

74. Rubinfeld H, Seger R. The ERK cascade: a proto-type of MAPK signaling. Mol Biotechnol. 2005;31(2):151-174.

75. Ricciardi MR, McQueen T, Chism D, et al. Quanti-tative single cell determination of ERK phosphor-ylation and regulation in relapsed and refractoryprimary acute myeloid leukemia. Leukemia.2005;19(9):1543-1549.

76. Lu CW, Yabuuchi A, Chen L, Viswanathan S, Kim K,Daley GQ. Ras-MAPK signaling promotes tro-phectoderm formation from embryonic stem cellsand mouse embryos. Nat Genet. 2008;40(7):921-926.

77. Hinoi T, Loda M, Fearon ER. Silencing of CDX2expression in colon cancer via a dominant re-pression pathway. J Biol Chem. 2003;278(45):44608-44616.

78. Rucker FG, Bullinger L, Schwaenen C, et al. Dis-closure of candidate genes in acute myeloid leu-kemia with complex karyotypes using microarray-based molecular characterization. J Clin Oncol.2006;24(24):3887-3894.

79. Wang WC, Shashikant CS. Evidence for positiveand negative regulation of the mouse Cdx2 gene.J Exp Zool B Mol Dev Evol. 2007;308(3):308-321.

80. Murohashi M, Nakamura T, Tanaka S, et al. AnFGF4-FRS2alpha-Cdx2 axis in trophoblast stemcells induces Bmp4 to regulate proper growth ofearly mouse embryos. Stem Cells. 2010;28(1):113-121.

81. Shimizu T, Bae YK, Muraoka O, Hibi M. Interac-tion of Wnt and caudal-related genes in zebrafishposterior body formation. Dev Biol. 2005;279(1):125-141.

82. Pilon N, Oh K, Sylvestre JR, Bouchard N, Savory J,Lohnes D. Cdx4 is a direct target of the canonicalWnt pathway. Dev Biol. 2006;289(1):55-63.

83. Heuser M, Yun H, Berg T, et al. Cell of origin inAML: susceptibility to MN1-induced transforma-tion is regulated by the MEIS1/AbdB-like HOXprotein complex. Cancer Cell. 2011;20(1):39-52.

PARAHOX GENES IN NORMAL AND MALIGNANT HEMATOPOIESIS 527BLOOD, 19 JULY 2012 � VOLUME 120, NUMBER 3

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom

online April 30, 2012 originally publisheddoi:10.1182/blood-2012-02-385898

2012 120: 519-527  

Vijay P. S. Rawat, R. Keith Humphries and Christian Buske hematopoiesisBeyond Hox: the role of ParaHox genes in normal and malignant 

http://www.bloodjournal.org/content/120/3/519.full.htmlUpdated information and services can be found at:

(696 articles)Review Articles    (1651 articles)Myeloid Neoplasia   

Articles on similar topics can be found in the following Blood collections

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

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

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

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

For personal use only.on April 12, 2017. by guest www.bloodjournal.orgFrom