of April 28, 2018.This information is current as

Hematopoietic Lineage Choice by HEBAltContext-Dependent Regulation of

Michele K. AndersonBraunstein, Amanda J. Moore, Mikael Sigvardsson and Duncheng Wang, Carol L. Claus, Paula Rajkumar, Marsela

http://www.jimmunol.org/content/185/7/4109doi: 10.4049/jimmunol.0901783September 2010;

2010; 185:4109-4117; Prepublished online 8J Immunol 

MaterialSupplementary

3.DC1http://www.jimmunol.org/content/suppl/2010/09/07/jimmunol.090178

Referenceshttp://www.jimmunol.org/content/185/7/4109.full#ref-list-1

, 36 of which you can access for free at: cites 70 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2010 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Context-Dependent Regulation of Hematopoietic LineageChoice by HEBAlt

Duncheng Wang,*,†,1 Carol L. Claus,*,2 Paula Rajkumar,* Marsela Braunstein,*,†

Amanda J. Moore,*,† Mikael Sigvardsson,‡ and Michele K. Anderson*,†

Hematopoietic development is controlled by combinatorial interactions between E-protein transcription factors and other lineage

regulators that operate in the context of gene-regulatory networks. The E-proteins HEB and E2A are critical for T cell and B cell

development, but the mechanisms by which their activities are directed to different genes in each lineage are unclear. We found

that a short form of HEB, HEBAlt, acts downstream of Delta-like (DL)-Notch signaling to promote T cell development. In this

paper, we show that forced expression of HEBAlt in mouse hematopoietic progenitors inhibited B cell development, but it allowed

them to adopt a myeloid fate. HEBAlt interfered with the activity of E2A homodimers and with the expression of the transcription

factor Pax5, both of which are critical for B cell development. However, when combined with DL-Notch signaling, HEBAlt

enhanced the generation of T cell progenitors at the expense of myeloid cells. The longer form of HEB, HEBCan, also inhibited

E47 activity and Pax5 expression, but it did not collaborate with DL-Notch signaling to suppress myeloid potential. There-

fore, HEBAlt can suppress B cell or myeloid potential in a context-specific manner, which suggests a role for this factor in

maintaining T lineage priming prior to commitment. The Journal of Immunology, 2010, 185: 4109–4117.

E-box binding transcription factors (E-proteins) are criticalregulators of lymphoid lineage choice (1). Three E-protein–encoding gene loci are present in mammals: E2A (Tcf3/

ITF1/TcfE2A),HEB (Tcf12/Alf-1/HTF4), and E2-2 (Tcf4/ITF2) (2).The E2A locus generates E47 or E12 proteins bymutually exclusivealternative splicing of two tandemly duplicated DNA-binding do-mains, whereas the HEB and E2-2 gene loci encode only one ba-sic helix-loop-helix (bHLH) domain each. E-proteins function asobligate dimers, and they can form homodimers or heterodimerswith each other, as well as with tissue-specific bHLH factors, suchas MyoD (3). Although it is known that E-protein homodimersand heterodimers are the primary bHLH complexes that act duringlymphocyte development, their specific roles in hematopoietic de-velopment are not well understood.HEB and E2-2 are more closely related to each other than either

is to E2A at the amino acid level and in terms of their genomic loci

structures (2). HEB and E2-2 can each give rise to two distinctproteins: a canonical form (HEBCan or E2-2Can) and an alter-native (Alt) form (HEBAlt or E2-2Alt) (4–6). The Alt forms ofHEB and E2-2 are generated by alternative transcriptional initia-tion sites and alternative splicing. However, the E2A gene locusdoes not contain an Alt domain. The 23-aa Alt domain, which isencoded in one exon between exons 8 and 9, is highly conservedwithin vertebrates, but it has not been found outside the verte-brates (6). Given the importance of E-proteins in cellular growthand development (7–9), it is remarkable that little is known aboutthe functions of the Alt forms of HEB and E2-2.All of the canonical E-proteins are important for lymphoid de-

velopment (1). E2A-deficient mice exhibit a complete block inB cell development prior to the B220+ pre–pro-B cell stage, andthey have aberrant T cell development (10–14). E2A is required forthe upregulation of key regulatory factors, such as EBF, Pax5, andIL-7Ra, and it is needed for the expression of proteins required fordifferentiation and mature B cell function, such as Rag-1, Rag-2,mb-1, l5, and Igk (15, 16). By contrast, HEB and E2-2 have beenthought to contribute to the total dosage of E-proteins requiredfor B cell development, because the absence of either one of thesefactors led to a decrease in the numbers of pro-B cells but no ob-vious developmental block (17). Indeed, a human HEB cDNAinserted into the E2A locus was able to partially rescue B cell de-velopment in E2A2/2 mice (18). However, the cDNA used in thisstudy was HEBCan, leaving open the question of whether HEBAltinfluences early B cell development.E2A and HEB factors are also important during T cell de-

velopment. E2A2/2 and HEB2/2 mice, which lack HEBAlt andHEBCan, exhibit partial blocks at the earliest stage of T cell de-velopment and at the pro-T to pre-T transition (19). Delta-like (DL)-Notch signaling is also critical for T cell development from theearliest stages (20–24), and it may help to direct E-protein activitytoward T lineage genes and away from B lineage genes. Moreover,the low basal levels of MAPK activity in T cell precursors protectE2A from DL-Notch–induced degradation (25, 26). HEB also col-laborates with E2A to control IL-7–mediated pro-T cell expansion,thus linking cell-cycle control to differentiation (27, 28). Interestingly,

*Division of Molecular and Cellular Biology, Sunnybrook Research Institute;†Department of Immunology, University of Toronto, Toronto, Ontario, Canada; and‡Department for Experimental Hematopoesis, Linkoping University, Linkoping,Sweden

1Current address: University of British Columbia Centre for Blood Research, Cana-dian Blood Services, Vancouver, British Columbia, Canada.

2Current address: Sanofi-Pasteur, Toronto, Ontario, Canada.

Received for publication June 4, 2009. Accepted for publication July 28, 2010.

This work was supported by the Canadian Institute of Health Research (GrantsMOP82861 and NIP79923), the Leukemia Research Fund, the Sunnybrook ResearchInstitute (to M.K.A.) and a Ontario Graduate Scholarship (to M.B.).

Address correspondence and reprint requests to Dr. Michele K. Anderson, Sunny-brook Research Institute, 2075 Bayview Avenue, Room A340, Toronto, Ontario M4N3M5, Canada. E-mail address: manderso@sri.utoronto.ca

The online version of this article contains supplemental material.

Abbreviations used in this paper: Alt, alternative; bHLH, basic helix-loop-helix; CLP,common lymphoid progenitor; DL, Delta-like; ETP, early T cell progenitor; IRES,internal ribosome entry site; MIGR1, murine stem cell virus-internal ribosome entrysite-GFP vector backbone; MIY, murine stem cell virus-internal ribosome entry site-yellow fluorescent protein; MPP, multipotent progenitor; MSCV, murine stem cellvirus; PI, propidium iodide; Q-PCR, quantitative real-time RT-PCR; SCF, stem cellfactor; YFP, yellow fluorescent protein.

Copyright� 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0901783

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

the third E-protein, E2-2, is not expressed at high levels in de-veloping B cells or T cells; instead, it seems to be a critical de-terminant of plasmacytoid dendritic cell development (29).By contrast, myeloid development depends on the inhibition of

E-protein activity, in part by an increase in Id factors (30, 31), whichsequester E-proteins in inactive heterodimers. Id-2 is induced byPU.1, and high-level expression of PU.1 promotes myeloid de-velopment, whereas low levels of PU.1, enforced by Ikaros-induced Gfi1 expression, are necessary for B cell development(32). PU.1 is also necessary for the earliest stages of T cell de-velopment (33, 34), but it must be downregulated in pro-T cells topermit T cell development (35–37). Interestingly, forced expressionof Id factors alone in precursors placed in the fetal thymic envi-ronment diverts cells to the NK lineage, rather than to the myeloidlineage, indicating that context is critical for the developmentaloutcome imposed by E-proteins and Id factors (38, 39).We first identified HEBAlt while screening an arrayed cDNA

library constructed from SCID thymocytes, which represent a ge-netically enriched source of pro-T cells (40). Additional work inour laboratory confirmed that it was expressed in pro-T cells, andit could promote the entry of uncommitted precursors into theT cell lineage (6). Because pro-T cells retain some myeloid po-tential (41), we evaluated the ability of HEBAlt and HEBCan toinfluence B cell development and myeloid development in thepresence and absence of DL-Notch signaling. We found thatHEBAlt inhibits B cell development in a DL-Notch–independentmanner, but it requires DL-Notch signaling to inhibit myeloiddevelopment. Thus, our results suggest that HEBAlt collaborateswith DL-Notch signals to promote early T cell development bysuppressing alternative lineage potentials.

Materials and MethodsAnimals

The mice used for the RT-PCR studies were CD1 (E14.5 fetal liver) andC57BL/6 mice (bone marrow). For OP9 cocultures, embryos were obtainedfrom National Institutes of Health Swiss timed matings (National CancerInstitute, Frederick, MD), CD1 timed matings (Charles River Laboratories,Wilmington, MA), or from C57BL/6 wild-type or C57BL/6 HEB+/2 3C57BL/6 HEB+/2 timed matings (Sunnybrook Research Institute animalfacility). The studies described were reviewed and approved by insti-tutional review committees at Sunnybrook Research Institute.

Isolation of hematopoietic progenitors from fetal liver andbone marrow

Fetal liver E14.5 lineage-negative (Lin2) cells were obtained by stain-ing cells with biotin-conjugated anti-Gr1, anti-Ter119, anti-CD19, andanti-F4/80 Abs, followed by binding to streptavidin-conjugated microbe-ads (Miltenyi Biotec, Auburn, CA). MACS was performed to obtain thelineage-negative fractions. Bone marrow subsets were obtained by de-pleting total bone marrow of cells expressing Gr1, Ter119, or CD11b byMACS, followed by sorting for populations representing early stages ofB cell development, as previously described (42).

Gene-expression analysis

Quantitative real-time RT-PCR (Q-PCR) was performed on sorted cell pop-ulations by extraction of RNA using TRIzol reagent (Invitrogen, Carlsbad,CA), followed by generation of first-strand cDNA using Superscript RT III(Invitrogen). Aliquots of cDNA, corresponding to ∼1000 cell equivalents,were used as templates in Q-PCR reactions using the SYBR Green MasterMix kit (Applied Biosystems, Foster City, CA) and specific primers (2.5 pmeach primer per 25-ml reaction). Primer sequences were reported previously.Reactions were run and analyzed using the Applied Biosystems SequenceDetection System 7000.

Retroviral transduction of fetal liver precursors

Lin2 E14.5 fetal liver cells were aliquoted at 2 million cells per well ofa six-well plate on Millipore (Bedford, MA) insert filters for retroviraltransduction with MIGR1-based constructs, generated as previously de-

scribed (6). Retroviral DNA was prepared using Endo-Free DNA extrac-tion reagents and columns (Qiagen, Valencia, CA) and cotransfected intofNX-Eco packaging cells with the pcEco plasmid, using a standard cal-cium phosphate-transfection method. Supernatant was collected after 24and 48 h and frozen at 280˚C until use. Cells were infected by modifiedspin infection (35) in the presence of polybrene (8 mg/ml) or Lipofect-amine (Invitrogen). For developmental assays, Lin2 cells were purified,transduced, and cultured overnight in OP9 media (24), supplemented with10 ng/ml IL-7, stem cell factor (SCF), and Flt3L (R&D Systems, Min-neapolis, MN). For the limiting-dilution assay, cells were cocultured withstably transduced GP+E.86 packaging cells (43), supplemented with 10 ng/ml IL-7, SCF, and Flt3L overnight. The next day (16–24 h postinfection),the Lin2Sca-1+c-kit+ (LSK) GFP+ (or YFP+) fractions were sorted.

Limiting-dilution assay for B lineage-commitment frequency

E14.5 Lin2 fetal liver cells were MACS enriched and transduced by co-culture with GP+E.86 packaging cells that had been stably transducedwith MIGR1-control or MIGR1-HEBAlt overnight. GFP+ LSK cells weresorted directly into 96-well round-bottom plates containing OP9-GFPstromal cells by a FACSAria (BD Biosciences, San Jose, CA). Threesets of cultures were performed, each consisting of 30 wells at 10 cells/well, 30 wells at 3 cells/well, and 30 wells at 1 cell/well, for a total of 90replicates of each dilution. After 6 d of culture, cells were harvested andanalyzed by flow cytometry using a FACSCalibur for expression of GFP,CD45, and CD19 using anti–CD45-allophycocyanin and anti–CD19-PE.All wells containing CD45+GFP+CD19+ cells were scored as positive.Frequencies of precursors able to give rise to GFP+CD19+ cells werecalculated by the maximum likelihood method applied to the Poissonmodel, using the Poisson9 program, as previously described (44).

OP9-GFP and OP9-DL1 cocultures

Sorted transduced cells were placed on OP9-GFP or OP9-DL1 monolayerssupplemented with 5 ng/ml IL-7, SCF, and Flt3L in OP9 medium (24) at∼3000 sorted cells/well in six-well plates. Cultures were split every 3–5 dand placed on freshly plated subconfluent monolayers. Cultures were an-alyzed by flow cytometry using a FACSCalibur, FACSDiva, FACSAria, orLSR II (BD Biosciences), and FACS data were analyzed using FloJo (TreeStar, Ashland, OR).

Immunoblotting

Fetal liver GFP+ LSK cells that had been transduced with MIGR1-control orMIGR1-HEBAlt were cultured in OP9-GFP cocultures for a total of 22 d.Cells were collected from cultures and subjected to immunoblotting withan anti-HEB Ab (Santa Cruz Biotechnology, Santa Cruz, CA; clone A-20 oranti-HEB Ab kindly provided by D. Littman, Skirball Institute of Bio-molecular Medicine, New York, NY). Blots were stripped and probed withan anti-tubulin Ab as a loading control. The anti-HEB Ab epitope is con-tained within the common portion of HEBAlt and HEBCan; thus, it detectsboth proteins, which are distinguishable by size.

Luciferase reporter assays

Luciferase assays were performed by cotransfection of 100 ng 8X E-boxreporter construct with expression plasmids encoding mouse E47, HEBAlt,and/or HEBCan into HeLa cells using Lipofectin (Invitrogen), as previouslydescribed (45, 46). Expression of E2A, HEBAlt, and HEBCan were drivenby CMV promoters in pCDNA3 and/or pBK-CMV constructs. HEBAlt andHEBCan sequences were identical to those used in the retroviral con-structs. Control plasmid DNA (pCDNA3) was included where necessary tobring each amount of expression plasmid up to 300 ng total. Renillaplasmids encoding pr10 were also added as transfection-efficiency con-trols, and the reporter activity was calculated and presented as relativeunits of luciferase to Renilla activity.

ResultsHEBAlt inhibits the generation of pro-B cells from multipotentprecursors

Our previous work showed that expression of HEBAlt in hema-topoietic precursors promoted their entry into the T cell lineage. Todetermine whether this was due, in part, to inhibition of alternativelineage potential, we retrovirally expressed HEBAlt or HEBCan infetal liver precursors and followed their development into the Blineage in the OP9-GFP coculture system (24, 47) (Fig. 1A, 1B).Although Lin2 fetal liver cells do not normally express HEBAlt,

4110 HEBAlt IN HEMATOPOIESIS

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

transduction with HEBAlt or HEBCan leads to an increase inHEBAlt mRNA levels, as previously shown (6) (SupplementalFig. 1A). Because this cross-regulation complicates analysis ofthe specific roles of HEBAlt versus HEBCan, we initially usedHEB2/2 precursors, which are devoid of HEBAlt and HEBCan(17). The HEB-null allele on the C57BL/6 background is em-bryonic lethal, necessitating timed matings of HEB+/2 mice toobtain HEB2/2 fetal precursors. E14.5 Lin2 fetal liver cells wereobtained by MACS depletion of cells expressing CD19, Gr1,Ter119, or F4/80. Lin2 cells were transduced with MIGR1-control, MIY-HEBAlt, or MIY-HEBCan retroviral constructs andwere cultured overnight to allow expression of GFP or YFP, whichwere used interchangeably. Sorted GFP+ or YFP+ LSK (DAPI2

Lin2Sca1+ckit+) precursors were cultured on OP9-GFP stromal

monolayers supplemented with SCF, IL-7, and Flt3L for 1 wk. Wethen assessed the appearance of CD19+-committed B cell pre-cursors by FACS analysis.HEBAlt-transduced precursors produced very few CD19+-

committed B cell precursors compared with the control-transduced and HEBCan-transduced precursors, which gave riseto nearly 50% CD19+ cells (Fig. 1C, 1D). GFP+ and YFP+ ex-pression was strongly retained in the HEB2/2 precursors (data notshown), and these cells were CD45+, confirming their hemato-poietic identity. Furthermore, most of the cells in the HEBAlt-transduced cultures remained CD45high, whereas the CD19+ cellsin the control and HEBCan-transduced cultures were CD45int,characteristic of B cell precursors (35, 48, 49). These results sug-gested that HEBAlt could inhibit the generation of B cell precur-sors independently of DL-Notch signaling. Six separate experi-ments using wild-type precursors confirmed that HEBAlt robustlyinterfered with early B cell development from LSK cells, even inthe presence of endogenous HEBCan (Fig. 1E). Total HEBAltmRNA levels in HEBAlt-transduced B lineage precursors weregenerally ∼5–10-fold higher than endogenous levels in control-transduced cells at the same stage of B cell development (Sup-plemental Fig. 1B, 1C). Although HEBCan did not inhibit B celldevelopment in the HEB2/2 background, it did impose a partialblock on early B cell development in the wild-type background(Supplemental Fig. 2). Therefore, HEBCan required HEBAlt toinhibit B cell development, whereas HEBAlt possessed this abilityindependently of HEBCan.

HEBAlt allows myeloid development from LSK cells in OP9-GFP coculture

Although HEBAlt expression led to a severe decrease in CD19+

cells, CD192 cells were still present. Myeloid development issupported from fetal liver-derived LSK cells by OP9-GFP cellsduring the first week of culture (50), suggesting that the CD192

cells could belong to the myeloid lineages. Therefore, we ana-lyzed the expression of CD11b on transduced sorted LSK cellsafter 6 d of OP9-GFP culture. All cultures contained somemyeloid cells, but HEBAlt-transduced cultures exhibited a mark-edly increased percentage of CD11b+ cells with high forwardscatter, consistent with a myeloid identity (Fig. 2A). The ratio ofmyeloid cells to B lineage cells was increased in response toHEBAlt even more in HEB2/2 cultures than in wild-type cultures,suggesting a dose-dependent inhibition of B cell development(Fig. 2A, 2B). Moreover, the paucity of CD19+ cells was not likelydue to a simple lack of CD19 expression, because there were noB220+ cells that were not also CD19+ at this time point (Fig. 2B).There was a moderate increase in cell death in HEBAlt-transducedcultures relative to controls (Fig. 2D). However, overall cellnumbers in HEBAlt-transduced cultures did not decrease relativeto control-transduced cultures (Fig. 2C). Therefore, the decreasein B lineage cells did not seem to be primarily a result ofincreased cell death. Interestingly, HEBCan-induced inhibition ofB cell development in wild-type precursors did not result in anincrease in the percentages of myeloid cells, suggesting that themechanisms of inhibition might not be identical.

HEBAlt expression in hematopoietic progenitors results ina reduced frequency of cells that can give rise to CD19+-committed B cell precursors

To better define the nature of the HEBAlt-induced B cell defect, weanalyzed the frequency of precursors able to develop into CD19+

cells in HEBAlt-transduced cultures compared with controls usinga limiting-dilution assay (Fig. 3) (44). Typical levels of over-expression were between 5- and 10-fold in B lineage cells (Sup-

FIGURE 1. HEBAlt inhibits the appearance of CD19+ B cell precursors.

A, Retroviral vector structures used in this study. Arrows indicate tran-

scriptional start sites driven by MSCV viral long terminal repeat pro-

moters. Blue segments are bHLH DNA-binding and dimerization domains.

The pink segment represents the Alt domain, and the gray segment rep-

resents the HEBCan-specific domain. The black segments are putative

transactivation domains. B, Experimental design. Lin2 (Gr-1, Ter119,

CD19, F4/80) cells from HEB2/2 E14.5 fetal liver were transduced with

retroviral constructs, cultured overnight, sorted for GFP+ LSK (Lin2Sca-

1+c-kit+GFP+ or YFP+) populations, and placed in OP9-GFP coculture for

7 d. C, FACS analysis of CD45 and CD19 expression. CD45int cells are

B lineage primed, and CD19+ cells are B lineage committed. D, Total cell

numbers after 7 d of OP9-GFP coculture. E, Scatter plot showing a com-

pilation of the decreases in the percentage of B lineage cells in response to

HEBAlt in six different experiments using wild-type precursors. IRES,

internal ribosome entry site; MIGR1, MSCV-IRES-GFP vector backbone;

MIY, MSCV-IRES-YFP; MSCV, murine stem cell virus; YFP, yellow

fluorescent protein.

The Journal of Immunology 4111

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

plemental Fig. 1B, 1C). If B-primed progenitors overexpressingHEBAlt were simply inhibited in their proliferation potential, thenequal numbers of cells would be expected to become GFP+CD19+

in HEBAlt-transduced and control cultures, with individual wellsproducing lower numbers of cells. However, if commitment or matu-ration of progenitors was indeed impeded, wells containing pre-cursors capable of giving rise to GFP+CD19+ cells (Fig. 3B) wouldoccur at a lower frequency in HEBAlt-expressing cultures comparedwith control-transduced cultures. To distinguish between these pos-sibilities, we transduced LSK cells with HEBAlt-encoding virus orcontrol virus and sorted GFP+ LSK cells into 96-well plates con-taining OP9-GFP stromal cells. The cultures were analyzed 6 d later.We found that although HEBAlt-expressing cells were capable ofproducing GFP+CD19+ cells, they were significantly less likely to doso than control cells (Fig. 3A). These results are consistent with a de-fect in B lineage commitment.

HEBAlt does not perturb the development of committed B cellprecursors

To overcome the severe block in early development and examinethe influence of HEBAlt on committed B cell precursors, we

cultured E14.5 Lin2 fetal liver cells with IL-7, SCF, and Flt3L for9 d to allow initiation of the B lineage gene-expression program.These cells were transduced with MIGR1-control or MIGR1-HEBAlt and analyzed for the expression of B220 and CD43 tofurther characterize the stages at which HEBAlt inhibited de-velopment. At an early time point, there clearly was a lowerpercentage of B220+ cells and a higher percentage of B2202

FIGURE 2. HEBAlt promotes myeloid development

in OP9-GFP coculture. Wild-type or HEB2/2 E14.5

Lin2 fetal liver cells were transduced with MIY-

control or MIY-HEBAlt and cultured overnight. YFP+

LSK cells were sorted, plated on OP9-GFP cells, and

analyzed by FACS after 6 d. A, Expression of the

myeloid marker CD11b. The numbers represent the

percentages of CD11b+ cells. High forward scatter is

indicative of more mature myeloid identity. B, Ex-

pression of B220 and CD19 on CD11b2 cells. C,

Numbers of cells in each culture, presented as the

mean 6 SD. D, Percentages of apoptotic cells mea-

sured by staining with Annexin V and propidium io-

dide (PI), followed by FACS analysis. Values are

shown as mean 6 SD.

FIGURE 3. B lineage commitment frequency analysis of HEBAlt-

transduced precursors by limiting-dilution assay. A, Chart showing the

frequency of precursors capable of giving rise to CD19+ cells within

control or HEBAlt-transduced precursors. E14.5 FL Lin2 cells were

transduced, cultured, and sorted into 96-well plates at 1, 3, and 10 cells/

well onto OP9-GFP stromal cells. After 6 d, the cells were analyzed by

FACS. Wells containing CD45+GFP+CD19+ cells were scored as positive,

and the frequency of precursors with B cell potential was calculated by the

maximum likelihood method applied to the Poisson model. B, Examples of

FACS plots scored as positives.

FIGURE 4. Committed B cell precursors tolerate high levels of HEBAlt

protein expression. E14.5 LSK cells were cultured in IL-7, SCF, and Flt3L

for 9 d. Cells were transduced, placed in OP9-GFP coculture, and analyzed

by FACS after 3 d (A) and 5 d (B). B220+CD43+ cells include pre–pro-B

and pro-B cells, whereas B220+CD432 cells include pre-B cells and im-

mature B cells. C and D, LSK cells transduced with control or HEBAlt

vectors were grown in solution culture for 6 d, transferred into OP9-GFP

cocultures for 22 d, and analyzed by FACS analysis (C) and Western

blotting (D) using anti-HEB Abs and anti-tubulin Abs as a loading control.

Coexpression of B220 and BP-1 are diagnostic for the late pro-B cell stage.

4112 HEBAlt IN HEMATOPOIESIS

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

CD43+ non-B lineage cells in the HEBAlt-transduced cultures,consistent with a very early block in development (Fig. 4A).However, after several additional days of culture, the B2202

CD43+ cells (and CD11b+ cells; data not shown) disappeared incontrol and HEBAlt-transduced cultures, consistent with the short

lifespan of myeloid cells coupled with the absence of appropriategrowth factors (Fig. 4B). In addition to B220 and CD43, thecontrol and HEBAlt-transduced cells that expanded in cultureexpressed BP-1 (Fig. 4C), which, in combination with B220,marks B cell commitment independently of CD19, placing them atthe late pro-B cell stage. To assess whether the recovery ofHEBAlt-transduced cells was due to downregulation or degrada-tion of HEBAlt protein in these cells, we performed an immu-noblot on protein isolated from control and HEBAlt-transducedpro-B cells cultured and expanded for 3 wk. We probed the blotwith an anti-HEB Ab that detects an epitope common to HEBAltand HEBCan and differentiated them based on size. HEBAlt andHEBCan were expressed in control cultures, indicating that theywere present at the protein level in committed B cell precursors,and levels of HEBAlt were elevated in HEBAlt-transduced cells(Fig. 4D, Supplemental Fig. 1C). Therefore, HEBAlt could nolonger perturb B cell development once the cells had committed tothe B cell lineage.

HEBAlt is expressed at low levels in early B cell precursors

Our results predicted that HEBAlt must be kept at low levels inearly B cell precursors to allow commitment to the B cell lineage.Therefore, we examined the mRNA expression of HEBAlt atprogressive stages of B cell development, using a sorting strategythat allowed precise isolation of multipotent precursors (MPPs),common lymphoid precursors (CLPs), and pre–pro-B cells (Sup-plemental Fig. 3A), as well as early pro-B and late pro-B cells (42)(Supplemental Fig. 3B). The mRNA expression levels of HEBAltin these subsets was determined by Q-PCR, and they were com-pared with HEBAlt levels in postnatal thymocytes at the earlyT cell progenitor (ETP), DN2 (early pro-T), DN3 (late pro-T), andDN4 (pre-DP) stages of T cell development (Fig. 5). HEBAltmRNA was nearly undetectable in MPPs, CLPs, and ETPs. Lowlevels of HEBAlt were present in pre–pro-B and early pro-B cells,and higher levels were found in DN2 (early pro-T) and DN3 (late

FIGURE 5. Expression of HEBAlt and HEBCan during lymphopoiesis.

Cell populations were sorted from postnatal bone marrow or thymus,

as specified, and subjected to Q-PCR measurements of HEBAlt and/or

HEBCan mRNA expression. Bone marrow precursors were sorted, as

previously described (42) (Supplemental Fig. 3). MPP, CLP, and pre–pro-B

cells were sorted from Lin2 (Lin1 = Ter119, Gr1, Mac1, CD19) depleted

bone marrow cells. MPP = Lin12AA4.1+CD43intB2202c-kithighIL-7Rlow;

CLP = Lin12AA4.1+CD43intB2202c-kitlowIL-7R+; pre–pro-B = Lin12

AA4.1+CD43intB220+. Early pro-B and late pro-B cells were sorted from

Lin-depleted bone marrow cells using lineage mixture #2 (Lin2 = Ter119,

Gr1, Mac1). Early pro-B = Lin22CD19+B220+CD24intCD43+ckitlow; late

pro-B = Lin22CD19+B220+CD24highCD43+c-kitlow; ETP = CD44+ckit+

CD252; DN2 = CD44+ckit+CD25+; DN3 = CD442ckit2CD25+; DN4 =

CD442ckit2CD252. Values were normalized to b-actin and are shown as

the mean 6 SD of triplicate readings. Data are representative of three

independent experiments. Circles represent developmental intermediates,

and the arrows represent putative precursor–progeny relationships.

FIGURE 6. HEBAlt and HEBCan interfere with E2A activity and Pax5 expression. A, Luciferase assays were performed to assess the activation of an 8X

E-box reporter construct by HEBAlt, HEBCan, E47, or coexpression of E47 with HEBAlt or HEBCan in epithelial HeLa cells. The amount and type of

expression plasmid is indicated on the x-axis, and the y-axis shows the values of luciferase activity relative to Renilla activity, presented as the mean 6 SD.

B, Q-PCR analysis of E2A, HEBCan, and HEBAlt expression in early B cell precursors sorted from bone marrow. Pre–pro-B, early pro-B, and late pro-B

cells were as specified in Fig. 4. Values are normalized to b-actin and presented as the mean 6 SD of triplicate readings. C, GFP+ LSK cells transduced

with MIGR1-control, MIGR1-HEBAlt, or MIGR1-HEBCan were sorted and placed in OP9-GFP coculture for 4 d. The CD45highCD192 cells and the

CD45intCD192 cells were harvested from the GFP+ populations and analyzed by Q-PCR for expression of Pax5 mRNA. D, GFP+ LSK cells were sorted

from MIGR1-control or MIGR1-HEBAlt–transduced precursors and placed in OP9-GFP coculture for 6 d. Pre–pro-B cells (B220+AA4.1+CD192) were

sorted from the GFP+ populations and subjected to Q-PCR to measure Pax5 expression. Values are expressed relative to b-actin and are the mean 6 SD of

triplicate readings.

The Journal of Immunology 4113

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

pro-T) cells, consistent with previous reports (6, 51). However, inlate pro-B cells, HEBAlt expression levels were similar to thoseobserved in pro-T cells. HEBCan was expressed in MPPs andCLPs and at higher levels than HEBAlt in pre–pro-B and earlypro-B cells. However, in late pro-B cells, HEBAlt levels increased,and HEBCan levels decreased. Therefore, HEBAlt expression washigher in early T cell precursors than in early B cell precursors.Intriguingly, pre-B cells expressed higher levels of HEBAlt thanpro-B cells at the mRNA (Supplemental Fig. 4B) and protein(Supplemental Fig. 4C) levels, whereas IgM+ bone marrow cellsexpressed very little. In addition, fetal liver CD19+ (pro-B and pre-B) mRNA levels were similar to those of early T cell precursors(Supplemental Fig. 4A). These observations suggest that HEBAltplays a role during normal B cell development at the late pro-B topre-B cell stages.

HEB factors can inhibit E2A activity

Given the importance of E2A in the initiation of the B lineage gene-expression program, it was possible that HEBAlt and/or HEBCaninhibited early B cell development by disrupting E2A homodimerfunction. Therefore, we tested the ability of HEBCan and HEBAltto inhibit E47 activity using an 8X E-box reporter construct (45).E47 alone activated expression of this construct by ∼90-fold inHeLa cells (Fig. 6A). HEBCan had only a limited ability to acti-vate transcription (3-fold), whereas HEBAlt did not activate thisreporter at all. However, cotransfection of HEBAlt or HEBCanwith E47 decreased reporter expression to ∼10-fold over control,indicating strong inhibition of E47 function by either factor. Theseresults suggested that initiation of B cell development might re-quire a high ratio of E47 to HEBAlt and/or HEBCan. Therefore,we analyzed the relative levels of E2A, HEBAlt, and HEBCanmRNA side-by-side in pre–pro-B, early pro-B, and late pro-B cells;E2A was indeed expressed at much higher levels than HEBCan orHEBAlt during these stages (Fig. 6B). These results confirm thatHEBAlt, HEBCan, and E2A do not play fully redundant roles in E-box activation and show that both HEB factors can negativelymodulate the activity of E47 in this context.

HEBAlt and HEBCan inhibit Pax5 expression in early B cellprecursors

Pax5 is one of the key regulators of B cell lineage commitment andmaintenance of B lineage identity (52). Because E2A activity canbe modulated by HEBAlt, and E2A activity is required for Pax5expression (53), we set out to determine whether Pax5 mRNAlevels could be perturbed in response to HEBAlt or HEBCanduring B cell specification. GFP+ LSK cells were sorted fromtransduced Lin2 progenitors and cultured for 4 d on OP9-GFPstromal cells, after which GFP+ cells from the CD45+ and CD45int

fractions were sorted. Q-PCR analysis revealed that HEBAlt andHEBCan decreased the levels of Pax5 mRNA compared withcontrols (Fig. 6C). This occurred in the CD45+ fractions, whichexpressed low levels of Pax5, as well as in the CD45int fractions,which expressed higher levels of Pax5, consistent with specificationtoward the B cell lineage. To confirm that HEBAlt decreased ex-pression of Pax5 specifically in B cell precursors, we next culturedtransduced LSK cells for 6 d onOP9-GFP stroma and sorted out pre-pro–B cells (B220+AA4.1+CD192). These cells also exhibitedlower levels of Pax5mRNA in the HEBAlt-transduced cultures thanin the controls (Fig. 6D). Therefore, HEB factors can inhibit Pax5mRNA expression in the context of early B cell precursors, possiblyby interfering with E47 activity.

DL-Notch signaling and HEBAlt collaborate to promote T celldevelopment over myeloid development

Although HEBAlt can inhibit E47 activity, it does not antagonizeT cell development, indicating that it does not act as a context-independent inhibitor of all E-protein activity. Because B celldevelopment is efficiently blocked by DL-Notch signaling, even inthe presence of Id factors (38), we turned to an analysis of myeloiddevelopment in the presence of DL-Notch signaling. E14.5 Lin2

fetal liver precursors were transduced with control, HEBAlt, orHEBCan-expressing retroviral constructs and cultured overnightto allow GFP expression. Sorted GFP+ LSK populations wereplaced on OP9-DL1 cells and analyzed after 4 d of coculture. Asexpected, the percentages and numbers of CD25+Thy-1+ (DN2/3;pro-T) cells were elevated over controls in the HEBAlt-transducedcultures but not in the HEBCan-transduced cultures (Fig. 7A, 7C).B cell progenitors did not arise in these cultures, as expected, butCD11b+ myeloid cells did. Under these conditions, unlike in OP9-GFP cocultures, fewer myeloid cells arose from uncommittedprecursors in HEBAlt-transduced cultures (Fig. 7B, SupplementalFig. 5). At this early time point in OP9-DL1 coculture, ∼60% ofthe control and HEBCan-transduced cells expressed CD11b com-pared with only 20% of the cells in the HEBAlt-transduced cul-ture. Cell numbers were not appreciably different (Fig. 7C).Therefore, HEBAlt promoted T cell development at the expense ofmyeloid development in the presence of DL-Notch signaling.Importantly, HEBCan did not inhibit the appearance of CD11b+

cells (Fig. 7B, Supplemental Fig. 5), suggesting that DL-Notchsignals collaborate uniquely with HEBAlt to enhance T cell

FIGURE 7. Increases in T cell precursors are accompanied by decreases

in myeloid cells in HEBAlt-transduced OP9-DL1 cocultures. E14.5 Lin2

fetal liver cells were transduced with retroviral constructs, cultured over-

night, and then sorted to obtain GFP+ LSK cells, which were placed in

OP9-DL1 cocultures. FACS analysis was performed 4 d later. A, Analysis

of Thy-1 and CD25 expression. Thy-1+CD25+ cells are pro-T cells. B,

Analysis of CD19 and CD11b expression. CD11b+ cells are myeloid. C,

Total cell numbers at day 4 of LSK OP9-DL1 coculture are shown for three

independent experiments.

4114 HEBAlt IN HEMATOPOIESIS

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

development over myeloid development and that this function isnot solely mediated by suppression of E47 activity.

DiscussionOur studies show that HEBAlt is able to repress alternative thymiclineage potential, as well as promote entry into the T cell lineage,and show that these functions are highly context dependent ina manner contingent on Notch signaling. HEBAlt expression inhematopoietic progenitors inhibited the appearance of CD19+ pro-B cells and allowed early progenitors to develop into myeloid cells,but only whenDL-Notch ligandswere not available. In the presenceof DL-Notch signals, HEBAlt inhibited myeloid development andpromoted T cell development. HEBAlt and HEBCan were able toinhibit E47 activity and Pax5 expression, providing a partial mech-anistic explanation for their ability to inhibit B cell development.However, HEBCan did not inhibit myeloid development or en-hance T cell development, suggesting that this activity was in-dependent of E47 inhibition and was specific to HEBAlt. Takentogether, our data indicate that the Alt domain is a powerful mod-ulator of E-protein activity and that HEBAlt may play an import-ant role in early hematopoietic lineage choice. Moreover, it is likelythat the Alt forms of HEB and E2-2 can impact the functions oftissue-specific class B bHLH proteins, such as MyoD, NeuroD, anddHAND, and such interactions might be particularly important inprocesses that also depend on Notch signaling, such as neural crestdevelopment.Our studies showed that several levels of context are critical for

determining the output of HEBAlt activity. Environmentally, highlevels of DL must be available for HEBAlt to inhibit myeloiddevelopment. It is likely that these signals change the context ofthe cell such that additional protein partners for HEBAlt becomeavailable. Signaling may also result in posttranslational mod-ifications of HEBAlt itself that would allow interaction with dif-ferent proteins. The importance of cellular context is also apparentin the differential ability of HEBAlt to influence B cell precursors atdifferent stages of development. Our frequency assay results areconsistent with a defect in B lineage commitment or in the earlymaturation of precursors expressing HEBAlt, suggesting that theprotein can perturb the genetic networks that regulate B lineage cellfate. In contrast, we repeatedly observed that direct transduction ofcommitted B cell precursors with HEBAlt did not decrease Pax5expression or perturb cell-surface expression of CD19, even wheninducible forms were used (data not shown). Coupled with theincrease in HEBAlt mRNA during normal B cell development,these results suggest that HEBAlt is tolerated, and is likely tofunction normally during B cell development, in capacities thathave yet to be determined.Like HEBAlt, Id factors can disrupt E47 function and suppress

B cell potential, whereas only Id factors enhance myeloid potential(38, 54–56). However, unlike Id factors, HEBAlt possesses a basicDNA-binding domain; we previously showed that HEBAlt is ca-pable of specific binding to an E-box from the pre-Ta–regulatoryregion (6), and it can restore pre-TamRNA expression to HEB2/2

precursors (57). Furthermore, although we found that HEBAlt andHEBCan are capable of inhibiting E47 activity, reminiscent of Idfactors, other studies clearly showed that HEBCan and E47 areable to activate specific genes, indicating that the context of theregulatory DNA of each gene is critical in the correct interpretationof specific E-protein dimers (58, 59).Moreover, E2A2/2 thymocytesdisplay a partial block at the DN1 to DN2 transition, whereasHEBAlt promotes this transition, indicating that HEBAlt does notact simply as an inhibitor of E2A in all cellular contexts. Indeed,E47 and HEBCan are able to substitute for each other in at leastsome contexts (60), consistent with earlier reports in whichHEBCan

was able to partially compensate for the loss of E2A in B cell de-velopment (18). Additional studies are needed to determine thecontexts in which E47, HEBAlt, and HEBCan work antagonisticallyor cooperatively with each other to drive the expression of lineage-specific genes.Because neither HEBAlt nor HEBCan is expressed at high levels

in uncommitted progenitors, they are unlikely to be involved inmediating the initial T/B fate choice during normal hematopoiesis,although they may act to inhibit B lineage regulators at the pro-T cell stage. ETPs, which express very low levels of HEBAlt, havealready lost most of their B cell potential, indicating that HEBAlt isnot strictly necessary for this function. However, some myeloidpotential is retained in fetal andadultDN2/3 thymocytes (61,62), andour results are consistent with a role for HEBAlt in the suppressionof myeloid development in developing T cell precursors. Candi-date targets of HEBAlt activity include PU.1 and C/EBPa, which canwork together to divert T cell precursors into the myeloid lineages(63–65). Preliminary studies in our laboratory also revealed latentmyeloid potential in HEB2/2 T cell precursors, and they showed thatthe addition of HEBAlt to myeloid progenitors could restore theirability to become T cells (M. Braunstein and M.K. Anderson, man-uscript in preparation). Therefore, our results agreewith other studiesthat support a close relationship between the T cell and myeloid lin-eages (49, 50, 61, 62, 66) and suggest that DL-Notch signals maysuppress myeloid development, in part, by upregulation of HEBAlt.Therefore, we propose that HEBAlt operates as a T lineage

fidelity factor during early T cell development in a DL-Notch–dependent manner. PU.1, GATA3, and Notch1 are all legacy genesfrom the stem cell stage, whereas HEBAlt is upregulated as T cellspecification occurs. This function is reminiscent of Pax5, whichmust be present to maintain B lineage fidelity in B cell precursorsand mature B cells (52, 67). Removal of Pax5 from B lineage cellsat any stage, including in mature peripheral B cells, leads tode-differentiation back to a primitive state, which is then able tobe induced to develop into alternative lineages. By contrast,HEBAlt is not expressed past b-selection checkpoint during T celldevelopment (6) and, therefore, it is not required to maintainlineage fidelity after the fundamental T lineage regulatory networkhas been established. Likewise, HEBAlt is not expressed at highlevels prior to DN2/3 pro-T stages of T cell development, sug-gesting that it is unlikely to be involved in prethymic restriction ofthe B cell lineage (68–70). As one of the few transcription factorsupregulated at the DN1 to DN2 transition (71), HEBAlt is uniquelypositioned to collaborate with DL-Notch signals in a feed-forwardloop that maintains T lineage identity, in part, by suppressing mye-loid lineage potential prior to T cell commitment. Indeed, we foundthat HEB2/2 precursors in OP9-DL1 cocultures undergo a pro-gressive loss of T lineage identity as they transit through the DN2and DN3 stages (M. Braunstein and M.K. Anderson, submitted forpublication). Future work will focus on defining the network con-nections between HEBAlt and other hematopoietic lineage regu-lators and on understanding the mechanism by which the Alt do-main confers unique functions to E-proteins.

AcknowledgmentsWe thank M. Ratcliffe and J. Rast for critical comments and helpful dis-

cussions. We also thank the Comparative Research Facility at Sunnybrook

Research Institute for excellent animal care. G. Knowles and A. Khandani

provided important sorting expertise. HEB+/2 mice were kindly provided by

Trang Hoang, and we thank J.C. Zuniga-Pflucker for the OP9-GFP and OP9-

DL1 stromal cell lines. We thank D. Littman for the anti-HEB Ab. We also

thank Matthew Chui and Gianna Vaccarelli for help with sorting and Q-PCR.

DisclosuresThe authors have no financial conflicts of interest.

The Journal of Immunology 4115

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

References1. Kee, B. L. 2009. E and ID proteins branch out. Nat. Rev. Immunol. 9: 175–184.2. Hu, J. S., E. N. Olson, and R. E. Kingston. 1992. HEB, a helix-loop-helix protein

related to E2A and ITF2 that can modulate the DNA-binding ability of myogenicregulatory factors. Mol. Cell. Biol. 12: 1031–1042.

3. Murre, C., P. S. McCaw, and D. Baltimore. 1989. A new DNA binding anddimerization motif in immunoglobulin enhancer binding, daughterless, MyoD,and myc proteins. Cell 56: 777–783.

4. Skerjanc, I. S., J. Truong, P. Filion, and M. W. McBurney. 1996. A splice variantof the ITF-2 transcript encodes a transcription factor that inhibits MyoD activity.J. Biol. Chem. 271: 3555–3561.

5. Anderson, M. K., and E. V. Rothenberg. 2000. Transcription factor expressionin lymphocyte development: clues to the evolutionary origins of lymphoid celllineages? Curr. Top. Microbiol. Immunol. 248: 137–155.

6. Wang, D., C. L. Claus, G. Vaccarelli, M. Braunstein, T. M. Schmitt, J. C. Zuniga-Pflucker, E. V. Rothenberg, and M. K. Anderson. 2006. The basic helix-loop-helix transcription factor HEBAlt is expressed in pro-T cells and enhances thegeneration of T cell precursors. J. Immunol. 177: 109–119.

7. Pagliuca, A., P. Gallo, P. De Luca, and L. Lania. 2000. Class A helix-loop-helixproteins are positive regulators of several cyclin-dependent kinase inhibitors’promoter activity and negatively affect cell growth. Cancer Res. 60: 1376–1382.

8. Prabhu, S., A. Ignatova, S. T. Park, and X. H. Sun. 1997. Regulation of theexpression of cyclin-dependent kinase inhibitor p21 by E2A and Id proteins.Mol. Cell. Biol. 17: 5888–5896.

9. Yang, Q., L. Kardava, A. St Leger, K. Martincic, B. Varnum-Finney,I. D. Bernstein, C. Milcarek, and L. Borghesi. 2008. E47 controls the de-velopmental integrity and cell cycle quiescence of multipotential hematopoieticprogenitors. J. Immunol. 181: 5885–5894.

10. Bain, G., E. C. Maandag, D. J. Izon, D. Amsen, A. M. Kruisbeek,B. C. Weintraub, I. Krop, M. S. Schlissel, A. J. Feeney, M. van Roon, et al. 1994.E2A proteins are required for proper B cell development and initiation of im-munoglobulin gene rearrangements. Cell 79: 885–892.

11. Zhuang, Y., P. Soriano, and H. Weintraub. 1994. The helix-loop-helix gene E2Ais required for B cell formation. Cell 79: 875–884.

12. Borghesi, L., J. Aites, S. Nelson, P. Lefterov, P. James, and R. Gerstein. 2005.E47 is required for V(D)J recombinase activity in common lymphoid progeni-tors. J. Exp. Med. 202: 1669–1677.

13. Dias, S., R. Mansson, S. Gurbuxani, M. Sigvardsson, and B. L. Kee. 2008. E2Aproteins promote development of lymphoid-primed multipotent progenitors.Immunity 29: 217–227.

14. Bain, G., I. Engel, E. C. Robanus Maandag, H. P. te Riele, J. R. Voland,L. L. Sharp, J. Chun, B. Huey, D. Pinkel, and C. Murre. 1997. E2A deficiencyleads to abnormalities in alphabeta T-cell development and to rapid developmentof T-cell lymphomas. Mol. Cell. Biol. 17: 4782–4791.

15. Kee, B. L., and C. Murre. 1998. Induction of early B cell factor (EBF) andmultiple B lineage genes by the basic helix-loop-helix transcription factor E12.J. Exp. Med. 188: 699–713.

16. Nutt, S. L., and B. L. Kee. 2007. The transcriptional regulation of B cell lineagecommitment. Immunity 26: 715–725.

17. Zhuang, Y., P. Cheng, and H. Weintraub. 1996. B-lymphocyte development isregulated by the combined dosage of three basic helix-loop-helix genes, E2A,E2-2, and HEB. Mol. Cell. Biol. 16: 2898–2905.

18. Zhuang, Y., R. J. Barndt, L. Pan, R. Kelley, and M. Dai. 1998. Functional re-placement of the mouse E2A gene with a human HEB cDNA. Mol. Cell. Biol.18: 3340–3349.

19. Barndt, R., M. F. Dai, and Y. Zhuang. 1999. A novel role for HEB downstream orparallel to the pre-TCR signaling pathway during alpha beta thymopoiesis. J.Immunol. 163: 3331–3343.

20. Radtke, F., A. Wilson, G. Stark, M. Bauer, J. van Meerwijk, H. R. MacDonald,and M. Aguet. 1999. Deficient T cell fate specification in mice with an inducedinactivation of Notch1. Immunity 10: 547–558.

21. Wilson, A., H. R. MacDonald, and F. Radtke. 2001. Notch 1-deficient commonlymphoid precursors adopt a B cell fate in the thymus. J. Exp. Med. 194: 1003–1012.

22. Feyerabend, T. B., G. Terszowski, A. Tietz, C. Blum, H. Luche, A. Gossler,N. W. Gale, F. Radtke, H. J. Fehling, and H. R. Rodewald. 2009. Deletion ofNotch1 converts pro-T cells to dendritic cells and promotes thymic B cellsby cell-extrinsic and cell-intrinsic mechanisms. Immunity 30: 67–79.

23. Pui, J. C., D. Allman, L. Xu, S. DeRocco, F. G. Karnell, S. Bakkour, J. Y. Lee,T. Kadesch, R. R. Hardy, J. C. Aster, and W. S. Pear. 1999. Notch1 expression inearly lymphopoiesis influences B versus T lineage determination. Immunity 11:299–308.

24. Schmitt, T. M., and J. C. Zuniga-Pflucker. 2002. Induction of T cell developmentfrom hematopoietic progenitor cells by delta-like-1 in vitro. Immunity 17: 749–756.

25. Nie, L., M. Xu, A. Vladimirova, and X. H. Sun. 2003. Notch-induced E2Aubiquitination and degradation are controlled by MAP kinase activities. EMBOJ. 22: 5780–5792.

26. Nie, L., S. S. Perry, Y. Zhao, J. Huang, P. W. Kincade, M. A. Farrar, andX. H. Sun. 2008. Regulation of lymphocyte development by cell-type-specificinterpretation of Notch signals. Mol. Cell. Biol. 28: 2078–2090.

27. Wojciechowski, J., A. Lai, M. Kondo, and Y. Zhuang. 2007. E2A and HEB arerequired to block thymocyte proliferation prior to pre-TCR expression. J.Immunol. 178: 5717–5726.

28. Anderson, M. K. 2006. At the crossroads: diverse roles of early thymocytetranscriptional regulators. Immunol. Rev. 209: 191–211.

29. Cisse, B., M. L. Caton, M. Lehner, T. Maeda, S. Scheu, R. Locksley,D. Holmberg, C. Zweier, N. S. den Hollander, S. G. Kant, et al. 2008. Tran-scription factor E2-2 is an essential and specific regulator of plasmacytoiddendritic cell development. Cell 135: 37–48.

30. Ishiguro, A., K. S. Spirin, M. Shiohara, A. Tobler, A. F. Gombart, M. A. Israel,J. D. Norton, and H. P. Koeffler. 1996. Id2 expression increases with differen-tiation of human myeloid cells. Blood 87: 5225–5231.

31. Cooper, C. L., and P. E. Newburger. 1998. Differential expression of Id genes inmultipotent myeloid progenitor cells: Id-1 is induced by early-and late-actingcytokines while Id-2 is selectively induced by cytokines that drive terminalgranulocytic differentiation. J. Cell. Biochem. 71: 277–285.

32. Spooner, C. J., J. X. Cheng, E. Pujadas, P. Laslo, and H. Singh. 2009. A recurrentnetwork involving the transcription factors PU.1 and Gfi1 orchestrates innate andadaptive immune cell fates. Immunity 31: 576–586.

33. Spain, L. M., A. Guerriero, S. Kunjibettu, and E. W. Scott. 1999. T cell de-velopment in PU.1-deficient mice. J. Immunol. 163: 2681–2687.

34. Franco, C. B., D. D. Scripture-Adams, I. Proekt, T. Taghon, A. H. Weiss,M. A. Yui, S. L. Adams, R. A. Diamond, and E. V. Rothenberg. 2006. Notch/Delta signaling constrains reengineering of pro-T cells by PU.1. Proc. Natl.Acad. Sci. USA 103: 11993–11998.

35. Anderson, M. K., A. H. Weiss, G. Hernandez-Hoyos, C. J. Dionne, andE. V. Rothenberg. 2002. Constitutive expression of PU.1 in fetal hematopoieticprogenitors blocks T cell development at the pro-T cell stage. Immunity 16: 285–296.

36. Georgescu, C., W. J. Longabaugh, D. D. Scripture-Adams, E. S. David-Fung,M. A. Yui, M. A. Zarnegar, H. Bolouri, and E. V. Rothenberg. 2008. A generegulatory network armature for T lymphocyte specification. Proc. Natl. Acad.Sci. USA 105: 20100–20105.

37. Rothenberg, E. V., and D. D. Scripture-Adams. 2008. Competition and collab-oration: GATA-3, PU.1, and Notch signaling in early T-cell fate determination.Semin. Immunol. 20: 236–246.

38. Heemskerk, M. H., B. Blom, G. Nolan, A. P. Stegmann, A. Q. Bakker, K. Weijer,P. C. Res, and H. Spits. 1997. Inhibition of T cell and promotion of naturalkiller cell development by the dominant negative helix loop helix factor Id3. J.Exp. Med. 186: 1597–1602.

39. Fujimoto, S., T. Ikawa, T. Kina, and Y. Yokota. 2007. Forced expression of Id2 infetal thymic T cell progenitors allows some of their progeny to adopt NK cellfate. Int. Immunol. 19: 1175–1182.

40. Anderson, M. K., G. Hernandez-Hoyos, R. A. Diamond, and E. V. Rothenberg.1999. Precise developmental regulation of Ets family transcription factors duringspecification and commitment to the T cell lineage. Development 126: 3131–3148.

41. Chi, A. W., J. J. Bell, D. A. Zlotoff, and A. Bhandoola. 2009. Untangling theT branch of the hematopoiesis tree. Curr. Opin. Immunol. 21: 121–126.

42. Rumfelt, L. L., Y. Zhou, B. M. Rowley, S. A. Shinton, and R. R. Hardy. 2006.Lineage specification and plasticity in CD192 early B cell precursors. J. Exp.Med. 203: 675–687.

43. Ciofani, M., T. M. Schmitt, A. Ciofani, A. M. Michie, N. Cuburu, A. Aublin,J. L. Maryanski, and J. C. Zuniga-Pflucker. 2004. Obligatory role for cooperativesignaling by pre-TCR and Notch during thymocyte differentiation. J. Immunol.172: 5230–5239.

44. de St. Groth, F. 1982. The evaluation of limiting dilution assays. J. Immunol.Methods 49: R11–R23.

45. Sigvardsson, M. 2000. Overlapping expression of early B-cell factor and basichelix-loop-helix proteins as a mechanism to dictate B-lineage-specific activity ofthe lambda5 promoter. Mol. Cell. Biol. 20: 3640–3654.

46. Smith, E. M., P. Akerblad, T. Kadesch, H. Axelson, and M. Sigvardsson. 2005.Inhibition of EBF function by active Notch signaling reveals a novel regulatorypathway in early B-cell development. Blood 106: 1995–2001.

47. Nakano, T., H. Kodama, and T. Honjo. 1994. Generation of lymphohemato-poietic cells from embryonic stem cells in culture. Science 265: 1098–1101.

48. Kirberg, J., and T. Brocker. 1996. CD45 up-regulation during lymphocytematuration. Int. Immunol. 8: 1743–1749.

49. Kawamoto, H., K. Ohmura, and Y. Katsura. 1997. Direct evidence for thecommitment of hematopoietic stem cells to T, B and myeloid lineages in murinefetal liver. Int. Immunol. 9: 1011–1019.

50. de Pooter, R. F., T. M. Schmitt, J. L. de la Pompa, Y. Fujiwara, S. H. Orkin, andJ. C. Zuniga-Pflucker. 2006. Notch signaling requires GATA-2 to inhibit mye-lopoiesis from embryonic stem cells and primary hemopoietic progenitors. J.Immunol. 176: 5267–5275.

51. David-Fung, E. S., R. Butler, G. Buzi, M. A. Yui, R. A. Diamond,M. K. Anderson, L. Rowen, and E. V. Rothenberg. 2009. Transcription factorexpression dynamics of early T-lymphocyte specification and commitment. Dev.Biol. 325: 444–467.

52. Cobaleda, C., A. Schebesta, A. Delogu, and M. Busslinger. 2007. Pax5: theguardian of B cell identity and function. Nat. Immunol. 8: 463–470.

53. Kwon, K., C. Hutter, Q. Sun, I. Bilic, C. Cobaleda, S. Malin, and M. Busslinger.2008. Instructive role of the transcription factor E2A in early B lymphopoiesisand germinal center B cell development. Immunity 28: 751–762.

54. Jaleco, A. C., A. P. Stegmann, M. H. Heemskerk, F. Couwenberg, A. Q. Bakker,K. Weijer, and H. Spits. 1999. Genetic modification of human B-cell de-velopment: B-cell development is inhibited by the dominant negative helix loophelix factor Id3. Blood 94: 2637–2646.

55. Thal, M. A., T. L. Carvalho, T. He, H. G. Kim, H. Gao, J. Hagman, and C. A. Klug.2009. Ebf1-mediated down-regulation of Id2 and Id3 is essential for specificationof the B cell lineage. Proc. Natl. Acad. Sci. USA. 2009: 106: 552–557.

4116 HEBAlt IN HEMATOPOIESIS

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

56. Cochrane, S. W., Y. Zhao, R. S. Welner, and X. H. Sun. 2009. Balance betweenId and E proteins regulates myeloid-versus-lymphoid lineage decisions. Blood113: 1016–1026.

57. Braunstein, M., and M. K. Anderson. 2010. Developmental progression of fetalHEB-/- precursors to the pre-T cell stage is restored by HEBAlt. Eur. J. Immunol.In press.

58. Tremblay, M., S. Herblot, E. Lecuyer, and T. Hoang. 2003. Regulation of pTalpha gene expression by a dosage of E2A, HEB, and SCL. J. Biol. Chem. 278:12680–12687.

59. Ghosh, J. K., W. J. Romanow, and C. Murre. 2001. Induction of a diverse T cellreceptor gamma/delta repertoire by the helix-loop-helix proteins E2A and HEBin nonlymphoid cells. J. Exp. Med. 193: 769–776.

60. Barndt, R. J., M. Dai, and Y. Zhuang. 2000. Functions of E2A-HEB heterodimersin T-cell development revealed by a dominant negative mutation of HEB. Mol.Cell. Biol. 20: 6677–6685.

61. Bell, J. J., and A. Bhandoola. 2008. The earliest thymic progenitors for T cellspossess myeloid lineage potential. Nature 452: 764–767.

62. Wada, H., K. Masuda, R. Satoh, K. Kakugawa, T. Ikawa, Y. Katsura, andH. Kawamoto. 2008. Adult T-cell progenitors retain myeloid potential. Nature452: 768–772.

63. Laiosa, C. V., M. Stadtfeld, H. Xie, L. de Andres-Aguayo, and T. Graf. 2006.Reprogramming of committed T cell progenitors to macrophages and dendriticcells by C/EBP alpha and PU.1 transcription factors. Immunity 25: 731–744.

64. Xie, H., M. Ye, R. Feng, and T. Graf. 2004. Stepwise reprogramming of B cellsinto macrophages. Cell 117: 663–676.

65. Heavey, B., C. Charalambous, C. Cobaleda, and M. Busslinger. 2003. Myeloidlineage switch of Pax5 mutant but not wild-type B cell progenitors by C/EBPalpha and GATA factors. EMBO J. 22: 3887–3897.

66. Ng, S. Y., T. Yoshida, J. Zhang, and K. Georgopoulos. 2009. Genome-widelineage-specific transcriptional networks underscore Ikaros-dependent lym-phoid priming in hematopoietic stem cells. Immunity 30: 493–507.

67. Mikkola, I., B. Heavey, M. Horcher, and M. Busslinger. 2002. Reversion ofB cell commitment upon loss of Pax5 expression. Science 297: 110–113.

68. Allman, D., A. Sambandam, S. Kim, J. P. Miller, A. Pagan, D. Well, A. Meraz,and A. Bhandoola. 2003. Thymopoiesis independent of common lymphoidprogenitors. Nat. Immunol. 4: 168–174.

69. Tan, J. B., I. Visan, J. S. Yuan, and C. J. Guidos. 2005. Requirement for Notch1signals at sequential early stages of intrathymic T cell development. Nat.Immunol. 6: 671–679.

70. Sambandam, A., I. Maillard, V. P. Zediak, L. Xu, R. M. Gerstein, J. C. Aster,W. S. Pear, and A. Bhandoola. 2005. Notch signaling controls the generation anddifferentiation of early T lineage progenitors. Nat. Immunol. 6: 663–670.

71. Tydell, C. C., E. S. David-Fung, J. E. Moore, L. Rowen, T. Taghon, andE. V. Rothenberg. 2007. Molecular dissection of prethymic progenitor entry intothe T lymphocyte developmental pathway. J. Immunol. 179: 421–438.

The Journal of Immunology 4117

by guest on April 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from