**4. Acute lymphoblastic leukemia**

286 Advances in Hematopoietic Stem Cell Research

Once the preBCR is expressed at the end of the proB stage, it can take over many of the functions performed by the IL-7 receptor signaling. Both receptors act individually and together to allow B cell development (Figure 5). Like IL-7R, the preBCR promotes mechanisms of positive selection, survival and proliferation (Ramadani et al., 2010; Yasuda et al., 2008). The CCND3 gene, which encodes for cyclin D3, is essential for PreB cell

Downstream the IL-7 and preBCR receptors, a handful of transcription factors (TF) are critical for commitment to the B cell lineage and early development; these include E2A/TCF3 (immunoglobulin enhancer binding factors E12/E47/transcription factor 3), EBF1 (Early B cell Factor 1) and PAX5 (Paired box 5) (Figure 5). Loss of E2A and EBF1 blocks entry into the B cell lineage, while loss of PAX5 redirects B cells into other lineages (Nutt et al., 1999; O'Riordan & Grosschedl, 1999). Acting together with E2A, EBF1 and STAT5, one of the main molecular functions of PAX5 is to allow VDJ recombination (Hsu et al., 2004). Also, E2A, PAX5, IKZF1 and RUNX1, among other TF, are responsible for RAG expression (Kuo & Schlissel, 2009). Moreover, IL-7R signaling fulfills an essential role in early B cell development, with STAT5 participating in the activation of the B cell regulatory genes E2A, EBF1 and PAX5. E2A encodes two TF via alternative splicing, E12 and E47. In mice lacking the E2A gene, the B cell lineage is lost, there is no heavy chain recombination, and the expression of the B cell-restricted

Enforced expression of EBF1 and PAX5 is sufficient to overcome the developmental block in mice deficient in E2A, IL-7 or IL-7R, further illustrating the transcriptional hierarchy of the B cell-specific program triggered by IL-7 receptor signaling (Nutt & Kee, 2007). EBF1 acting together with PAX5 drives the expression of many genes critical for early B cell development and B cell function, including FOXO1, MYCN, LEF1, BLNK, CD79A (MB-1),

Although PAX5 is a positive regulator of B-cell specific genes, also functions as a repressor of non B-lineage genes such as M-CSFR, NOTCH1 and FLT3 (Cobaleda et al., 2007) so B cell

Also important for lymphoid development are members of the Ikaros family of TFs, mainly IKZF1 (which encodes Ikaros) and IKZF3 (which encodes Aiolos). Ikaros activates B cell genes and represses genes that are unrelated to the B lineage. Expression of IKZF1 and IKZF3 is regulated by alternative splicing, which produces long isoforms (Ik-1, Ik-2, Ik-3, Aio-1, Aio-3, Aio-4 and Aio-6) that efficiently bind to DNA, and short isoforms (Ik-4, Ik-5/7, Ik-6, Ik8, Aio-2, Aio-5) that are unable to bind DNA with high affinity and do not activate transcription (Liippo et al., 2001). Ikaros is activated in early stages of lymphopoiesis and is required for both early and late events in lymphocyte differentiation. Aiolos is not required during the early specification of the B and T lineages but is essential during further B cell maturation. They also act in concert to promote preB cell cycle exit and transition to small

Selection processes operating on developing B cells are similar in all mammals. Thus, early B cell development in humans is also mainly guided by VDJ recombination and by the

expansion and integrates IL-7R and preBCR signals (Cooper et al., 2006).

genes EBF1, PAX5, CD79A/B and VPREB1 (CD179A) is also affected.

PreB stage (Ma et al., 2010).

**3.4 Human B cell development** 

RAG2, CD19 and CR2 (CD21) (Nutt & Kee, 2007; Smith & Sigvardsson, 2004).

development is unidirectional and mostly irreversible in homeostatic conditions.

Acute lymphoblastic leukemia (ALL) is a disorder characterized by the monoclonal and/or oligoclonal proliferation of hematopoietic precursor cells of the lymphoid series within the bone marrow. At present, ALL is the most frequent malignancy in children worldwide and a serious problem of public health, constituting 25% of all childhood cancers and 75%-85% of the cases of childhood leukemias (Perez-Saldivar et al., 2011). Near to 80% of ALL cases have precursor B-cell immunophenotype, while approximately 15% show T-cell immunophenotype. Even when a relatively high efficiency of therapeutic agents has been demonstrated (Pieters & Carroll, 2010), there has been a slight but gradual increase in the incidence of ALL in the past 25 years, and appears to be highest in Hispanic population, which also show superior rates of high risk patients (Fajardo-Gutiérrez et al., 2007; Abdullaev et al., 2000; Perez-Saldivar et al., 2011; Mejía-Aranguré et al., 2011). Factors such as drug resistance, minimal residual disease, cell lineage switch, and the rise of mixed lineages often put the success of treatment at risk and change the prognosis of the illness. The molecular mechanism involved in these phenomena and the identities of the target hematopoietic populations have not been completely defined, due in part, to the fact that neither the precise origin of the disease, nor the susceptibility of primitive leukemic cells to extrinsic factors, is known.

#### **4.1 The origin of ALL**

Over the last two decades, cancer stem cells (CSC) have been defined as cells within a tumor that possess the capacity to self-renew and to cause heterogeneous lineages of cancer cells that comprise the tumor (Clarke et al., 2006). According to MF Greaves, who proposed the original hypothesis for leukemogenesis, multiple consecutive carcinogenic hits in hematopoietic cells may drive the malignant transformation (Greaves, 1993; Greaves & Wiernels, 2003), where the second oncogenic event on pre-leukemic clones could be indirectly promoted by delayed infections (Greaves, 2006; Mejía-Aranguré et al., 2011). Our general current view suggests the occurring of oncogenic lesions in early development or in

From HSC to B-Lymphoid Cells in Normal and Malignant Hematopoiesis 289

leukemia initiating cells have undifferentiated characteristics (Espinoza-Hernandez et al., 2001; Cobaleda et al., 2000; Cox et al., 2004; Cox et al., 2009). To characterize ALL progenitor cells, Blair and colleagues have purified by flow cytometry a number of cell fractions based on the expression of CD34 and the B-lymphoid marker CD19. Regardless the risk stratum of

disease in NOD/SCID models (Cox et al., 2004). Moreover, CD133+CD38-CD19- primitive cells residing in ALL BM are suggested to be the leukemia-initiating cells and responsible of drug-resistant residual disease (Cox et al., 2009). However, recent studies have remarkably shown that precursor blasts can also reestablish leukemic phenotypes *in vivo*, conferring them stem cell properties (Heidenreich & Vormoor, 2009; Bomken et al., 2010). Using novel intrafemoral xenotransplantation strategies, Vormoor's Lab has found that all differentiation

successfully engraft and recapitulate the original patient's disease in long-term systems, suggesting that committed cells in ALL do not lose the self-renewal stem cell property while they mature (le Viseur et al., 2008) (Figure 5), though their multi-lineage potential is

These discordant results unveil that key questions regarding leukemic stem cells and the earliest steps of the lymphoid program in ALL still to be solved. Recently, the combination of clonal studies and alterations on genetic copies along with xenotransplant models, showed unsuspected genetic diversity, supporting multiclonal evolution of leukemogenesis rather than lineal succession (Dick, 2008). Thus, a less rigid structure of CSC models should further take account of functional plasticity and clonal evolution to understand CSC biology

and to develop novel, stem/progenitor cell-directed therapies (Bomken et al., 2010).

**4.2 Genes, cytogenetic alterations and transcription factors in B-cell leukemogenesis**  The leukemogenic program is characterized by arrest of differentiation pathways, increased cell proliferation, enhanced self-renewal, decreased apoptosis rates and telomere maintenance. It is thought that together these alterations result in production of highly proliferative clones of immature leukemic blast cells with intrinsic survival advantage and

Gain or loss of function of transcription factors such as E2A, EBF1, PAX5 and Ikaros affect homeostatic B cell lymphopoiesis in murine models, and are often associated with malignant transformation in humans, supporting conserved roles for these TFs and their

A high frequency of ALL patients has genetic lesions -mostly chromosomal translocationsassociated with leukemic cells. E2A is often translocated with several partners, including PBX1 [t(1;19)(q23;p13)] and HLF [t(17;19)(q22;p13)], which are detected in 5-6% and 1% of ALL children, respectively. E2A-PBX1 is a potent transcriptional activator of the WNT16 oncogene (McWhirter et al., 1999), while E2A-HLF functions as a survival factor of early B cells by activating expression of the anti-apoptotic genes SNAI2 (SLUG) and LMO2. Accordingly, gene silencing of LMO2 in an E2A-HLFpos cell line induced apoptotic cell death (Hirose et al., 2010). RUNX1 is also a frequent target for chromosomal rearrangements and mutations in ALL. 25% of children and 2% of adults of ALL patients carry the ETV6/RUNX1 fusion as a result of the translocation t(12;21)(p12;q21), which may play a role

stages of B precursor cells within CD34+CD19+ and CD34-

limitless replicative potential (Warner et al., 2004).

activating signaling pathways (Figure 5) (Pérez-Vera et al., 2011).

cells, but not committed B precursors, were able to reconstitute the

CD19+ fractions are able to

the patient, CD34+CD19-

uncertain.

a primitive cell that result in the abnormal differentiation of leukemic stem cells. Among the various factors that hit the HSC fraction, anomalous microenvironmental cues may contribute to trigger and support the leukemic behaviour of precursor cells (Figure 6).

Although CSC in myeloid leukemias have been strictly depicted as the responsible cells for tumour maintenance, which clearly keep the biological hierarchy within the hematopoietic structure (Dick, 2008), identification of a rare primitive and malignant cell with intrinsic stem cell properties and the ability to recapitulate the acute lymphoblastic leukemia has been more complicated (Bomken et al.,2010), particularly due to the genetic diversity of the disease and the lack of appropriate *in vitro* and *in vivo* models.

Fig. 6. Leukemic stem cell model. Normal hematopoietic stem cells (HSC) give rise to progenitors and mature blood cells within a hierarchical structure in the bone marrow. As a result of multiple and consecutive oncogenic hits on HSC including genetic and microenvironmental alterations, a malignant counterpart (the leukemic stem cell, LSC) emerge, which maintains some degree of developmental potential, generating the leukemic progenitor and blast cells.

Cell culture systems revealing alterations in early hematopoiesis, the existence of leukemic clones with unrelated DJ rearrangements and cytogenetic abnormalities on cells lacking lineage markers, have strongly suggested the participation of primitive cells in ALL. Moreover, data showing cells with immature phenotypes capable of engrafting and reconstituting leukemia in immunodeficient mice, lead to believe that, as in AML & CML, the hierarchy structure of the hematopoietic system is kept in ALL, and infant B cell-

a primitive cell that result in the abnormal differentiation of leukemic stem cells. Among the various factors that hit the HSC fraction, anomalous microenvironmental cues may contribute to trigger and support the leukemic behaviour of precursor cells (Figure 6).

Although CSC in myeloid leukemias have been strictly depicted as the responsible cells for tumour maintenance, which clearly keep the biological hierarchy within the hematopoietic structure (Dick, 2008), identification of a rare primitive and malignant cell with intrinsic stem cell properties and the ability to recapitulate the acute lymphoblastic leukemia has been more complicated (Bomken et al.,2010), particularly due to the genetic diversity of the

Fig. 6. Leukemic stem cell model. Normal hematopoietic stem cells (HSC) give rise to progenitors and mature blood cells within a hierarchical structure in the bone marrow. As a

Cell culture systems revealing alterations in early hematopoiesis, the existence of leukemic clones with unrelated DJ rearrangements and cytogenetic abnormalities on cells lacking lineage markers, have strongly suggested the participation of primitive cells in ALL. Moreover, data showing cells with immature phenotypes capable of engrafting and reconstituting leukemia in immunodeficient mice, lead to believe that, as in AML & CML, the hierarchy structure of the hematopoietic system is kept in ALL, and infant B cell-

result of multiple and consecutive oncogenic hits on HSC including genetic and microenvironmental alterations, a malignant counterpart (the leukemic stem cell, LSC) emerge, which maintains some degree of developmental potential, generating the leukemic

progenitor and blast cells.

disease and the lack of appropriate *in vitro* and *in vivo* models.

leukemia initiating cells have undifferentiated characteristics (Espinoza-Hernandez et al., 2001; Cobaleda et al., 2000; Cox et al., 2004; Cox et al., 2009). To characterize ALL progenitor cells, Blair and colleagues have purified by flow cytometry a number of cell fractions based on the expression of CD34 and the B-lymphoid marker CD19. Regardless the risk stratum of the patient, CD34+CD19- cells, but not committed B precursors, were able to reconstitute the disease in NOD/SCID models (Cox et al., 2004). Moreover, CD133+CD38-CD19- primitive cells residing in ALL BM are suggested to be the leukemia-initiating cells and responsible of drug-resistant residual disease (Cox et al., 2009). However, recent studies have remarkably shown that precursor blasts can also reestablish leukemic phenotypes *in vivo*, conferring them stem cell properties (Heidenreich & Vormoor, 2009; Bomken et al., 2010). Using novel intrafemoral xenotransplantation strategies, Vormoor's Lab has found that all differentiation stages of B precursor cells within CD34+CD19+ and CD34- CD19+ fractions are able to successfully engraft and recapitulate the original patient's disease in long-term systems, suggesting that committed cells in ALL do not lose the self-renewal stem cell property while they mature (le Viseur et al., 2008) (Figure 5), though their multi-lineage potential is uncertain.

These discordant results unveil that key questions regarding leukemic stem cells and the earliest steps of the lymphoid program in ALL still to be solved. Recently, the combination of clonal studies and alterations on genetic copies along with xenotransplant models, showed unsuspected genetic diversity, supporting multiclonal evolution of leukemogenesis rather than lineal succession (Dick, 2008). Thus, a less rigid structure of CSC models should further take account of functional plasticity and clonal evolution to understand CSC biology and to develop novel, stem/progenitor cell-directed therapies (Bomken et al., 2010).

#### **4.2 Genes, cytogenetic alterations and transcription factors in B-cell leukemogenesis**

The leukemogenic program is characterized by arrest of differentiation pathways, increased cell proliferation, enhanced self-renewal, decreased apoptosis rates and telomere maintenance. It is thought that together these alterations result in production of highly proliferative clones of immature leukemic blast cells with intrinsic survival advantage and limitless replicative potential (Warner et al., 2004).

Gain or loss of function of transcription factors such as E2A, EBF1, PAX5 and Ikaros affect homeostatic B cell lymphopoiesis in murine models, and are often associated with malignant transformation in humans, supporting conserved roles for these TFs and their activating signaling pathways (Figure 5) (Pérez-Vera et al., 2011).

A high frequency of ALL patients has genetic lesions -mostly chromosomal translocationsassociated with leukemic cells. E2A is often translocated with several partners, including PBX1 [t(1;19)(q23;p13)] and HLF [t(17;19)(q22;p13)], which are detected in 5-6% and 1% of ALL children, respectively. E2A-PBX1 is a potent transcriptional activator of the WNT16 oncogene (McWhirter et al., 1999), while E2A-HLF functions as a survival factor of early B cells by activating expression of the anti-apoptotic genes SNAI2 (SLUG) and LMO2. Accordingly, gene silencing of LMO2 in an E2A-HLFpos cell line induced apoptotic cell death (Hirose et al., 2010). RUNX1 is also a frequent target for chromosomal rearrangements and mutations in ALL. 25% of children and 2% of adults of ALL patients carry the ETV6/RUNX1 fusion as a result of the translocation t(12;21)(p12;q21), which may play a role

From HSC to B-Lymphoid Cells in Normal and Malignant Hematopoiesis 291

cells (mesenchymal cells, osteoblasts, fibroblasts, adipocytes, endothelial cells, etc) and their products (extracellular matrix molecules, cytokines and chemokines) that support hematopoiesis. Under physiological conditions, the appropriate production of mature blood cells throughout life is sustained by special niches that provide stem and progenitor cells with regulatory signals essential for their maintenance, proliferation and differentiation (Nagasawa et al., 2011). Among secreted factors, CXCL12, FLT3-L, interleukin 7 and stem cell factor are critical for commitment to the lymphoid program and normal B cell development is supported by two stage specific cellular niches within central bone marrow: a CXCL12/SDF1 expressing niche, and a IL-7 expressing niche. B cell precursors are thought to move from one to another as differentiation progresses (Tokoyoda et al., 2004; Nagasawa, 2006). The role of the bone marrow microenvironment in carcinogenesis has been conceived through three possible mechanisms: competition of tumor cells for normal HSC niches, which may allow their maintenance and survival; manipulation of the environment to promote tumor progression and disruption of hematopoietic-niche communication that drives oncogenesis (Raaijmakers, 2011). Although these potential mechanisms are tempting, their contribution to ALL remains formally unexplored. It has been proposed by Sipkins and colleagues that the leukemic cells derived-tumor microenvironment impairs the behavior of normal hematopoietic cells (Colmone et al., 2008). Furthermore, a number of alterations have been recorded in the marrow microenvironment of ALL, including chromosomal aberrations in mesenchymal stem cells, anomalous expression of adhesion molecules, abnormal levels of CXCR4 and growth factors, as well as prevalence of pro-inflammatory cytokines (Menendez et al., 2009; Geijtenbeek et al., 1999; Juarez et al., 2009; and our unpublished results). Whether an abnormal microenvironment anticipates the leukemic stage or is a consequent fact, is still an open issue.

Much has been learned about identity, function and intercommunication of seminal cells within the hematopoietic system from animal models. However, our understanding of the hierarchy and regulation of human stem/progenitor cells is still incomplete and the hematopoietic charts have been in constant re-construction over the last few years. Furthermore, while it has long been recognized that intrinsic abnormalities in primitive hematopoietic cells may cause hematological disorders, it has also become clear that changes in both cell composition and function of the bone marrow microenvironment might govern stem cell activity and lead to disease. Future progress in these areas will be decisive to

R.P. is recipient of funding from the National Council of Science and Technology, CONACYT (grant CB-2010-CO1-152695) and from the Mexican Institute for Social Security,

Abdullaev, F., Rivera-Luna, R., Roitenburd-Belacortu, V., & Espinosa-Aguirre, J. (2000).

Pattern of childhood cancer mortality in Mexico. *Arch Med Res.* Vol. 31, No.5,

IMSS (grant FIS/IMSS/852). EDA and EV are scholarship holders from CONACYT.

(September-October 2000), pp. (526-31), ISSN 0188-4409

suggest novel classification, prognosis and treatment venues.

**5. Conclusion** 

**6. Acknowledgment** 

**7. References** 

regulating the B lineage-specific transcriptional program at an early stage (Durst & Hiebert, 2004). SNP array analysis of ETV6-RUNX1 samples has recently identified multiple additional genetic alterations, but the role of these lesions in leukemogenesis remains undetermined (van der Weyden et al., 2011).

Genome-wide analysis has recorded abnormalities in PAX5 and EBF1 in up to 32% of children and 30% of adults with B ALL, and in 35% of relapsed cases (Mullighan et al., 2007). Currently, five PAX5 fusions have been identified with the gene partners LOC392027 (7p12.1), SLCO1B3 (12p12), ASXL1 (20q11.1), KIF3B (20q11.21) and C20orf112 (20q11.1), with the resulting chimeric proteins expressing lower levels of PAX5 and its target genes (An et al., 2008). EBF1 alterations are common in patients with poor outcomes and are particularly frequent (25%) in relapsed children (Harvey et al., 2010).

The MLL (mixed lineage leukemia) gene is often rearranged in leukemias with myeloid and lymphoid phenotype, probably indicating a very early multipotent progenitor origin. More than 50 fusions involving MLL have been documented. Among them, the MLL-AF4 [t(4;11)(q21;q23)] translocation is present in 80% of infant, 2% of children, and 5-10% of adult ALL (McCarthy, 2010).

The BCR-ABL1 translocation [t(9;22)(q34;q11), also known as Philadelphia chromosome] is found in 5% of pediatric and 25% of adult B cell ALL. An important consequence for this translocation is the over-expression of STAT5. STAT5 inactivation results in cell cycle arrest and apoptosis of BCR-ABLpos malignant B cells and BCR–ABL1pos STAT5 knockout mice do not develop leukemia (Malin et al., 2010). Interestingly, genome-wide analysis of B cell ALL has identified mutations in the STAT5 upstream regulators JAK1 and JAK2 in up to 10% of patients, and patients BCR-ABLpos or with JAK1&2 mutations have a similar gene expression profile and prognosis (Malin et al., 2010). JAK2 mutations lead over-expression of CRLF2 (also known as thymic stromal lymphopoietin receptor) which forms a heterodimeric complex with the IL-7R (Harvey et al., 2010). In a subset of cases, CRLF2 promotes constitutive dimerization and cytokine-independent proliferation. Finally, high expression levels of the short Ikaros isoforms, particularly the dominant negative Ik-6, are also associated with high risk leukemia (Sun et al., 1999). Most of the BCR-ABLpos B ALL patients have deletions in IKZF1 and increased levels of the short isoforms; however, Ik-6 has also been found to be elevated in BCR-ABLneg patients (Mullighan et al., 2008). It has been proposed that the high level of Ikaros short isoform expression is due to genetic lesions. Supporting this idea, IKZF1 somatic deletions have been found in a number of recurrences and are strongly associated with minimal residual disease (Mullighan et al., 2009). A summary of homeostatic and leukemic expression of transcription factors along the B cell pathway is shown in Figure 5.

Despite these important advances in the definition of genetic abnormalities that are prevalent in ALL, the disease is heterogeneous at the molecular level, and possibly it is the result of combination of genetic and epigenetic alterations. Furthermore, high frequencies of ALL cases seem not to be associated to intrinsic genetic abnormalities, opening the possibility of microenvironmental cues leading to disease.

#### **4.3 Leukemic microenvironmental cues?**

The complexity of leukemogenesis increases when we consider the indubitable influence of the bone marrow microenvironment in the hematopoietic development, which is a network of cells (mesenchymal cells, osteoblasts, fibroblasts, adipocytes, endothelial cells, etc) and their products (extracellular matrix molecules, cytokines and chemokines) that support hematopoiesis. Under physiological conditions, the appropriate production of mature blood cells throughout life is sustained by special niches that provide stem and progenitor cells with regulatory signals essential for their maintenance, proliferation and differentiation (Nagasawa et al., 2011). Among secreted factors, CXCL12, FLT3-L, interleukin 7 and stem cell factor are critical for commitment to the lymphoid program and normal B cell development is supported by two stage specific cellular niches within central bone marrow: a CXCL12/SDF1 expressing niche, and a IL-7 expressing niche. B cell precursors are thought to move from one to another as differentiation progresses (Tokoyoda et al., 2004; Nagasawa, 2006). The role of the bone marrow microenvironment in carcinogenesis has been conceived through three possible mechanisms: competition of tumor cells for normal HSC niches, which may allow their maintenance and survival; manipulation of the environment to promote tumor progression and disruption of hematopoietic-niche communication that drives oncogenesis (Raaijmakers, 2011). Although these potential mechanisms are tempting, their contribution to ALL remains formally unexplored. It has been proposed by Sipkins and colleagues that the leukemic cells derived-tumor microenvironment impairs the behavior of normal hematopoietic cells (Colmone et al., 2008). Furthermore, a number of alterations have been recorded in the marrow microenvironment of ALL, including chromosomal aberrations in mesenchymal stem cells, anomalous expression of adhesion molecules, abnormal levels of CXCR4 and growth factors, as well as prevalence of pro-inflammatory cytokines (Menendez et al., 2009; Geijtenbeek et al., 1999; Juarez et al., 2009; and our unpublished results). Whether an abnormal microenvironment anticipates the leukemic stage or is a consequent fact, is still an open issue.
