Notch (Notch1, Notch2, Notch3, Notch4)
Notch mutants in Drosophila melanogaster were originally reported in 1919 (Morgan and Bridges 1919). Notch was named based on the Drosophila melanogaster mutants that exhibited irregular notches of missing tissue at the tips of their wing blades. This phenotype was caused by heterozygous loss-of-function mutations in a gene subsequently named “Notch” that was cloned in 1983 (Artavanis-Tsakonas et al. 1983).
Notch and Notch Signaling
The Notch receptor is synthesized as a 300-kDa precursor and is cleaved in the trans-Golgi compartment. The resulting extracellular and lumenal N-terminal fragment and the transmembrane domain and intracellular domain C-terminal fragment are assembled into the mature heterodimer receptor through a noncovalent linkage. The extracellular and lumenal portion of Notch undergoes extensive N- and O-linked glycosylation during synthesis and secretion, which is critical for proper folding of the receptor and its subsequent interactions with ligands (Fortini 2009).
Canonical Notch signaling is activated by binding of a ligand from the DSL family, that includes Delta and the Serrate/Jagged subfamily that is located on the adjacent cell’s surface. Ligand binding activates proteolysis of Notch between gly1743 and val1744 (termed site 3 or S3) by γ-secretase. One fragment is the Notch intracellular domain (NICD) that is translocated to the nucleus. NICD functions in transcriptional regulation of the Hairy/E(spl) family (Hes genes in mammals, her genes in zebrafish), that codes for inhibitory basic helix-loop-helix (bHLH) transcriptional regulators that control many different secondary targets, including Notch ligand genes and the Hes/her genes themselves. The other fragment is the Notch extracellular domain (NECD) that is endocytosed by the ligand-expressing cell.
Lateral inhibition is a feedback loop in which adjacent and developmentally equivalent cells assume completely different fates. Binding of Delta and Notch across the two cells in the opposite direction generates transcriptional feedback. Delta expression by the signal-sending cell is stronger than by the signal-receiving cell, therefore, Notch signaling by the signal-receiving cell is activated. This signaling results in transcription of Hes/her genes, thereby producing an achaete-scute complex (AS-C) to block expression of differentiation genes and Delta transcription. Decreased Delta expression by the signal-receiving cell diminishes the Delta-Notch binding from the signal-receiving to signal-sending cell, thus inactivating Notch signaling by the signal-sending cell. As a result, the Hes/her genes are not transcribed by the signal-sending cells, enhancing differentiation and upregulating Delta expression. The final outcome of this feedback loop is that the signal-receiving cell maintains strong expression of the Notch receptor Notch expression and Notch signaling through Hes/her-dependent transcriptional feedback, while the signal-sending cell maintains strong ligand expression and repressed Notch signaling through AS-C-dependent transcriptional feedback (Lewis et al. 2009).
Physiological Roles of Notch Signaling
Notch and Development and Differentiation
The Notch pathway, with TGF-β, Wnt, and Hedgehog, is a representative pathway that regulates developmental and differentiation gene expression programs, so that at the correct time and place, cells with the same propensity for a particular cell fate can give rise to daughters that exhibit differences in morphology and protein expression.
A variety of strategies are used to achieve developmental goals. One strategy is to form gradients of signaling proteins either on cell surfaces or in extracellular spaces that help determine cell fates when they activate cellular receptors. These signaling proteins are known as morphogens. Another strategy is to utilize hierarchical sequences of gene expression so that over time different progeny will become different types of cells. Notch signaling is involved with these strategies to achieve proper cell development and differentiation (Beckerman 2005).
Notch signaling coordinates and synchronizes the cell clocks of individual cells when somites form. Segmentation and antero-posterior polarity of somites are formed under the control of Notch signaling and its associated pathways.
Somites are musculoskeletal segments, derived from presomitic mesoderm (PSM). Wnt and FGF are produced at the tail end of the PSM and spread to the anterior portion, generating a morphogen gradient. Anteriorly, somites form in response to lowered Wnt and FGF levels, and at the tail PSM grows caudally, expanding the embryo. This growth is a rhythmic process that results in segmentation, and originates from cell-intrinsic oscillation. The oscillation is paced by autoregulation of Hes/her genes, which results in a negative feedback loop. Delays in transcription and translation within the feedback loop determine the period of oscillation.
Blocking of Notch signaling disrupts segmentation. However, Notch signaling does not produce oscillation itself, but rather functions in the coordination. Even Notch-dissociated cells show oscillating expression, though in a less regular pattern (Lewis et al. 2009).
Mesp2 is a central mediator of Notch signaling in the somite mesoderm, generating antero-posterior polarity in the presumptive somite via a complex signaling network involving Dll1- and Dll3-Notch signaling and the Notch regulator presenilin 1 (Kuan et al. 2004).
Notch signaling controls neural cell fates, both the early-fate decisions between neural and epidermal and the late-fate decisions between different subtypes of neural cells (Cau and Blader 2009). Proneural clusters are developed in neural cells or epidermal cells. Following several iterations of the lateral inhibition feedback loop as cluster communication, only one cell of the cluster downregulates the Notch pathway and becomes a neural precursor. The remaining cells are blocked from reaching the neural fate and, are either reselected during a second wave of neurogenesis or secondarily adopt an epidermal fate.
Notch is also required during the specification between two different neural subtypes. These binary decisions can either involve sister cells, as during the formation of Drosophila sense organs, or cells that are not linearly related, such as the R3 and R4 photoreceptors of the Drosophila eye. Once specified as a neural progenitor, the sensory organ precursor (SOP or pI) divides to generate two cells, pIIa and pIIb, which communicate via Notch. Subsequent divisions generate the four cells of the sensory organs, as well as a glial cell that undergoes apoptosis.
Canonical Notch signaling is essential for the generation of definitive embryonic hematopoietic stem cells, but is dispensable for their maintenance during adult life (Sandy and Maillard 2009). Notch controls several early steps of T-cell development, as well as specific cell fate and differentiation decisions in other hematopoietic lineages. In addition, emerging evidence indicates that Notch is a potent, context-specific regulator of T-cell immune responses, which includes several disease models relevant to patients.
Notch signaling intensity varies throughout T-cell development. The Notch target genes Deltex1 and Hes1 are expressed at very low levels in bone marrow hematopoietic stem cells (HSCs). Upon arrival in the thymic environment, early T progenitors (ETPs) strongly upregulate the expression of Notch target genes. Expression levels of Notch target genes gradually increase during development from the ETP stage to the double negative 3a (DN3a) stage. After the β-selection checkpoint, during which Notch signaling is significantly downregulated, the intensity of Notch signaling steadily decreases from the DN3b stage to the CD4+/CD8+ double positive (DP) stage. Thymic single positive (SP) CD4+ and CD8+ T cells, as well as naive peripheral T cells, express low amounts of Notch target genes. Upon T-cell activation in the periphery, Notch signaling increases sharply in a context-dependent manner.
Notch and Cancer
The highly conserved Notch signaling pathway plays pleiotropic roles during embryonic development and is important for the regulation of self-renewing tissues (Koch and Radtke 2007). The physiological functions of this signaling cascade range from stem cell maintenance and influencing cell fate decisions of slightly differentiated progenitor cells, to the induction of terminal differentiation processes, all of which are recapitulated in different forms of cancers. Although Notch signaling is mainly associated with oncogenic and growth-promoting roles, depending on the tissue type, it can also function as a tumor suppressor.
The first data describing the oncogenic consequences of aberrant Notch signaling in solid tumors were derived from animal studies characterizing a frequent insertion site, named int3, of the mouse mammary tumor virus (MMTV). The int3 site was later identified as the Notch4 locus. MMTV insertions have also been found in the Notch1 locus, albeit with a lower frequency. Together, these result on aberrant Notch signaling and mouse mammary tumorigenesis lead to the question of how significant aberrant Notch signaling is for human breast cancer. To date, only correlative evidence for the involvement of Notch signaling in human breast cancer is available.
Medulloblastoma has primarily been associated with aberrant sonic hedgehog signaling (Shh), which induces N- MYC expression. The primitive nature of medulloblastoma tumor cells and the fact that Notch signaling is involved in the maintenance of neural stem and progenitor cells motivated several groups to investigate the potential role of Notch in medulloblastoma. Expression studies using primary medulloblastoma tumor samples showed increased mRNA expression of NOTCH2, but not of NOTCH1. Increased expression of the target gene Hes1 correlated with a poor patient survival prognosis. Blocking of Notch signaling resulted in increased apoptosis and a reduction of viable cell in tumor cells.
While the causative role of aberrant Wnt signaling for the development of colorectal cancer is well established, it is currently less clear whether Notch signaling might have a similar oncogenic function within the gut. Since gene expression profiles of crypt cells and colorectal cancer cell lines appear to be very similar, colorectal cancer cells may represent the transformed counter part of crypt cells. Since Notch is a gate keeper of crypt cells, it is likely that Notch and Wnt signaling occur simultaneously in adenomas and crypt cells. Indeed, expression of the Notch target gene Hes1 has been observed in adenomas of APCMin mice, as well as in primary human colorectal tumors.
Notch signaling has been shown to play an important role during embryonic pancreas development by maintaining an undifferentiated precursor cell type. Notch receptors, ligands, and downstream targets, such as Hes1, were found to be upregulated in preneoplastic lesions, as well as in invasive pancreatic cancers in humans and mice. This suggests that Notch signaling in pancreatic cancers might be an early event leading to the accumulation of undifferentiated precursor cells.
Global gene expression profiling and immunohistochemistry have revealed the expression of multiple Notch receptors and ligands in primary lesions of human malignant melanomas, therefore, expanding the list of possible pathways involved in melanoma development. Subsequent studies using established melanoma cell lines showed that pharmacological blocking of Notch signaling can have growth suppressive effects. The oncogenic function of Notch signaling within these cell lines was linked to increased β-catenin-mediated signaling, as well as increased MAPK and AKT signaling.
Historically, human NOTCH was identified at the chromosomal breakpoint of a subset of T cell lymphoblastic leukemias/lymphomas containing a t(7;9)(q34;q34.3) chromosomal translocation. The translocation fuses the 3′ portion of NOTCH1 to the T-cell receptor Jb locus. This translocation results in a truncated NOTCH1 protein (N1ICD) that is constitutively active and aberrantly expressed. However, this seminal discovery did not reveal the full oncogenic potential of the truncated version of N1ICD. Less than 1% of all human T-cell leukemias or lymphomas contain this translocation. However, more importantly, aberrant Notch signaling was subsequently found in several human leukemias and lymphomas that lacked genomic rearrangements, signifying that upregulated Notch signaling might have a common role in human leukemogenesis.
Definitive proof for a central role of NOTCH1 in human T-cell acute lymphoblastic leukemia (T-ALL) cell lines came from a recent study that identified somatic activating mutations in the NOTCH1 receptor independent of the t(7;9) translocation. These mutations were detected in more than 50% of human T-ALL cases. Additionally, these mutations were found in all previously defined T-ALL subtypes.
Notch signaling was constitutively active in human clear cell renal cell carcinoma (CCRCC) cell lines. Blocking Notch signaling attenuated proliferation and restrained anchorage-independent growth of CCRCC cell lines and inhibited growth of xenotransplanted CCRCC cells in nude mice. Notch1 knockdown was accompanied by elevated levels of the negative cell-cycle regulators p21(Cip1) and/or p27(Kip1). Moreover, Notch1 and the Notch ligand Jagged1 were expressed at significantly higher levels in CCRCC tumors than in normal human renal tissues. Growth of primary CCRCC cells was attenuated upon inhibition of Notch signaling.
In adults, blood vessels in most organs are quiescent, but the growth of solid tumors requires specific embryonic signaling pathways to direct new blood vessels to grow around and into the tumor. VEGF is important in this process. Notch signaling has a strong effect on angiogenesis as seen in gain- or loss-of-function studies of Notch1, Notch1/4, Jagged1, Dll1, Dll4, Hey1/Hey2, and Presenilins (PS1 and PS2) (Iso et al. 2003; Li and Harris 2005).
By examining the whole-genome transcriptome profile of a xenograft model of breast cancer and its metastasis form to brain, activation of Notch signaling was found to be crucial in brain metastasis. Over 2,000 genes were differentially expressed in brain metastatic cells, which included various metastasis-related genes and many genes related to angiogenesis, migration, tumorigenesis, and cell-cycle regulation. Interestingly, the Notch signaling pathway was activated in correlation with increased Jag2 mRNA expression, activated NICD, and NICD/CLS promoter-luciferase activity. Increased migration and invasion of brain metastatic cells as compared with primary breast cancer cells were inhibited by inactivation of Notch signaling using DAPT, a γ-secretase inhibitor, and RNAi-mediated knockdown of Jag2 and Notch1 (Nam et al. 2008; Jeon et al. 2008).
Tumor Suppressor Function
Instead of maintaining progenitor cells in an undifferentiated state, or influencing their cell fate decisions, in some tissues such as skin, prostatic epithelium, hepatocellular carcinoma, and small-cell lung cancer, Notch can also induce differentiation that is associated with growth suppression (Koch and Radtke 2007). However, the growth inhibitory role of Notch has mainly (with the exception of the prostate) been based on activated Notch1 overexpression studies. Thus, further experiments are needed in these tissues and cancer types to clarify whether Notch indeed has tumor suppressive functions.
Cancer Stem Cells (CSCs)
The strongest evidence to date for a role of Notch in CSCs is in breast cancer, embryonal brain tumors, and gliomas. γ-secretase inhibitors (GSIs) abolish the formation of secondary mammospheres from a variety of human breast cancer cell lines, as well as in primary patient specimens. In breast ductal carcinoma in situ (DCIS), the ability to form multilineage spheroids termed “mammospheres,” indicators of stem-like cells, is dramatically decreased by GSIs, Notch-4 monoclonal antibodies, or Gefitinib. This finding suggests cooperation between epidermal growth factor receptor (EGFR) and Notch-4 in DCIS “stem cell” maintenance. There is evidence for a feedback loop between Her2/Neu and Notch, which may maintain CSCs in Her2/Neu-overexpressing tumors. Sansone and colleagues showed that in mammospheres from human breast cancers, IL-6 induces Notch-3 signaling, increases expression of Jagged-1, and, through Notch-3, promotes a hypoxia-resistant phenotype. The same group described the p66Shc-Notch-3 pathway as essential for maintaining the hypoxia-resistant phenotype of human breast cancer mammospheres. Fan and colleagues showed that Notch inhibition selectively depletes medulloblastoma CSCs as determined by CD133-high status or dye exclusion. The same group has described very similar findings in glioblastoma CSCs. Importantly, in gliomas, Notch confers radioresistance to CSCs. GSI treatment selectively enhanced radiation-induced death of glioma CSCs, but not bulk glioma cells. This effect was replicated by Notch-1 or Notch-2 knockdown, and was accompanied by AKT inhibition and reduced Mcl-1 expression. Other malignancies are being actively investigated. A role of Notch, STAT3, and TGF-β in hepatocellular carcinoma CSC maintenance has been suggested. In Gemcitabine-resistant pancreatic carcinoma cells, EMT (epithelial-mesenchymal transition) is associated with activation of Notch signaling, potentially linking Notch to the “Weinberg model” of stemness acquisition through EMT and to treatment resistance. Inhibition of Notch signaling through GSIs or Delta-4 monoclonal antibodies decreased the number of CSC and/or their tumorigenicity in some preclinical models (Pannuti et al. 2010).
The Notch signaling pathway in various cancers can be targeted at various levels, including receptor-ligand binding, release of NICD, as well as the coactivator complex. A promising strategy to block receptor-ligand binding employs inhibitory antibodies directed against Jagged1 or DLL4. Blocking DLL4 led to dysfunctional neovascularization and inhibition of tumor growth. The most promising results have been achieved using small-molecule inhibitors of the γ-secretase complex (GSI) that prevented the release of NICD. A phase I clinical trial using the GSI MK0752 inhibitor was initiated in 2005. The third protein component of the Notch signaling complex that may possibly provide a suitable drug target is the coactivator complex consisting of CSL, MAML, and CBP/p300. Small inhibitory peptides acting as dominant negative forms of MAML or CLS decrease the transcriptional activation of target genes (Koch and Radtke 2007).
Even targeting developmental pathways such as Notch, will most likely not give the elusive “magic bullet,” and will require the development of rational drug combinations. Such cocktails will be made possible only through a thorough understanding of cross-talk between Notch and other developmental and nondevelopmental pathways that may play roles in CSCs in specific malignancies. Our knowledge is rapidly evolving, but there is evidence to support some combinations of treatments. The following examples are not meant to be all-inclusive. However, these classes of agents are reasonable candidates for combination with Notch inhibitors: (1) Inhibitors of the PI3-kinase-AKT-mTOR pathway; (2) NF-κB inhibitors; and (3) Her2/Neu inhibitors, platinum compounds, EGFR inhibitors, and Hedgehog inhibitors. In breast cancer, a newly discovered feedback loop between Notch and ERα supports combining Notch inhibitors with antiestrogens. Antiestrogens plus GSI and Hedgehog-inhibitors plus GSI combinations are being investigated in ongoing clinical trials. In the case of the Hedgehog inhibitor-GSI combination, anti-CSC effects are being specifically measured.
Ultimately, the best use of Notch inhibitors and other CSC-targeted agents will be in the context of personalized medicine. To that end, it must be determined: (1) which cancers and specific cancer subtypes contain Notch-dependent CSCs; (2) what role specific components of Notch signaling play in these CSCs; (3) what pathways cross-talk with Notch in specific CSCs; and (4) how Notch activity can be measured in CSCs from individual patients (e.g., in biopsy material).
The design of clinical trials for CSC-targeted agents will have to consider that anti-CSC effects will not necessarily translate into rapid tumor volume changes. Disease-free or recurrence-free survival will be the most informative endpoints. For situations when this would require prohibitively long follow-ups, it will be important to develop accurate surrogate biomarkers that reflect anti-CSC effects. These may include spheroid formation assays, flow cytometry, and molecular tests, but posttreatment tumor tissue will be required in most cases. A question of potentially great interest is whether it is possible to assess CSC numbers or the relative “stemness” of individual tumors by studying CTCs. These cells can be isolated from patient blood by several methods, one of which is US Food and Drug Administration approved. Although these trials may be challenging, the payoff may be novel treatments that eliminate or greatly reduce treatment resistance in a broad range of malignancies (Pannuti et al. 2010).
Notch and Genetic Diseases
Loss of function of Notch pathway components can cause inherited genetic diseases such as Alagille syndrome, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), and spondylocostal dysostosis (SCD) (Fiuza and Arias 2007).
Mutations in the Jagged1 gene are responsible for Alagille syndrome, which is normally diagnosed in the first 2 years of life. This is an autosomal dominant mutation that causes defects in bile duct formation leading to liver problems, and is also responsible for kidney, eye, heart, and skeleton development problems. The great diversity of the disease presentation suggests that other factors may influence the outcome, such as genetic properties of Notch signaling regulators.
Mutations in human Notch 1 and 3 are responsible for CADASIL syndrome. These mutations lead to an autosomal vascular disorder resulting in the loss of the arteriolar vascular smooth muscle cells that are substituted by granular eosinophilic material. One specific feature of CADASIL syndrome is its late onset around the age of 45 years. This disease is linked with a variety of symptoms ranging from migraines and subcortical ischemic strokes to progressive dementia and premature death.
SCD is a family of diseases that results in vertebral defects. Essentially, SCD is caused by mutations in Dll3 resulting in rib defects that lead to abnormalities in vertebral segmentation and trunk size. Understanding the mechanisms of Notch signaling regulation is crucial in the development of therapeutic approaches for the treatment of these diseases.
Notch and Other Diseases
Notch is expressed by neurons in the adult brain where it is present at particularly high levels in the hippocampus. The prospect that Notch is the substrate of γ-secretase/presenilin and plays a role in learning and memory suggests a potential link between Notch signaling and the pathogenesis of Alzheimer’s disease (Woo et al. 2009). In post-mitotic neurons, Notch proteins interact with PSs (presenilins) and with APP (amyloid precursor protein), which have roles in the memory deficits associated with Alzheimer’s disease. In some cases, mutations in the genes encoding APP, PS1, and PS2 are responsible for early-onset Alzheimer’s disease. The phenotype of PS1 deletion mice is similar to that observed in Notch knockout mice. The PS1/PS2 double knockout phenotype is even more similar, suggesting closely related functions for these proteins. Indeed, knockouts of any one of several c-secretase components cause developmental abnormalities that are similar to those caused by Notch 1 and Notch 2 knockouts.
Albuminuria associated with sclerosis of the glomerulus affects millions of people and leads to a progressive decline in renal function. Activation of the Notch pathway, which is critical to glomerular patterning, contributes to the development of glomerular disease. Expression of the intracellular domain of Notch1 (ICN1) was increased in glomerular epithelial cells in diabetic nephropathy and in focal segmental glomerulosclerosis. Conditional in vivo re-expression of ICN1 exclusively in podocytes caused proteinuria and glomerulosclerosis. In vitro and in vivo studies showed that ICN1 induced apoptosis of podocytes through the activation of p53. Genetic deletion of a Notch transcriptional partner (Rbpj) specifically in podocytes or pharmacological inhibition of the Notch pathway (with a c-secretase inhibitor) protected rats with proteinuric kidney diseases (Niranjan et al. 2008).
Connection to Other Signaling Pathways
The Drosophila Disheveled gene, which encodes a component of the Wingless signaling pathway, interacts antagonistically with Notch and one of its ligands, Delta. Notch1 activation induced p21 in differentiating mouse keratinocytes. The induction was associated with the targeting of Rbpjk (RBPSUH) to the p21 promoter. Notch1 also activated p21 through a calcineurin-dependent mechanism acting on the p21 TATA box-proximal region. Notch signaling through the calcineurin/ NFAT pathway also involved calcipressin and Hes1.
Oncogenic Ras activates Notch signaling. Wild-type Notch1 is necessary to maintain the neoplastic phenotype in Ras-transformed human cells in vitro and in vivo. The oncogenic effect of NOTCH1 on primary melanoma cells was mediated by beta-catenin, which was upregulated following NOTCH1 activation. Inhibiting beta-catenin expression reversed NOTCH1-enhanced tumor growth and metastasis.
Microarray studies of the mouse presomitic mesoderm transcriptome demonstrated that the segmentation clock drives the periodic expression of a large network of cyclic genes involved in cell signaling. Mutually exclusive activation of the Notch-fibroblast growth factor ( FGF) and Wnt pathways during each cycle suggested that coordinated regulation of these 3 pathways underlies the clock oscillator. Another study identified two clusters, the first cluster contains the known cyclic genes of the Notch pathway: Hes1, Hes5, Hey1, Id1, and Nrarp, a direct target of Notch signaling. In the same cluster as the Notch pathway were members of the FGF-MAPK pathway, including Spry2 and Dusp6. The second cluster of periodic genes contained genes cycling in an opposite phase to the Notch- FGF cluster. This cluster included a majority of the cyclic genes associated with Wnt signaling, including Dkk1, c Myc, Axin2, Sp5, and Tnfrsf19.
NOTCH and MYC regulate two interconnected transcriptional programs containing common target genes that regulate cell growth in primary human T-cell lymphoblastic leukemias. In bone marrow progenitor cells and T-cell acute lymphoblastic leukemia (T-ALL) cell lines, constitutively active NOTCH1 transcriptionally activated the NFKB pathway via the IKK complex, thereby causing increased expression of NFKB target genes.
Expression of NOTCH1 in human keratinocytes was under the control of P53. NOTCH1 suppressed tumor formation through negative regulation of ROCK1/ROCK2 and MRCK-α (CDC42BPA), which are effectors of small RHO GTPases implicated in neoplastic progression.
Some T-ALL cells show resistance to γ-secretase inhibitors that act by blocking NOTCH1 activation. Using microarray analysis, PTEN was identified as the gene most consistently downregulated in γ-secretase inhibitor-resistant T-cell lines. Studies in normal mouse thymocytes indicated that Notch1 regulated Pten expression downstream.
Genes in the Notch pathway were expressed in mature podocytes in humans and in rodent models of diabetic nephropathy and focal segmental glomerulosclerosis.
Notch activity via cell–cell contacts generates molecular differences between adjacent cells. The Notch pathway can mediate both instructive and lateral signaling in neural differentiation and tumorigenesis. Many Notch regulatory processes have been identified, but are not yet truly characterized. Notch activity regulation by ligand inhibitory effects is well described, but its mechanism of action is still unclear. The role and mechanisms of Notch and ligand trafficking are not well understood. CSL-independent Notch signaling remains undefined, both as a molecular pathway and in its effects. Further work is necessary to understand Notch signaling in all its complexity, and to provide insight into how to tackle Notch signaling in a more specific way to better approach different clinical contexts.
This study was supported by a grant of the Korea Healthcare technology R&D Project, Ministry for Health, Welfare and Family Affairs (A084022) and BRL (Basic Research Laboratory) program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (2010–0001200).
- Beckerman M. Cell Fate and Polarity. In: Molecular and cellular signaling. New York: Springer; 2005. p. 305–29.Google Scholar