The GLI2 gene encodes one of the three zinc finger transcription factors that are involved in Hedgehog signaling. It was identified by using GLI1 cDNA as a probe (Ruppert et. al, 1988). Human GLI2 gene comprises 13 exons (Roessler et al., 2005).

Protein: Human GLI2 was originally identified as a Tax-helper protein (THP) that binds to Tax-responsive element in the long terminal repeat of the human T-cell leukemia virus (Tanimura et al., 1998). However, when compared to orthologous GLI2 genes from different species, the human mRNAs lacked a part of the 5 region encoding the evolutionarily conserved N-terminus of GLI2. Roessler et al discovered a 5 sequence encoding 328 amino acids and showed that this previously undescribed amino-terminal repressor domain was essential for the dominant negative activity of the human GLI2 (Roessler et al., 2005). Thus, the full length GLI2 contains 1586 amino acids and in vitro transcriptional activity of the full length GLI2 is about 30 fold lower than the N-delta GLI2, previously known as GLI2 (Roessler et al., 2005). The predicted molecular weight of the full length GLI2 is 197 kDa.
GLI2 promoter: The promoter region of GLI2 has been cloned and defined by Dennler et al. (Dennler et al., 2007). 5-RACE was used to identify the transcription start site and the promoter sequence was defined to be about 1600-bp upstream of the transcription initiation site (Dennler et al., 2007).
Binding sequence: The consensus binding sequence GACCACCCA for the three GLI transcription factors was first described by Kinzler et al. (Kinzler, et al., 1988) and many direct GLI target genes have been identified using the consensus binding sequence. Later, other natural non-consensus sequences have been found to be comparably active in the luciferase reporter assays in response to activated GLI2 expression (Winklmayr et al., 2010).
Structure: GLI2 belongs to the C2H2 type zinc finger domain protein family. It has a DNA binding domain of five zinc fingers, which are very highly homologous in the three GLIs. The crystal structure revealed by Pavlevich et al. (1993) showed that the fourth and fifth zinc fingers of GLI are responsible for all but one of the protein-DNA base contacts in a conserved nine base-pair region (Pavletich et al., 1993).
GLI2 protein has the repressor domain at N-terminus (Roessler et al., 2005) and activation domain at its C-terminus (Figure1). Unmasking of the strong activation potential of GLI2 through modulation of the N-terminal repression domain is one of the key mechanisms of the Shh signaling (Sasaki et al., 1999).
There is also a processing determinant domain of GLI2 at the C-terminal end of the protein and this domain was found to be the determining factor for the differential processing of the protein by proteasomes (Pan et al., 2007). It is a 197 aa residue between the zinc-finger DNA binding domain and first of 6 Protein Kinase A (PKA) site of GLI2 protein (Li et al., 2011).
In addition to proteolytic processing, alternative splicing may be another important regulatory mechanism for the modulation of repression and activation properties of the GLI2 protein and the generation of protein isoforms with different activities (Speek et al., 2006).
There is also a predicted nuclear export signal and nuclear localization signal domains in the GLI2 protein as depicted in Figure 1.

In the integrative genome analysis of GLI2 orthologs, it was found that human GLI2 mRNA was expressed in embryonic stem cells, NT2 cells, fetal lung, fetal heart, regenerating liver, gastric cancer, and other tumors. Mouse GLI2 mRNA was expressed in unfertilized egg, ES cells, and EG cells (Katoh et al., 2008).

In a study tracking the cellular localization of GLI2, it has been described that GLI2 shuttles in and out of cilium and with the activation of HH signaling, the protein localizes to the nucleus from its primary cytoplasmic localization. However, exogenous GLI2 localizes into the cytoplasm (Kim et al., 2009a).

Hedgehog signaling mediated functions
GLI2 is one of the components of Hedgehog signaling and one of the three zinc finger transcription factors involved in this signaling pathway. GLI2 is thought to be the primary transcription activator of HH signaling along with GLI1 and to lesser extent GLI3 (Li et al., 2011). Hedgehog signaling pathway is a well characterized pathway and the role of GLI transcription factors are modeled distinctly in the absence or presence of HH ligands. In the absence of HH ligand, GLI proteins are processed and repressor forms are translocated into the nucleus whereas in the presence of HH ligand with the activation of SMO, the active forms of GLI proteins are translocated to the nucleus where they bind and activate their target gene expression (Javelaud et al., 2012). Processing of GLI2 and GLI3 N-terminus represents a critical mechanism allowing regulation of target genes. GLI2 functions as a primary activator of HH signaling along with GLI1 while GLI3 exhibits weak activity and is considered to function as a repressor. Due to the lack of repressor domain, GLI1 acts also as an activator where it is a direct target of GLI2 and has been shown to complement some of GLI2 functions in vivo (Dahmane et al., 1997; Ikram et al., 2004; Bai and Joyner, 2001).
There are some other proteins that participate in the transduction of the HH signaling by interacting with GLI proteins. An example is a serine/threonine kinase ULK3 is able to phosphorylate GLI proteins and promote their transcriptional activity but in the absence of HH signaling, ULK3 interacts with the inhibitory protein SUFU promoting GLI2 to its repressor form (Maloverjan et al., 2010). Another such proteins are DYRK, dual-specificity tyrosine-regulated kinases, DYRK2 acts as a scaffold protein for E3 unbiquitin ligase complex and phosphorylates GLI2 for proteasome degradation (Maddika et al., 2009) while DYRK1B inhibits GLI2 activity and promotes GLI3 repressor processing (Lauth et al., 2010).
Hedgehog signaling independent functions
GLI2 protein has also been found to be affected by other signaling pathways in hedgehog independent way stressing the importance of this protein. There are increasing evidences that GLI2 expression and activation is regulated by other signaling cascades and molecules.
In a study by Dennler et al. (2007), GLI2 has been shown to be a direct transcriptional target of the TGF-β/SMAD pathway independent from HH receptor signaling, and requires a functional Smad pathway in human normal fibroblasts and keratinocytes and cancer cell lines.
It has also been shown that GLI2 can induce the secretion of IL6 in stromal cells as a model of tumor microenvironment via sonic hedgehog independent signaling cascade and mediate increased immunoglobulin secretion in B-cell malignancies (Elsawa et al., 2011).
In a study on pancreatic cancer cells oncogenic KRAS strongly stimulated GLI function where this effect was ligand independent and occurred downstream of SMO since Cyclopamine treatment did not inhibit the KRAS induced GLI superactivation (Ji et al., 2007).
TGF-β, β-catenin and hyperactive RAS/RAF/MEK/ERK-mediated signaling upregulates GLI2 expression/activity in tumor cells independent of the presence of hedgehog ligand and hyperactivity of these alternate cell signaling pathways is known to occur in many different types of cancer (Jenkins, 2009; Lauth and Toftgard, 2007).
GLI2 and the tumor microenvironment.
Metastasizing tumor cells hijack many of the pathways that play major roles during normal development. Many of the embryonic developmental signaling pathways, such as Wnt, Hedgehog (HH), and Notch pathways, affect the survival of tumor stem cells and orchestrate a complex microenvironment that promotes tumor survival and progression (Das et al., 2012). In this context, several studies have addressed a putative role of GLI2 in the tumor microenvironment in mediating the cellular crosstalk in certain cancers.
Evidence of the crosstalk between TGF-beta and HH signaling pathways exists in non-canonical HH signaling pathway which could challenge cancer therapy (Javelaud et al., 2012). TGF-beta signalling is known to modulate the tumor microenviroment by orchestrating fibroblast chemotaxis and activation of immune cells and stromal-epithelial cross-talk (Stover et al., 2007). TGF-β also induces GLI2 expression in various human cell types, including normal fibroblasts and keratinocytes, as well as various cancer cell lines affecting tumor progression, apoptosis and cell cycle (Javelaud et al., 2011a). To date, the precise role of TGF- β-GLI2 signaling axis in the tumor microenvironment has not been addressed.
One study stressing the role of GLI2 on tumor microenvironment showed that GLI2 can mediate secretion of IL-6 in bone marrow stromal cells via a SHH-independent signaling cascade. The cytokine CCL5 initiates signaling through the CCR3 receptor leading to activation of the PI3K/AKT signaling pathway ultimately increasing the expression of IL-6. IL-6 in turn leads to increased immunoglobulin secretion by malignant B-cells (Elsawa et al., 2011).
A role for GLI2 in the tumor microenvironment in breast cancer metastasizing to the bone has been described. The production of parathyroid hormone-related peptide (PTHrP), a major factor involved in tumor-induced osteolysis caused by metastatic breast cancers, is driven at least in part by GLI2 both in vitro and in vivo (Sterling et al., 2006).
In B-cell chronic lymphocytic leukemia (B-CLL), it has been demonstrated that inhibiting HH signaling in stroma inhibits bone marrow stromal cell-induced survival of B-CLL cells, suggesting a role for HH signaling in the survival of B-CLL cells evidence supported by gene expression profiling of primary B-CLL cells and disease progression of B-CLL patients with clinical outcome (Hegde et al., 2008).

GLI2 shares 44 % sequence identity with GLI3 and is structurally more similar to GLI3.

Loss-of-function mutations in the human GLI2 gene are associated with a distinctive phenotype whose primary features include defective anterior pituitary formation and pan-hypopituitarism, with or without overt forebrain cleavage abnormalities, and HPE-like mid-facial hypoplasia (Roessler et al., 2003).