Dickkopf 3 (Dkk3) belongs to the Dkk family of proteins, which are secreted glycoproteins that regulate the canonical Wnt/β-catenin (“Wnt”) signaling pathway. Wnt signaling is an essential cellular communication pathway that mediates a diverse array of cellular and molecular activities in embryonic development, tissue homeostasis, and disease pathogenesis. The prototypic Dkk family member, Dkk1, has been well characterized as an essential modifier of Wnt signaling. In contrast, the function of Dkk3 was initially unknown because it was a weak regulator of Wnt signaling, and mice deficient in Dkk3 had only mild phenotypes (see below), despite its sequence similarity to the other Dkk genes. However, in recent years, new functions for Dkk3 have been identified in immune regulatory pathways, malignancies, and neurogenesis, raising the exciting possibility that Dkk3 is a critical regulator of these processes and may be an important novel therapeutic target for several types of diseases.
The Dkk family includes five different proteins: Dkk1, Dkk2, Dkk3, Dkk4, and Soggy (Sgy). Dkk1–4 share 37–50% protein identity and contain two highly conserved cysteine-rich domains (CRD; cys-1 and cys-2) separated by a variable sequence linker region and also contain a signal peptide sequence, colipase region, and putative sites for N-linked glycosylation. The cysteine-rich CRD domains contain ten conserved cysteine residues in Dkk1–4, although the CRD of Dkk3 has lower sequence similarity to the other Dkk family members and has a relatively shorter linker region between the two CRDs (12 amino acids in Dkk3 and 50–55 amino acids in the Dkk1, Dkk2, and Dkk4 genes). Dkk3 also has longer N- and C-terminal regions compared to the other Dkks. The fifth family member, the Sgy gene, has sequence homology only to Dkk3 and lacks the CRD domains, indicating that it is most likely evolutionarily derived from Dkk3. Comparative sequence analysis of the Dkk family members across the evolutionary tree indicated that Dkk3 separated from the other Dkk family members prior to the divergence of cnidarians and bilaterians, suggesting an ancient phylogenetic separation, and raising the possibility that Dkk3 may have unique physiological roles that are distinct from the other Dkk family members.
Dkk3 was initially identified in mouse tissues by Glinka et al., in a study using EST database mining to search for sequences similar to the embryogenesis regulator Dkk1 (Glinka et al. 1998). Subsequently, a Dkk3 cDNA containing the complete open reading frame was cloned from a newborn mouse library by PCR amplification (Monaghan et al. 1999). Shortly after its initial isolation, Dkk3 was also cloned and characterized in an analysis of genes that were decreased in human tumor-derived cell lines; this Dkk3 clone was called REIC, which stood for reduced expression in immortalized cells (Tsuji et al. 2000). Therefore, Dkk3 is often referred to as REIC/Dkk3 to denote both its association with tumor cells and its membership in the Dkk family.
Dkk3 is expressed during embryonic development in many organs, including the neural epithelium, limb bud, bone, and heart, particularly in regions of epithelial-mesenchymal transformation (EMT). Expression of Dkk3 in fetal liver may be important for immune cell development, especially B cells. Dkk3 is also widely expressed in adult tissues, with the highest levels found in the heart, retina, adrenal cortex, and brain. In adult mouse, Dkk3 is expressed in neurons in distinct areas of the brain, including the hippocampus, medulla, and the visual area of the cerebral cortex, with weaker expression in the somatomotor area of the cerebral cortex and substantia nigra of the midbrain. Dkk3 is not expressed in GFAP-positive astroglia in the brain (Meister et al. 2015), although it is expressed in the GFAP-expressing Muller glia in the retina (Nakamura et al. 2007), implying a unique Dkk3 activity in retinal glia that is not observed in the brain. Despite the widespread expression distribution of Dkk3, Dkk3 knockout mice (Dkk3 −/− ) develop normally, are fertile, and have a mild phenotype that includes hyperactivity, increased immunological and hematological markers, and a slight decrease in lung ventilation (Barrantes Idel et al. 2006). The absence of severe phenotypes in Dkk3 −/− mice may be due to compensation from the Dkk3 homolog Soggy (Barrantes Idel et al. 2006). Alternatively, physiological stress or injury may be required for the appearance of a Dkk3-dependent phenotype, as demonstrated in a study on cardiac stress, described below (Lu et al. 2016).
The Role of Dkk3 in the Heart
Cardiac hypertrophy is an important cause of congestive heart failure and sudden death. Hypertrophy is often caused by hemodynamic overload, which leads to structural remodeling and dysfunction of the heart muscle. Recent evidence demonstrated that Dkk3 expression in the heart is reduced in patients with end-stage heart failure and in mice with pressure-overloaded cardiac hypertrophy (Zhang et al. 2014). Furthermore, overexpression of Dkk3 in rodent cardiac myocyte cultures led to reduced hypertrophic responses, and siRNA-mediated knockdown of Dkk3 led to increased hypertrophic responses. These findings were confirmed in vivo, in which cardiac hypertrophy was enhanced in mice with genetic loss of Dkk3, and the phenotype was reversed by transgenic overexpression of Dkk3. Therefore, these data provided evidence for a role of Dkk3 in cardiac protection. The authors also demonstrated in an elegant series of experiments that Dkk3 reduces cardiac hypertrophy by blocking activation of apoptosis signal-regulating kinase 1 (Ask1) signaling (Zhang et al. 2014). Therefore, Dkk3 has an important role in protecting the heart from pathological hypertrophy and acts as a negative modulator of cardiac hypertrophy in response to pressure overload. Whether Dkk3 functions as a cardiac protector through regulating Wnt signaling and Krm1/2 is an important mechanistic question that was not addressed in the study. Additional evidence supporting a role for Dkk3 as a cardioprotector came from a study by Lu et al., who demonstrated that Dkk3 reduced cardiac pathologic changes in a mouse model of familial dilated cardiomyopathy (FDCM), and the mechanism of protection involved activation of the canonical Wnt pathway and inhibition of the noncanonical Wnt pathway (Lu et al. 2016). In the Dkk3 −/− mice, loss of Dkk3 caused enhanced cardiac pathology and dysfunction and led to decreased survival compared with wild-type mice, whereas overexpression of Dkk3 in a mouse model of FDCM improved cardiac morphology and echocardiographic parameters and led to increased survival (Lu et al. 2016). Together, these studies demonstrated that Dkk3 is important for protecting cardiac tissue from physiological damage.
The Role of Dkk3 in Immune Regulation
An important route by which Dkk3 could influence tissue homeostasis throughout the body is by regulating immune cell maturation and activation processes. Initial analyses of the Dkk3 −/− mice suggested a regulatory role for Dkk3 in the immune system because the knockout mice had higher levels of IgM and increased natural killer cells compared with wild-type controls (Barrantes Idel et al. 2006). Indeed, subsequent studies demonstrated that Dkk3 is an important modulator of several immune cell types, including B cell and CD4+ and CD8+ T cells. In a recent study by Ludwig et al., Dkk3 was shown to regulate B-cell development, survival, proliferation, and autoreactivity (Ludwig et al. 2015). Loss of Dkk3 led to changes in B-cell-mediated immune responses, including altered antibody production and cytokine release. Notably, the role of Dkk3 in influencing B-cell fate was shown to be important to autoimmune disease because blocking Dkk3 with a neutralizing antibody increased the severity of disease in a mouse model of systemic lupus erythematosus (SLE, lupus) (Ludwig et al. 2015). Furthermore, Dkk3 is expressed in mouse tolerant to CD8+ T cells, and Dkk3 controls peripheral CD8+ T-cell tolerance to self-antigens, which is critical to preventing autoimmunity. Adoptive T-cell transfer experiments and blocking Dkk3 function demonstrated that Dkk3 is essential for inhibiting T-cell proliferation and IL-2 production. Inhibiting Dkk3 reversed CD8+ T-cell tolerance, leading to rejection of tumors and autologous skin grafts in mouse models, whereas soluble Dkk3 inhibited CD8+ T-cell responses (Papatriantafyllou et al. 2012). Similarly, Dkk3 was also shown to function within the local microenvironment to control CD4+ and CD8+ T-cell responses by suppressing activation and differentiation of T cells in the periphery, which is consistent with its high expression in immune-privileged sites, such as the eye, placenta, and brain (Meister et al. 2015). Inhibiting Dkk3 led to enhanced T-cell responses and increased disease severity in a mouse model of experimental autoimmune encephalomyelitis, and Dkk3 was shown to act locally during the T-cell effector phase, rather than systemically in the spleen and lymph nodes. Therefore, these findings demonstrate that Dkk3 is a novel regulator of immune responses in mice and suggest that Dkk3 may be a target for controlling immunosuppression in inflammatory diseases and transplantation (Meister et al. 2015). Although Wnt signaling has been implicated in various aspects of immune system regulation, it is not known yet whether Dkk3 regulates B- and T-cell fates by modifying the Wnt signaling pathway, or whether it acts independently from Wnt signaling.
The Role of Dkk3 in the Central Nervous System (CNS)
Dkk3 is highly expressed in developing and adult neurons in multiple regions in the brain. As expected from its widespread distribution in the CNS, Dkk3 has been implicated in essential neuronal processes, including survival, neurogenesis, and differentiation. For example, Dkk3 increases survival of mutant dopaminergic neurons (Zhang et al. 2015) and differentiating ventral midbrain neurons (Fukusumi et al. 2015), which is consistent with its anti-apoptotic activity in cell lines (Nakamura et al. 2007). Also, a recent paper by Fukusumi et al. demonstrated that Dkk3 is necessary and sufficient for differentiation and survival of a neuronal precursor subset into rostrolateral mesodiencephalic dopaminergic neurons (mdDA) in the developing mouse ventral midbrain. Loss of Dkk3 impeded mdDA differentiation, and Dkk3 was required for expression of the transcription factors LMX1A and PITX3 in neuronal precursors. Furthermore, treating pluripotent stem cells with Dkk3 and Wnt1 ligand in vitro increased the proportion of differentiated neurons of the SNcDA type (Fukusumi et al. 2015). Because dopaminergic neurons are lost during Parkinson’s disease (PD), these studies suggest that Dkk3 may be developed as a therapeutic strategy for replacing neurons in PD patients. The demonstration of Dkk3 expression in many other regions of the CNS suggests that Dkk3 has additional roles in the developing and/or adult brain and retina. Indeed, the hyperactivity phenotype in Dkk3 −/− mice suggests that there are novel functions of Dkk3 in the brain that are yet to be discovered.
Dkk3 and Cancer
Numerous studies have reported that loss of Dkk3 is associated with various cancers, leading to the conclusion that Dkk3 functions as a tumor suppressor. It is notable that the original description of Dkk3 function was as a molecular marker with reduced expression in immortalized cells (REIC) compared to expression in normal cells (Tsuji et al. 2000). Reduced Dkk3 expression is implicated in oncogenesis in many cancer types, including the prostate, testes, ovarian, breast, gastrointestinal, malignant glioma, bone, renal, liver, non-small cell lung cancer, and many others (for a complete review, please see (Veeck and Dahl 2012)). Dkk3 expression levels are decreased in tumor cells due to epigenetic changes (hypermethylation) of its promoter region, as well as by direct reduction of Dkk3 transcripts via transcriptional inhibition by the oncogene MycN. Thus, loss or reduction of Dkk3 protein levels would relieve cancer cells of the tumor-suppressing activity of Dkk3 and permit aberrant proliferation and survival of the cancer.
Because of the association of Dkk3 expression with cancer, efforts have been made by several research groups to utilize Dkk3 levels as a potential biomarker for tumor progression and disease prognosis. For example, methylation levels in the Dkk3 promoter were quantified and associated with survival prediction in cervical cancer, and downregulated Dkk3 expression was a marker of poor prognosis in endometrial cancer (Dellinger et al. 2012). Additionally, Dkk3 expression in serum was shown to have diagnostic application for detecting early stage disease in colorectal cancer patients before symptoms are evident, when used in a biomarker panel along with two other gene markers (Fung et al. 2015).
Targeted overexpression of Dkk3 to tumors has shown promising results in experimental gene therapy studies in rodent models of several cancers. Dkk3 was incorporated into an adenovirus for sustained gene expression, and it was shown to inhibit tumor growth, induce cancer cell-specific apoptosis, and reduce EMT in cancers of the pancreas, prostate, and breast. The mechanisms of action for antitumor activity of Dkk3 include systemic immunity effects on the tumor, maintenance of tumor cells in a dedifferentiated state (Zenzmaier et al. 2012), enhanced anticancer immune activity of splenocytes, reduced Wnt signaling, and localized apoptotic and ER-stress-mediated JNK activity in the tumor. Based on these preclinical results for adenoviral Dkk3 in animal models, a clinical trial is currently in phase 1/2a to test Dkk3 therapy for localized prostate cancer (ClinicalTrials.gov identifier NCT01931046).
In contrast, there are several reports that showed that Dkk3 acts as a tumor-promoting gene. Increased Dkk3 expression was observed in esophageal adenocarcinoma, and Dkk3 treatment of adenocarcinoma cells increased proliferation and invasion in cultured cells and in NOD/SCIDƔ mice by modulating the TGFβ signaling pathway (Wang et al. 2015). Dkk3 also promoted tumor invasion and neoangiogenesis via TGFβ in esophageal cancer and was oncogenic in oral squamous cell carcinoma independent of Wnt signaling. Because TGFβ can act as a tumor suppressor early in oncogenesis yet promotes metastasis in advanced cancer, the authors concluded that Dkk3 activities in cancer, similar to TGFβ, depend on tumor type and stage (Wang et al. 2015).
Dickkopf 3 (Dkk3), also known as REIC, is the least conserved member of the Dkk family of proteins, which are secreted glycoproteins that regulate the canonical Wnt/β-catenin signaling pathway. The prototypic member Dkk1 is a powerful inhibitor of Wnt signaling, but Dkk3 binds atypically to Dkk1 receptors, showing an interaction with Krm1/2 but not with LRP5/6, and induces Wnt signaling activation, inhibition, or even has no effect, depending on the cell type and tissue. Dkk3 is expressed in numerous cell types in developing and adult tissues, but the Dkk3 −/− mouse had a very mild phenotype, which initially made the physiological role for Dkk3 unclear. However, in recent years, a flurry of publications has described new roles for Dkk3 in cardiac protection, immune regulation, and neurogenesis. Furthermore, numerous studies have demonstrated that Dkk3 acts as a tumor suppressor in a large number of cancer types and induces tumor cell-specific apoptosis, and Dkk3 expression levels have been proposed as a biomarker for cancer prognosis. Additionally, a clinical trial is currently underway that uses adenoviral vectors to overexpress Dkk3 in tumors. Because of the role of Dkk3 in regulating key homeostatic events, such as B- and T-cell maturation and neuronal survival, there has been increasing interest in Dkk3 action and its potential as a therapeutic target. An unexplored area of research is the precise role of Dkk3 in the CNS and whether it acts by regulating the Wnt pathway or independent of Wnt signaling. Future studies will further characterize mechanisms of action of Dkk3 and identify interacting proteins, all of which will lead to improved understanding of this interesting molecule.
- Dellinger TH, Planutis K, Jandial DD, Eskander RN, Martinez ME, Zi X, et al. Expression of the Wnt antagonist Dickkopf-3 is associated with prognostic clinicopathologic characteristics and impairs proliferation and invasion in endometrial cancer. Gynecol Oncol. 2012;126:259–67.PubMedPubMedCentralCrossRefGoogle Scholar
- Fukusumi Y, Meier F, Gotz S, Matheus F, Irmler M, Beckervordersandforth R, et al. Dickkopf 3 promotes the differentiation of a rostrolateral midbrain dopaminergic neuronal subset in vivo and from pluripotent stem cells in vitro in the mouse. J Neurosci: Off J Soc Neurosci. 2015;35:13385–401.CrossRefGoogle Scholar