Dickkopf-3 Causes Neuroprotection by Inducing Vascular Endothelial Growth Factor.

Carla Letizia Busceti,Giovanna Bozza,Giuseppe Battaglia,Massimo Volpe,Francesco Fornai,Federica Mastroiacovo,Valeria Bruno,Ferdinando Nicoletti,Anna Traficante,Paola Di Pietro,Christof Niehrs,Franca Bianchi,Luisa Di Menna,Speranza Rubattu

doi:10.3389/fncel.2018.00292

Abstract

Dickkopf-3 (Dkk3) is an atypical member of the Dkk family of Wnt inhibitors, which has been implicated in the pathophysiology of neurodegenerative disorders. However, the role of Dkk3 in mechanisms of cell degeneration and protection is unknown. We used Dkk3 knockout mice to examine how endogenous Dkk3 influences ischemic brain damage. In addition, we used primary cultures of astrocytes or mixed cultures of astrocytes and neurons to investigate the action of Dkk3 on cell damage and dissect the underlying molecular mechanisms. In a model of focal brain ischemia induced by permanent middle cerebral artery (MCA) occlusion (MCAO) Dkk3−/− mice showed a significantly greater infarct size with respect to their wild-type counterparts at all time points investigated (1, 3 and 7 days after MCAO). Immunohistochemical analysis showed that Dkk3 expression was enhanced at the borders of the ischemic focus, and was predominantly detected in astrocytes. This raised the possibility that Dkk3 produced by astrocytes acted as a protective molecule. We tested this hypothesis using either primary cultures of cortical astrocytes or mixed cortical cultures containing both neurons and astrocytes. Genetic deletion of Dkk3 was permissive to astrocyte damage induced by either oxidative stress or glucose deprivation. In addition, application of human recombinant Dkk3 (hrDkk3) was highly protective against oxidative stress in cultured astrocytes. We tested the hypothesis that the protective activity of Dkk3 was mediated byvascular endothelial growth factor (VEGF). Interestingly, glucose deprivation up-regulated both Dkk3 and VEGF in cultured astrocytes prepared from wild-type mice. VEGF induction was not observed in astrocytes lacking Dkk3 (i.e., in cultures prepared from Dkk3−/− mice). In mixed cultures of cortical cells, excitotoxic neuronal death induced by a brief pulse with N-methyl-D-aspartate (NMDA) was significantly enhanced when Dkk3 was lacking in astrocytes, whereas post-NMDA addition of hrDkk3 was neuroprotective. Neuroprotection by hrDkk3 was significantly reduced by pharmacological blockade of type-2 VEGF receptors and was mimicked by hrVEGF. These data offer the first evidence that Dkk3 protects both neurons and astrocytes against a variety of toxic insults, and at least in culture, protection involves VEGF induction.

Highlights

The Dickkopf (Dkk) family of Wnt inhibitors includes four secreted proteins (Dkk-1 to -4), of which Dkk3 diverges from the other members for the presence of a soggy-like domain in addition to common DKK_N and colipase fold domains (Kawano and Kypta, 2003)
To examine whether endogenous Dkk3 shapes brain vulnerability to ischemic damage, we used mice subjected to permanent middle cerebral artery (MCA) occlusion (MCAO, see section ‘‘Materials and Methods’’)
Double fluorescent staining for GFAP and Dkk3 at this time point indicated that astrocytes largely contributed to the increased expression of Dkk3 at the borders of the ischemic focus (Figure 1D), other cell types expressed Dkk3. This evidence raised the possibility that induction of Dkk3 in astrocytes could represent a defensive mechanism against ischemic neuronal death. To explore this mechanism in further detail, we examined the protective activity of Dkk3 in primary cultures of astrocytes and in mixed cortical cultures containing both astrocytes and neurons

Summary

Introduction

The Dickkopf (Dkk) family of Wnt inhibitors includes four secreted proteins (Dkk-1 to -4), of which Dkk diverges from the other members for the presence of a soggy-like domain in addition to common DKK_N and colipase fold domains (Kawano and Kypta, 2003). The role of Dkk in mechanisms of cell death and survival is context-dependent. Dkk behaves as an oncosuppressor protein by inducing apoptotic death in a variety of cancers (Hoang et al, 2004; Hsieh et al, 2004; Ueno et al, 2011; Dellinger et al, 2012; Veeck and Dahl, 2012; Yang et al, 2012; Eskander et al, 2016; Lorsy et al, 2016; Sawahara et al, 2016), it might favor cancer cell spreading by promoting angiogenesis (Untergasser et al, 2008; Zitt et al, 2008). Dkk induces kidney tubular cell death in proteinuric nephrosis (Wong et al, 2016), but, in contrast, it supports cell survival in the mouse retina (Nakamura et al, 2007). The following observations raised our interest on the involvement of Dkk in mechanisms of neurodegeneration/neuroprotection in the CNS: (i) Dkk co-localizes with β-amyloid peptide both in diffuse and classic plaques in brain tissue from patients affected by Alzheimer’s disease (AD; Bruggink et al, 2015); (ii) Dkk levels are reduced in brain tissue from AD patients and transgenic mouse models of AD (Zhang et al, 2017); (iii) brain-specific enhancement of Dkk expression in AD mice improves amyloid pathology, cognitive dysfunction and cerebral glucose metabolism (Zhang et al, 2017); and (iv) Dkk induces the expression of vascular endothelial growth factor (VEGF) in cultured endothelial cells (Busceti et al, 2017) and VEGF was shown to exert protective activity against hypoxic/ischemic and excitotoxic neuronal injury (Svensson et al, 2002; Moser and Humpel, 2005; Taoufik et al, 2008; Inada et al, 2014)

Methods

Results

Conclusion