خط مشتق سطوح کاهش یافته عامل نوروتروفیک سلول گلیال در بیماران مبتلا به افسردگی: مطالعه فراتحلیلی

Glial cell-line derived neurotrophic factor (GDNF) has been shown to promote development, differentiation, and protection of CNS neurons and was thought to play an important role in various neuropsychiatric disorders. Several studies have examined the GDNF levels in patients with depression but shown inconsistent results. In this study, we compared blood GDNF levels between depressive patients and control subjects through meta-analytic method. The effect sizes (ESs) from all eligible studies were synthesized by using a random effect model. In this meta-analysis, we included 526 patients and 502 control subjects from 12 original articles. Compared to control subjects, blood GDNF levels are significantly decreased in patients with depression (ES = −0.62, p = 0.0011). However, significant heterogeneity was found among included studies. Through subgroup analysis, we found that GDNF was still decreased in studies with major depressive disorder (ES = −0.73, p = 0.0001); in studies with non-old-age depression (ES = −1.25, p = 0.0001), but not with old-age depression; and in studies using serum samples (ES = −0.86, p < 0.0001), but not in studies using plasma sample. Meta-regression did not show moderating effects of mean age of subjects, gender distribution, and age of onset of depression. Our findings support blood GDNF levels as a biomarker of depression as a whole, but the results were modulated by psychiatric diagnosis, age of included subjects, and sampling sources. With these results, future studies are required to examine whether effective antidepressant treatment is associated with an increase in serum GDNF levels.

Depression is one of the most serious mental disorders in the world, and is associated with social and occupational disability, heavy financial burden, and high risk of suicide (Greden, 2001). Although the etiologies of depression are still unclear, evidence has supported that neurotrophic dysregulation plays an important role in the pathogenesis of depression (Duman and Li, 2012). Altered expression of multiple types of neurotrophic factors has been found in patients with depression, including brain-derived neurotrophic factor (BDNF) (Bocchio-Chiavetto et al., 2010 and Molendijk et al., 2014), nerve growth factor (de Azevedo Cardoso et al., 2014 and Pallavi et al., 2013), neurotrophin-3 (Otsuki et al., 2008 and Pallavi et al., 2013), vascular endothelial growth factor (Berent et al., 2014 and Fornaro et al., 2013), and insulin-like growth factor-1 (Sievers et al., 2014 and Weber-Hamann et al., 2009). Glial cell line-derived neurotrophic factor (GDNF) is a member of the transforming growth factor-β superfamily and is extensively distributed in mammalian brains, including hypothalamus, substantia nigra, and thalamus (Golden et al., 1998). It first binds to the GDNF-family receptor α1 (GFRα1) and then forms a complex with receptor tyrosine kinase receptor, activating intracellular tyrosine kinase domain and downstream signaling protein cascade (Airaksinen and Saarma, 2002). GDNF promotes the differentiation of dopaminergic neurons (Christophersen et al., 2007 and Lin et al., 1993) and serotonergic neurons (Ducray et al., 2006), and can increase neurite growth in various neuronal types (Ducray et al., 2006, Takaku et al., 2013 and Wakeman et al., 2014). In addition, GDNF has been shown to protect mesencephalic neuron-derived cells (Ortiz–Ortiz et al., 2011), serotonergic neurons from dorsal raphe nucleus (Hochstrasser et al., 2011), and mesencephalic dopaminergic neurons (Jaumotte and Zigmond, 2014), and glial cells (Uzdensky et al., 2013) from oxidative or neuro-inflammatory injury. Considered together, these findings suggest that GDNF may play a role in the pathogenesis of neuropsychiatric diseases through its neuroprotective effects in the brain (Pascual et al., 2008). In past few years, changes in GDNF activity have been examined in patients with depression. Takebayashi et al. (2006) first found that whole blood GDNF levels were significantly decreased in remitted patients with major depressive disorder (MDD). And the findings were replicated in some other studies (de Azevedo Cardoso et al., 2014, Diniz et al., 2012, Pallavi et al., 2013, Tseng et al., 2013, Zhang et al., 2014 and Zhang et al., 2008). On the contrary, Rosa et al. (2006) found an increase in GDNF immunocontent in patients with bipolar depression. In addition, Wang et al. (2011) reported a significant increase in plasma GDNF concentrations in patients with late-onset depression. The inconsistency of these results may be related to the psychiatric diagnosis of the patients, age of subjects, gender distribution, severity of depression, sample sources (plasma, serum, or whole blood), or concomitant physical or psychiatric illnesses. The goals of current meta-analysis are (1) to determine the difference in blood GDNF levels between patients with depression and control subjects, and (2) to examine potential variables that can modulate the difference.

The search in PubMed resulted in 57 results for initial consideration in the meta-analysis. By examining their titles and abstracts, 31 articles were excluded because they were review articles (n = 10), non-human studies (n = 19), comments on other studies (n = 1), or case report (n = 1). Next we examined the text of remaining 26 articles by inclusion criteria, 6 of them were excluded because they did not include patients with depressive disorder (Barbosa et al., 2011, Blasko et al., 2006, Dmitrzak-Weglarz et al., 2013, Fontenelle et al., 2012, Kotyuk et al., 2013a and Ricci et al., 2010), one article was excluded because they measured brain GDNF level (Michel et al., 2008), 3 articles without control subjects (Rybakowski et al., 2013, Sun et al., 2013 and Wang et al., 2014), 2 articles of genetic association but not GDNF protein level (Kotyuk et al., 2013b and Ma et al., 2013), and one article examining GDNF mRNA but not protein level (Otsuki et al., 2008) were excluded. Finally, 12 articles (de Azevedo Cardoso et al., 2014, Diniz et al., 2012, Heberlein et al., 2010, Marksteiner et al., 2011, Pallavi et al., 2013, Rosa et al., 2006, Takebayashi et al., 2006, Tseng et al., 2013, Wang et al., 2011, Zhang et al., 2014, Zhang et al., 2010 and Zhang et al., 2008) were included into the current meta-analysis, and the selection process was shown in Fig. 1. The characteristics of included articles were described in Table 1. The patient subjects of study by Tseng et al. (2013) included two independent subgroups, depressed and remitted patients, so this article was regarded as two studies.