Effects of early life stress on biochemical indicators of the dopaminergic system: A 3 level meta-analysis of rodent studies
Introduction
During the perinatal period, the brain matures and is rapidly wired (Semple et al., 2013), rendering it particularly vulnerable to negative life experiences that might lastingly impact brain function and behavior (Levine, 2005). This may contribute to the well-established observation that exposure to adverse conditions during childhood is a major risk factor for the later development of psychopathologies (Teicher et al., 2016), including schizophrenia and substance abuse (Agid et al., 1999; Dube et al., 2003; Enoch, 2011; Scheller-Gilkey et al., 2002).
Prevailing evidence highlights that the dopamine (DA) system may be a prime candidate in mediating the influence of adverse events early in life on vulnerability to psychopathology (Gatzke-Kopp, 2011). The DA system develops early during the embryonic period, matures throughout adolescence, and forms stable patterns during young adulthood (Money and Stanwood, 2013). This prolonged development provides an extensive window of time in which adverse conditions early in life can tip the balance towards dysfunction (Money and Stanwood, 2013). Indeed, alterations in this system have been consistently associated with mental disorders (for a review: (Money and Stanwood, 2013)). For example, genetic variations of the DA degradation enzyme COMT are associated with schizophrenia and bipolar disorder as well as an increased risk for psychosis, autism and anxiety (Money and Stanwood, 2013). In line, the DA receptor 2 is a major target for antipsychotics.
Overall, the associative studies in humans have led to the assumption that childhood adversities result in developmental alterations of the dopaminergic system. To investigate causality, preclinical studies using animal models have adopted behavioral early life stress (ELS) paradigms to mimic negative childhood conditions, aiming to understand the neurobiological substrate by which ELS adds to the development of DA system dysfunction. Although extensive, the existing literature is quite heterogeneous: it uses disparate models and outcome measures, and each study focuses on only a limited number of variables; moreover, preclinical studies are frequently underpowered (Button et al., 2013). The resulting findings are rather incoherent and difficult to interpret. This limitation hinders our understanding of the entire biological system and its development, and delays translational applicability.
To overcome these limitations, we performed a meta-analysis, a powerful method still sparsely applied to preclinical research which allows to systematically synthesize the scientific knowledge of a specific topic. Recent advances in the field of statistics such as the 3-level approach (Assink and Wibbelink, 2016; Cheung, 2014a) along with their implementation in R packages (Cheung, 2014b; Viechtbauer, 2010) now enable researchers to use more sophisticated and robust methodology when analyzing meta-data. This method allows to include multiple data-points from a single study (nesting), without necessarily knowing their (often unreported) covariance. Ultimately, this substantially increases the flexibility of meta-analysis applications and improves the validity of the conclusions drawn.
Here, we aimed to investigate whether preclinical studies support an effect of ELS on dopaminergic signaling. We included diverse types and timings of ELS models (Fig. 1), and we operationalized the dopaminergic system by quantifying several biochemical markers in mice and rats (Fig. 2), across brain areas (Fig. 3), considering possible confounders.
We determined whether the quality of the studies affected the estimation of the results. To make our knowledge readily available to others, we organized all information in a freely accessible open-source dataset and created a user-friendly web-app as a tool to guide future (pre)clinical research (e.g. power analysis calculation), thereby avoiding unnecessary replication and limit animal experimentation.
Section snippets
Materials and methods
The review adhered to SYRCLE (Systematic Review Center for Laboratory animal Experimentation) guidelines for protocol (De Vries et al., 2015), search strategy (Leenaars et al., 2012), and risk of bias assessment (Hooijmans et al., 2014).
Study selection and data extraction
The process of study selection is illustrated in the flow chart (Fig. 4). The search string identified a total of 979 unique research papers. Statistical measurements (e.g. mean, SD and N) for quantitative analysis were extracted from 90 peer-reviewed publications that met our pre-specified inclusion criteria as described in the methods section. Three publications (Arborelius and Eklund, 2007; Kippin et al., 2008; Kosten et al., 2005) were excluded from the analysis as it was not possible to
Discussion
Schizophrenia and addiction are examples of psychiatric disorders reported to be linked to DA dysfunction. Childhood trauma is a well-documented risk factor (Gatzke-Kopp, 2011; Grace, 2016; Money and Stanwood, 2013; Teicher et al., 2016). This clinical observation led to the hypothesis that the dopaminergic system mediates the risk of ELS. Although this link has been causally investigated in more than 90 rodent publications over 20 years, no consensus has yet been reached on the extent,
Acknowledgments
This work was supported by the Consortium on Individual Development (CID), which is funded through the Gravitation program of the Dutch Ministry of Education, Culture, and Science and Netherlands Organization for Scientific Research (NWO grant number 024.001.003). R.A.S. was supported Netherlands Organization for Scientific Research (NWO Veni grant 863.13.02).
We thank Prof. Joop Hox for his input on multilevel methods, and Erik-Jan van Kesteren for his programming expertise. We also gratefully
References (147)
- et al.
Permanent dopaminergic alterations in the n. accumbens after prenatal stress
Pharmacol. Biochem. Behav.
(1994) - et al.
Both long and brief maternal separation produces persistent changes in tissue levels of brain monoamines in middle-aged female rats
Neuroscience
(2007) - et al.
Prenatal lipopolysaccharide treatment enhances MK-801-induced psychotomimetic effects in rats
Pharmacol. Biochem. Behav.
(2011) - et al.
Exogenous prenatal corticosterone exposure mimics the effects of prenatal stress on adult brain stress response systems and fear extinction behavior
Psychoneuroendocrinology
(2013) - et al.
Maternal separation followed by early social deprivation affects the development of monoaminergic fiber systems in the medial prefrontal cortex of Octodon degus
Neuroscience
(1999) - et al.
Effects of postnatal stress on dopamine mesolimbic system responses to aversive experiences in adult life
Brain Res.
(1993) - et al.
Sex-differences in the effects of early experience on the development of behavioral and brain asymmetries in rats
Physiol. Behav.
(1984) - et al.
Prenatal restraint stress: an in vivo microdialysis study on catecholamine release in the rat prefrontal cortex
Neuroscience
(2010) - et al.
The impact of maternal separation on the number of tyrosine hydroxylase-expressing midbrain neurons during different stages of ontogenesis
Neuroscience
(2011) - et al.
Attenuated disruption of prepulse inhibition by dopaminergic stimulation after maternal deprivation and adolescent corticosterone treatment in rats
Eur. Neuropsychopharmacol.
(2008)