Hippocampal neuroplasticity in major depressive disorder

Highlights

• Hippocampus volume is decreased in patients with major depressive disorder.

• Localized differences in the hippocampus are more prominent than global differences.

• Hippocampal volume reductions are specific to the cornu ammonis and dentate gyrus.

Abstract

One of the most replicated findings has been that hippocampus volume is decreased in patients with major depressive disorder (MDD). Recent volumetric magnetic resonance imaging (MRI) studies suggest that localized differences in hippocampal volume may be more prominent than global differences. Preclinical and post-mortem studies in MDD indicated that different subfields of the hippocampus may respond differently to stress and may also have differential levels of plasticity in response to antidepressant treatment. Advances in high-field MRI allowed researchers to visualize and measure hippocampal subfield volumes in MDD patients in vivo. The results of these studies provide the first in vivo evidence that hippocampal volume reductions in MDD are specific to the cornu ammonis and dentate gyrus hippocampal subfields, findings that appear, on the surface, consistent with preclinical evidence for localized mechanisms of hippocampal neuroplasticity. In this review we discuss how recent advances in neuroimaging allow researchers to further understand hippocampal neuroplasticity in MDD and how it is related to antidepressant treatment, memory function, and disease progression.

Introduction

Major depressive disorder (MDD) is a major challenge for society affecting 2–5% of the population and is a major cause of disability worldwide (Murray and Lopez, 2013). At least 30% of patients do not remit after a year of multiple antidepressant trials (Warden et al., 2007) and this overestimates positive outcomes (Frank et al., 1991).

The causes of MDD remain uncertain, although a number of factors are known to increase risk including abuse during childhood and chronic stress as an adult (Paolucci et al., 2001, Widom et al., 2007, Bradley et al., 2008, Danese et al., 2009, Risch et al., 2009). The well-described effects of stress on risk of developing MDD have been supported by findings that there are abnormalities in the hypothalamic–pituitary–adrenal (HPA) axis in patients with MDD, and this may impact the release of glucocorticoids. That the HPA axis is dysregulated is evidenced by studies examining cortisol hypersecretion, dexamethasone non-suppression, and exaggerated responses to dexamethasone–corticotropin-releasing hormone challenges (Barden, 2004). In particular several components of the HPA axis have been implicated in the development of MDD, specifically the hippocampus (HC), amygdala and prefrontal cortex (PFC) (Pittenger and Duman, 2008, Ulrich-Lai and Herman, 2009).

One of the most replicated findings has been that HC volume is decreased in patients with MDD, with the degree of change having been confirmed by several meta-analyses of magnetic resonance imaging (MRI) studies (Videbech and Ravnkilde, 2004, McKinnon et al., 2009). Based on preclinical studies, several mechanisms, including neuronal and glial remodeling or loss, neuronal death and suppressed adult neurogenesis, apparently involving elevated levels of glucocorticoids, have been suggested as potential causative factors in low HC volume (Sapolsky, 2000, Czéh and Lucassen, 2007). MRI studies have consistently shown that the reductions in HC volumes in MDD have been associated with episode recurrence (MacQueen et al., 2003, McKinnon et al., 2009), history of childhood maltreatment (Vythilingam et al., 2002, Frodl et al., 2010) deficits in memory performance (Lee et al., 2012). Only a few MRI studies have analyzed the HC in medication-free MDD (MacQueen et al., 2003, Posener et al., 2003, Vythilingam et al., 2004, Frodl et al., 2010) while the majority of studies included participants on antidepressant treatment (Videbech and Ravnkilde, 2004, McKinnon et al., 2009). Several genetic associations have been suggested to play an important role with associations between mood, memory and HC volume (Eker et al., 2011, Kohli et al., 2011, Price et al., 2013, Dunn et al., 2015).

There is preclinical evidence that stress and glucocorticoids negatively impact HC neuroplasticity, neuronal survival, and glial survival (Czéh and Lucassen, 2007, Pittenger and Duman, 2008). Other preclinical studies have suggested that antidepressants have stress-protective effects on HC neuroplasticity (Pittenger and Duman, 2008), and such a positive effect also appears to occur in humans (Boldrini et al., 2009). Therefore, these findings might suggest that stress, possibly acting via glucocorticoids, may negatively affect HC neuronal plasticity, which in turn is reflected in decreased HC volumes (Dranovsky and Hen, 2006). This information may also suggest that one effect of antidepressant treatment would be to reverse some of these changes. Clearly, if this were known to be the case it could open up significant new possibilities for both the etiology and treatment of MDD. However, the information from most previous MRI studies has been inadequate to allow measurement of any such effects (McKinnon et al., 2009). In this review we discuss how recent advances in neuroimaging allow researchers to further understand HC neuroplasticity in MDD and how it is related to antidepressant treatment, memory function, and disease progression.