Brain Activation during Memory Retrieval is Associated with Depression Severity in Women

Highlights

• Depression is associated with structural and functional brain alterations

• Morris water task used to probe memory in women with varying depression severity

• Hippocampal dysregulation was evident during spatial memory performance

• Functional activation of the hippocampus correlated with depression severity level

• Probing hippocampal activation during memory may help elucidate depression etiology

Abstract

Major depressive disorder (MDD) is a debilitating disorder that interferes with daily functioning, and that occurs at higher rates in women than in men. Structural and functional alterations in hippocampus and frontal lobe have been reported in MDD, which likely contribute to the multifaceted nature of MDD. One area impacted by MDD is hippocampal-mediated memory, which can be probed using a spatial virtual Morris water task (MWT). Women (n=24) across a spectrum of depression severity underwent functional magnetic resonance imaging (fMRI) during MWT. Depression severity, assessed via Beck Depression Inventory (BDI), was examined relative to brain activation during task performance. Significant brain activation was evident in areas traditionally implicated in spatial memory processing, including right hippocampus and frontal lobe regions, for retrieval > motor contrast. When BDI was included as a regressor, significantly less functional activation was evident in left hippocampus, and other non-frontal, task relevant regions for retrieval > rest contrast. Consistent with previous studies, depression severity was associated with functional alterations observed during spatial memory performance. These findings may contribute to understanding neurobiological underpinnings of depression severity and associated memory impairments, which may have implications for treatment approaches aimed at alleviating effects of depression in women.

Introduction

Major depressive disorder (MDD) is a debilitating disorder, which interferes with daily functioning, and occurs at a markedly higher rate in women relative to men. Without treatment, MDD may become a chronic burden associated with an increasing disability over time, particularly in women (Kessler, 2003; Stegenga et al., 2012; Whiteford et al., 2015). According to the Global Burden of Disease report, depression was ranked as the third leading cause of disability by the World Health Organization and was predicted to continue to be a leading cause of burden in 2030 (WHO, 2004). The impact of depression is 50% higher in women than men,15-44 years of age, which contributes to an important source of lost years of healthy life (e.g., lost years of life represents the years of healthy life lost through time spent in states of less than full health). Further, it is the leading cause of disease burden for women in low and middle as well as high income countries (WHO, 2004).

Among the impacts of MDD on daily functioning, cognitive impairments have been reported across several domains, which include attention, working memory, other executive functions, emotional processing, and learning and memory (e.g., verbal memory) (Darcet et al., 2016). Structural and functional brain alterations associated with MDD likely contribute to such cognitive deficits, as well as to other multifaceted impacts of this condition (Belleau et al., 2019). Alterations include reduced brain volume (Bora et al., 2012; Bremner et al., 2000; Han et al., 2016; Roddy et al., 2019) and aberrant functional activation (Castanheira et al., 2019; Fonseka et al., 2018). Frontolimbic regions have frequently been implicated in brain changes associated with MDD (Bora et al., 2013; Jamieson et al., 2019; Pizzagalli, 2011). For instance, increased white matter volume in uncinate fasciculus and decreased gray matter density in ventromedial prefrontal cortex were observed in patients with MDD, the extent of which was correlated with depression scores (Klauser et al., 2015). Importantly, there are mixed findings related to memory impairments associated with depression, suggesting heterogeneity across MDD, with differential impacts being related to depression subtype and severity (Burt et al., 1995; King et al., 2010; Quinn et al., 2012).

While there are a large number of studies assessing and reporting executive functions and associated frontal lobe circuitry deficits in MDD, hippocampal structural and function have been strongly linked to MDD pathology (Roddy et al., 2019). In a study using voxel-based morphometry to investigate gray matter volume, via data collected using magnetic resonance imaging (MRI), patients with recurrent depression exhibited more pronounced reductions in substructure volume in the left CA1 region of the hippocampus relative to healthy controls and those with first-presentation MDD, interpreted as a potential marker of disease progression (Roddy et al., 2019) . There have been a number of studies using functional magnetic resonance imaging (fMRI), which have demonstrated some mixed findings. Patients with MDD demonstrated increased bilateral hippocampus and parahippocampal activation during performance of a process-dissociation task, a task designed to separate contributions of recollection and familiarity to memory performance (Alders et al., 2019; Jacoby, 1991). Hippocampal activation differences also were observed in the absence of group performance differences (patients with MDD versus healthy comparison subjects) on all of the memory domains assessed (e.g., recollection memory, habit memory trials and non-item trials). These findings potentially reflect greater hippocampal recruitment for those in the early course of MDD, relative to those with a prolonged illness and repeated number of depressive episodes (Alders et al., 2019). These findings also are consistent with earlier findings of altered hippocampal activation during a recollection memory performance task in patients with extensive history of major depression (3 or more previously treated depressive episodes) (Milne et al., 2012). Decreased recruitment, however, was reported in the right hippocampus and left parahippocampal gyrus in patients with MDD relative to non-depressed controls. Further, impairments in recollection performance in patients with MDD was associated with diminished hippocampal engagement, suggesting that the hippocampus is particularly vulnerable to the impact of MDD.

Given that difficulties in attention and emotional processing can modulate cognitive domains such as learning and memory and decision-making (Robinson et al., 2017), additional focus on better characterizing the impact of MDD on learning, memory and hippocampal function is warranted. Studies examining memory deficits have shown mixed results, particularly for spatial memory (Quinn et al., 2012). One commonly used tool that has been applied to investigate memory in preclinical animal models are maze type tasks, which require frontally-mediated spatial working memory, and hippocampally-mediated spatial learning and memory to be solved. As such, the Morris water task (MWT) (Morris, 1984), a maze type task, has been used extensively in animal research to investigate spatial memory ability and related hippocampal circuitry. Adapted for use in humans during fMRI, the virtual MWT performed in conjunction with fMRI has been applied to study healthy adults (Sneider et al., 2011) and adolescents (Sneider et al., 2018), and adults with a history of marijuana use (Sneider et al., 2013), revealing evidence that the task elicits hippocampal and prefrontal brain activation. While MWT has been used to document learning and memory associated with MDD in human adults with whole-head magnetoencephalography (MEG) (Cornwell et al., 2010), to date, the MWT combined with fMRI has not been applied to study adults with MDD.

Sex differences in prevalence and severity of depression have been established to be greater in women (Kuehner, 2017; Stegenga et al., 2012), as well as sex differences in performance (Sneider et al., 2015) and hippocampal activation (Sneider et al., 2011) observed during the MWT. Accordingly, the objective of the current study was to use the MWT with fMRI to investigate neurocircuitry engaged during task performance in women across a range of depression severity, assessed via Beck Depression Inventory (BDI) scores. It was predicted that greater depression severity in women would be associated with less MWT-related fMRI activation in hippocampus and frontal lobe regions, brain areas involved in spatial memory and also implicated in pathophysiology of MDD.