Abnormal T2 relaxation time in the cerebellar vermis of adults sexually abused in childhood:

Abstract

Recent studies suggest that childhood sexual abuse (CSA) elicits a cascade of neurohumoral events that affect brain development and is also a risk factor for the later development of substance abuse. We hypothesize that the cerebellar vermis may be a key region linking these observations. The vermis has a protracted ontogeny and a high density of glucocorticoid receptors, rendering it highly susceptible to early stress. The vermis modulates dopamine turnover in the accumbens and receives direct dopamine input through fibers with dopamine transporters. To test this hypothesis, steady-state functional magnetic resonance imaging (fMRI) (T2 relaxometry) was performed to assess resting blood flow in the vermis of 24 young adults (18–22 years) selected by screening from a large community sample. Eight subjects had a history of repeated CSA but were unmedicated and not under psychiatric care. Sixteen subjects were age-matched controls who had no personal or family history of Axis I psychiatric disorders. All subjects were screened to exclude known abnormalities affecting brain development, and any history of drug or alcohol abuse. CSA subjects had higher T2 relaxation time (T2-RT) than controls in the vermis but not in cerebral or cerebellar hemispheres. Vermal T2-RT correlated strongly with Limbic System Checklist (LSCL-33) ratings of temporal lobe epilepsy (TLE)-like symptomatology. From 537 prescreened young adults we found that their frequency of substance use was associated with a monotonic increase in LSCL-33 ratings and depression scores. Together these findings suggest that early trauma may interfere with the development of the vermis, and produce neuropsychiatric symptoms associated with drug use.

Introduction

Early stress or maltreatment is an important risk factor for the later development of substance abuse (Ellason et al., 1996, Janikowski et al., 1997, Kendler et al., 2000, McClanahan et al., 1999, Najavits et al., 1997, Simpson et al., 1994, Wilsnack et al., 1997). We have spent the last several years pursuing the hypothesis that childhood abuse or neglect induces a cascade of neurohumoral and physiological events that affect brain development (Ito et al., 1993, Ito et al., 1998, Teicher et al., 1993, Teicher et al., 1994, Teicher et al., 1997, Schiffer et al., 1995). Briefly, we found that subjects reporting childhood abuse had markedly increased scores on the Limbic System Checklist-33 (LSCL-33), which was devised to ascertain the frequency of occurrence of symptoms suggestive of temporal lobe epilepsy (TLE) (Teicher et al., 1993). Childhood abuse was associated with a two-fold increased incidence of clinically significant electroencephalographic (EEG) abnormalities, which were restricted to the left hemisphere (Ito et al., 1993). Abused children had highly abnormal measures of left hemisphere EEG coherence, suggesting deficient cortical maturation and differentiation (Ito et al., 1998; Teicher et al., 1997). Bremner et al. (1997) and Stein (1997) found that women with post-traumatic stress disorder (PTSD) who were abused in childhood had a smaller left, but not right, hippocampus. Abuse was also associated with prominent thinning of the middle portion of the corpus callosum, affecting males more than females (DeBellis et al., 1999, Teicher et al., 1997). These anatomic anomalies may contribute to the psychiatric syndromes reported to occur with increased frequency in childhood trauma survivors (Teicher et al., 1994, Teicher, 2000).

Preclinical studies have experimentally established the enduring effects of early experience on brain development and have elucidated potential mechanisms and key target areas. One major finding has been the discovery that excessive levels of glucocorticoids, or stress, damage brain regions with high densities of glucocorticoid receptors, such as the hippocampus (e.g. Sapolsky et al., 1988, McEwen, 1999).

A brain region that should also be extraordinarily sensitive to the effects of early maltreatment is the cerebellar vermis. Like the hippocampus, the vermis has a protracted period of postnatal ontogeny and may produce granule cells postnatally (Altman and Bayer, 1997). The vermis also has the highest density of glucocorticoid receptors during development, exceeding that of the hippocampus (Lawson et al., 1992, Pavlik and Buresova, 1984, Sanchez et al., 2000) and may be particularly vulnerable to the effects of stress hormones (Schapiro, 1971, Ferguson and Holson, 1999).

Our interest in the cerebellar vermis stems from the work of Harlow (Harlow et al., 1965) on the deleterious effects of maternal separation and early isolation. Mason and Berkson (1975) showed that swinging wire primate surrogates greatly diminished the degree of psychopathology (aggression, self-stimulation) seen in adults as a result of these adverse early experiences. Prescott (1980) suggested that both proprioceptive and vestibular stimulation was protective, and Berman (1997) found that lesions of the vermis, which receives major input from the vestibular system, eliminated aggressive behavior. Heath (1972) found that primates reared in this manner had epileptiform EEG patterns in their fastigial nuclei, which project from the vermis to the limbic system and modulate seizure susceptibility (Heath, 1977, Cooper and Upton, 1985, Maiti and Snider, 1975, Strain et al., 1979).

The purpose of this study was to ascertain whether there were alterations in blood flow in the cerebellar vermis in a carefully selected group of young adults with a history of repeated sexual abuse using a novel functional magnetic resonance imaging (fMRI) procedure (T2 relaxometry) to derive steady state blood flow measures (Teicher et al., 2000) that correlate strongly with cerebral blood volume (CBV) as determined by Dynamic Susceptibility Contrast MRI (Anderson et al., 2000b). T2 Relaxometry (T2-RT) provides an indirect assessment of changes in steady-state blood flow as a result of magnetic susceptibility effects due to regional changes in deoxyhemoglobin concentration. This is known as the blood oxygenation level dependent (BOLD) effect or BOLD effect. In contrast to standard BOLD techniques that use background subtraction to detect acute task-dependent blood flow changes, T2-RT can be used to assess long-term steady-state blood flow changes, which may occur with chronic drug adminstration or as a response to environmental stressors. For example, during a typical fMRI experiment, the acute BOLD signal response to enhanced neuronal activity is due, in large part, to a mismatch between blood flow and oxygen extraction which does not persist under steady-state conditions (Punwani et al., 1998). Instead, regional blood flow is regulated to appropriately match perfusion with ongoing metabolic demand (Buxton et al., 1998), and deoxyhemoglobin concentration becomes constant between regions in the steady-state. Therefore, regions with greater continuous activity would be perfused at a greater rate, and these regions would receive, over time, a greater volume of blood and a greater number of deoxyhemoglobin molecules per volume of tissue. Conversely, with long-term decreases in the activity of a region, there should be a lessening of the paramagnetic properties of the region, which would be detectable as an increased T2 relaxation time.

There are compelling reasons to hypothesize that the vermis plays a role in modulating response to addictive drugs. The vermis, through its fastigial projections to the ventral tegmental area and locus coeruleus, exerts strong effects on the turnover of dopamine and norepinephrine in the caudate and nucleus accumbens (Albert et al., 1985, Nieoullon et al., 1978, Snider et al., 1976, Snider and Snider, 1977, Tellerman et al., 1979). There are also important afferent and efferent pathways between the vermis and the hypothalamus (Dietrichs and Haines, 1986, Supple, 1993). The vermis receives direct monoamine projections from the midbrain (Ikai et al., 1992, Steindler, 1981, Kerr and Bishop, 1991) and has dopamine receptors (Khan et al., 2000) and transporters (Melchitzsky and Lewis, 2000). The vermis is affected by stimulants, cocaine and ethanol (Cavanagh et al., 1997). Methylphenidate exerts robust effects on blood flow in these regions (Schweitzer et al., 2000, Anderson et al., 2000a). The putative anti-addictive agent, ibogaine, exerts profound effects on the vermis (O'Hearn and Molliver, 1993) and induces transient ataxia (Anderson, 1998). Attentional deficit, hyperactivity disorder (ADHD) is a serious risk factor for development of substance abuse (Schubiner et al., 2000, Wilens et al., 1998), and the most consistent anatomical finding in ADHD is reduced vermal size (Berquin et al., 1998, Castellanos et al., 2001). Hence, it is plausible that the cerebellar vermis may be affected by early stress, and that it may be a component of a neural circuit modulating risk for substance abuse.

The purpose of this study was to ascertain whether there were alterations in blood flow in the cerebellar vermis in a carefully selected group of young adults with a history of repeated sexual abuse. Further, we sought to determine whether there were meaningful associations between the degree of vermal blood flow and certain categories of psychiatric symptoms. Finally, we sought to examine the relevance of these associations by ascertaining whether these psychiatric symptoms were associated with substance use in a large sample of college students.