Myo-inositol in depressive and healthy subjects determined by frontal ¹H-magnetic resonance spectroscopy at 1.5 tesla

Abstract

Myo-inositol (mI) as a precursor in the phosphatidylinositol second messenger system has been reported to be reduced in depression. By means of proton-magnetic resonance spectroscopy (¹H-MRS) the mI levels in the frontal brain were investigated in vivo in the present study. Twenty-two patients (mean age: 42.8±10.7 years) with depressive episodes according to ICD 10 (HAMD score > 17) were compared to 22 healthy subjects (28.0±5.3 years). Two voxels (30×20×20 mm³) in the frontal lobes were examined in a Siemens Magnetom SP 4000 at 1.5 T (STEAM sequence: TR = 3500 ms, TE = 55 ms). With the total creatine (Cr) as an internal standard, mI\Cr ratios were calculated to follow the mI levels. In the left frontal lobe, mI\Cr was 0.43±0.06 in depressive patients and 0.46±0.07 in healthy subjects; concerning the right frontal lobe, mI\Cr was 0.46±0.08 and 0.48±0.06, respectively. There were neither significant differences between the two groups nor between the hemispheres. Since there was a significant positive correlation (R = 0.6) between the age and the mI\Cr in the right frontal lobe of depressed patients, age matched pairs analysis was performed (n = 2×10, in each group: nine females, one male, < 40 years). In the right frontal lobe, the patients mI\Cr of 0.40±0.05 was now significantly lower than the controls mI\Cr of 0.45±0.06. However, most of the patients were on antidepressive medication. Interestingly, it was exactly this group of patients which showed significantly lower mI levels.

We regard our investigation as a pilot study which suggests an influence of age and antidepressants on mI levels and should be taken into consideration in further investigations in depressive patients.

Introduction

Phosphatidylinositol (PI) is a major component of neuronal cell membranes. The phosphoinositide cycle (Fig. 1) has been discovered as a major second messenger system (Berridge and Irvine, 1989). Receptor stimulation by neurotransmitters activates a phospholipase C enzyme in a number of membrane receptor signalling pathways. Phospholipase C triggers the breakdown of phosphatidylinositol-bis-phosphate (PIP2) to inositol 1,4,5-trisphosphate (IP₃), which releases calcium from internal stores. A series of phosphatases remove the phosphate groups from IP₃ sequentially, releasing free inositol. Inositol is a 6-carbon cyclic polyol carbohydrate chemically closely related to glucose. The biologically active stereoisomer is called myo-inositol (mI). It is the theories about the action of lithium which have aroused the interest in the phosphoinositide (PI) cycle (Fig. 1) and its role in affective disorders. Lithium reduces the brain levels of inositol by inhibiting inositol monophosphatase (Hallcher and Sherman, 1980). Its therapeutic function in mania and depression might be a result of the mI depletion which slows down the PI cycle in overactive synaptic systems (Berridge and Irvine, 1989; Snyder, 1992; Kofman and Belmaker, 1993).

In cerebrospinal fluid, a markedly reduced level of mI has been reported in depressed patients with unipolar and bipolar affective disorder (Barkai et al., 1978). Myo-inositol levels in the human brain were analysed by gas-liquid chromatography post mortem (Shimon et al., 1997). These data revealed significantly reduced mI levels in the frontal cortex of bipolar affective disorder patients (n = 8) and suicide victims (n = 10) as compared with controls (n = 10). Inositol levels were 8.7±2.4 mmol\kg in the normal sample, but only 5.9±2.1 and 6.7±2.2 mmol\kg in bipolar disorder and suicides, respectively, i.e., reduced by 20–30%. There were similar but statistically non-significant intergroup differences in the occipital cortex. The authors found no correlation between age and cerebral inositol levels and no significant difference between inositol levels in males and females.

High doses of exogenously administered inositol may pass the blood–brain barrier sufficiently (Spector, 1988; Levine et al., 1993). Since mI has been widely appreciated as a key metabolic precursor of a second messenger system for numerous neurotransmitters, controlled trials of inositol treatment have been published (review paper: Levine, 1997). Under double-blind conditions, the intake of inositol 12 g\day led to an improvement in depression, which was significantly superior to placebo at week 4 (Levine et al., 1995).

Magnetic resonance spectroscopy (MRS) offers unique possibilities for non-invasive studies of biochemistry in the human brain in vivo. An atomic nucleus with a spin distinguished from zero has properties like a little magnetic dipole. In a strong external magnetic field, due to the quantum nature of matter, each magnetic dipole or spin aligns parallel or antiparallel to this field. As a result the spins are split into two distinct groups with different energy levels and slightly different population densities (Nuclear Zeeman effect). Transitions between these energy levels may occur by absorption or emission of electromagnetic radiation, in particular at the so called Larmor or resonance frequency. As the external magnetic field is shielded by the electron configuration of each molecule, the differences in energy, or frequency absorbed, depend not only on the external magnetic field strength and the particular species of nucleus (e.g., ¹H or ³¹P) but also on the specific chemical compound. Protons within different molecular structures absorb and reemit energy in the radio frequency range at slightly different frequencies because of small differences in the local magnetic field induced by chemical binding. Thus, proton (¹H)–MRS allows to detect biochemical compounds due to substance specific differences in resonance frequency. This frequency shift is referred to as chemical shift. A graph of the magnitude of energy absorption as a function of frequency is called MR spectrum (Fig. 2). It is well known, that the main resonance of mI appears at a chemical shift of 3.56 ppm (parts per million as a measure of normalized frequency) in human brain (Michaelis et al., 1991). Compounds must be present at concentrations of approximately 0.5–1.0 mmol\kg or greater to be visible with MR techniques used in human subjects in vivo. ¹H-MRS studies published so far indicate that the free (MRS visible) mI concentration ranges between 3.9–8.1 mmol\kg in adults (review paper: Henriksen, 1995). In contrast to mI, the isomer scyllo-inositol yields a resonance at 3.35 ppm. Scyllo-inositol concentrations of about 0.5 mmol\kg were identified in ¹H-MRS of human brain at 2.0 T (Michaelis et al., 1993).

Phosphorous-31-MRS showed abnormal phosphorous metabolism in bipolar disorder (Kato et al., 1991; Kato et al., 1992; Kato et al., 1993; Deicken et al., 1995) in the frontal lobe, presumably due to deviations in the phosphatidylinositol cycle. Moreover, the most consistent (but not specific) SPECT and PET findings in depressives have been regional deficits in the cerebral blood flow and metabolism of the frontal lobe (Hurwitz et al., 1990; Lesser et al., 1994; Kasper et al., 1994; Mozley et al., 1996). Thus, the left and right frontal lobes are the regions of interest in this investigation.

We hypothesized that reduced mI concentrations might be detected by means of ¹H-MRS in the frontal lobe of depressive patients when compared to healthy subjects.