This item is under embargo and not available online per the author's request. For access information, please visit

Date of Award

Winter 12-15-2014

Author's School

Graduate School of Arts and Sciences

Author's Department


Degree Name

Doctor of Philosophy (PhD)

Degree Type



Magnetic resonance imaging (MRI) is one of the most powerful tools for the detection and diagnosis of brain injuries. Several types of brain injuries can be detected using diffusion-sensitive MRI because the apparent diffusion coefficient (ADC) of brain tissue water decreases rapidly (within minutes) and dramatically (30-50%) in the injured region. Remarkably, the biophysical mechanism(s) underlying this phenomenon remains poorly understood.

In developing a better understanding of the biophysical determinants that underlie this phenomenon, it would be useful to separately characterize the ADC in both intra- and extracellular spaces. However, water molecules can transit across the cell membrane separating these spaces (transcytolemmal water exchange). It follows that knowledge of the average residence time for water molecules within intracellular compartment -- referred to herein as the intracellular water preexchange lifetime (τin) -- is required if one is to design experiments that monitor compartment-specific parameters such as the ADC. For example, in the simplest case, ADC in one (intra- or extracellular) compartment should be determined in a measurement time that is short relative to τin, i.e., before transcytolemmal exchange averaging can occur.

Herein, τin was measured for microbead-adherent neurons and astrocytes using magnetic resonance (MR) methods. Slice-selective inversion-recovery spin-echo spectroscopy was performed on newborn rat neural cell cultures enriched for neurons and astrocytes, respectively. The rapid and turbulent flow of oxygenated media perfusing (flowing) about the microspheres, which is necessary to maintain cell viability, allowed the MR 1H signal from extracellular water (the perfusate) to be largely suppressed by the time-of-flight effect of slice-selective spin-echo detection. Because of rapid flow through the thin slice(s), the extracellular water, whose polarization had been inverted by the slice-selective inversion pulse, is rapidly replaced (in 10-50 ms) by fully equilibrium-state polarized extracellular water. Under these circumstances, the normally inefficient longitudinal relaxation of intracellular water is dominated by physical water exchange (transcytolemmal) with the fully equilibrium-state polarized extracellular water, allowing τin to be determined.

Under normal culture conditions, τin was determined to be 0.7 ± 0.1 s for neurons (n = 12) and 0.5 ± 0.1 s for astrocytes (n = 21). This infers relatively low cell-membrane water apparent permeability among neural cells and implies that the typical clinical diffusion-sensitive MR protocol, with diffusion times of ca. 50-100 ms, is in the slow-exchange regime for intracellular water. Within 45-min under combined oxygen-glucose deprivation, a widely-used model of hypoxia-ischemia in cultured cells, τin decreased to 0.53 ± 0.08 s for neurons (n = 3) and 0.20 ± 0.06 s for astrocytes (n = 5). The decrease in τin implies a significant increase in the cell membrane water apparent permeability upon injury and, thus, may shift clinical diffusion-sensitive MR experiments into an intermediate- or even fast-exchange regime.

This is the first report of a near-direct MR measurement of τin in central nervous system cells. The results of this dissertation provide new insights into the development and interpretation of diffusion MR studies in the brain.


English (en)

Chair and Committee

Joseph J.H. Ackerman

Committee Members

Joel R Garbow, James E Huettner, Jacob Schaefer, Dewey Holten


Permanent URL:

Available for download on Saturday, December 15, 2114

Included in

Chemistry Commons