Characterization of Carbonic Anhydrase In Vivo Using Magnetic Resonance Spectroscopy

Abstract

Carbonic anhydrase is a ubiquitous metalloenzyme that catalyzes the reversible interconversion of CO₂/HCO₃^-. Equilibrium of these species is maintained by the action of carbonic anhydrase. Recent advances in magnetic resonance spectroscopy have allowed, for the first time, in vivo characterization of carbonic anhydrase in the human brain. In this article, we review the theories and techniques of in vivo ¹³C magnetization (saturation) transfer magnetic resonance spectroscopy as they are applied to measuring the rate of exchange between CO₂ and HCO₃^- catalyzed by carbonic anhydrase. Inhibitors of carbonic anhydrase have a wide range of therapeutic applications. Role of carbonic anhydrases and their inhibitors in many diseases are also reviewed to illustrate future applications of in vivo carbonic anhydrase assessment by magnetic resonance spectroscopy.

Keywords: GABAergic transmission; carbonic anhydrase; in vivo MRS; neurological diseases; psychiatric diseases.

Figure 1

The two-site exchange diagram for CO₂ ↔ HCO₃⁻ catalyzed by carbonic anhydrase. M_0A and M_0B denote the magnetization of CO₂ and HCO₃⁻ at thermal equilibrium. T_1A, T_1B and T_2A, T_2B are their respective longitudinal and transverse relaxation times without any chemical exchange. k_AB and k_BA represent the pseudo-first-order rate constants of the unidirectional CO₂ → HCO₃⁻ hydration and HCO₃⁻ → CO₂ dehydration reactions, respectively.

Figure 2

Radiofrequency pulse sequence for the ¹³C saturation transfer experiments. ¹H→¹³C heteronuclear Nuclear Overhauser Enhancement (NOE) was generated by saturating proton signals using evenly spaced non-selective hard pulses. A continuous wave (CW) ¹³C pulse or a train of spectrally selective shaped ¹³C pulses was used for radiofrequency saturation at CO₂ resonance or at the control frequency on the opposite side of bicarbonate. For excitation, a ¹³C block pulse was used. Δ: Delay between proton pulses (48 ms).

Figure 3

A typical time-course of control spectra acquired from a single subject after oral administration of [U-¹³C₆] glucose without proton decoupling. Each spectrum was acquired with recycle delay = 30 s, spectral width = 8 kHz, number of data points = 2048, number of averages = 12, and line broadening = 8 Hz. The time interval indicates the beginning and end of acquisition following oral glucose intake. Lipid: carboxylic carbons of natural abundance lipids (172.5 ppm), Glu5: glutamate C5 (182.0 ppm), Glu1: glutamate C1 (175.4 ppm), Gln5: glutamine C5 (178.5 ppm), Gln1: glutamine C1 (174.8 ppm), Asp4: aspartate C4 (178.3 ppm), Asp1: aspartate C1 (175.0 ppm) (reprinted from ref. [43]. https://creativecommons.org/licenses/by/4.0/).

Figure 4

Bicarbonate signal intensities as a function of time after oral administration of [U-¹³C₆] glucose. The glucose level was different in different subjects during the scan time. Bicarbonate signal increased monotonically (reprinted from ref. [43]. https://creativecommons.org/licenses/by/4.0/).

Figure 5

¹³C saturation transfer effect catalyzed by carbonic anhydrase (CA) in the human brain. Spectra were measured from a single subject between 118 and 130 minutes after oral administration of 20% [U-¹³C₆] glucose. (a) control spectrum with ¹³C irradiation at 228 ppm; (b) with saturation of carbon dioxide at 125.0 ppm; (c) difference spectrum.

Figure 6

In vivo ¹³C magnetization transfer effect of carbon dioxide–bicarbonate exchange in rat brain before (left) and after (right) carbonic anhydrase inhibition. Upper traces: no saturation of carbon dioxide. Middle traces: with saturation of carbon dioxide at 125.0 ppm. Lower traces: difference spectra. The ¹³C magnetization transfer effect of the carbon dioxide–bicarbonate exchange was significantly reduced after blockade of carbonic anhydrase (adapted from ref [40] with permission from John Wiley and Sons Ltd.).

Figure 7

Pseudo first-order unidirectional bicarbonate dehydration rate constant k_BA determined from healthy human subjects, control rats, rats treated with acetazolamide, and a phantom (standard deviation is plotted as error bars).