the plasma membrane of a cell provides a physical barrier to the free movement of water, ions, and organic molecules. It also hampers the diffusion of dissolved gases, including oxygen (O2), ammonia (NH3), and carbon dioxide (CO2). For each molecular species that passively crosses the plasma membrane, there are several key factors that affect its movement: 1) the permeability coefficient of the membrane (with its complement of lipids and proteins) to the particular molecule, 2) the concentration gradient of the molecule between the two compartments, and 3) an additional barrier created by unstirred layers of solution adjacent to the membrane, which reduces diffusion from the bulk solution to the plasma membrane. Reduced diffusion through unstirred layers often prevents a direct measurement of the true membrane permeability. In fact, because of their large size, which usually exceeds the membrane thickness by several orders of magnitude, unstirred layers may make an appreciable contribution to the overall resistance to the flux of gases (4). Note that this is in contrast to ions and other solutes (for which the membrane is often the only significant resistance), because the membrane permeability to them is much smaller than that of CO2, for example.
Gutknecht and Tosteson (3) were the first to demonstrate that, by using a chemical reaction within the unstirred layers, it was possible to measure the true membrane permeability of an artificial bilayer and that pH buffers facilitated the movement of weak acids across lipid bilayers. While salicylic acid was used in their original experiments, they indicated that the diffusion of other weak acids, such as CO2, might also be affected by chemical reactions in the unstirred layers, a proposition that they later verified (2). Indeed, using artificial planar lipid bilayers, they demonstrated that large unstirred layers were rate-limiting for CO2 movement, unless chemical reactions were introduced by adding carbonic anhydrase (CA) (2). Burg later extended these experiments to short segments of rabbit proximal convoluted tubules (10).
In a series of three companion papers published in this issue of the American Journal of Physiology-Cell Physiology, the group of Walter Boron examined how the inward flux of CO2 is affected by the presence of intracellular and extracellular CA or the presence of a pH buffer. By simultaneously using two pH-sensitive microelectrodes, one inserted inside the cell and the other at the cell surface, and the use of a system that allows rapid exchange of the external solution, they introduced CO2 outside of a Xenopus laevis oocyte and meticulously measured the pH response on both sides of the membrane. The authors varied the experimental paradigm many different ways, collected a large number of pH traces, and used simulations to refine a sophisticated mathematical model.
The concentration of a gas such as CO2 is affected by its conversion along with water to bicarbonate and protons (and vice versa): CO2 + H2O ↔ HCO3− + H+, a reaction greatly facilitated by the presence of CA (Fig. 1). Note that the oocyte is a preferred model for such analyses as its HCO3− permeability is extremely low and therefore, its movement can be ignored.
In the first paper, the authors examined the effect of microinjecting oocytes with soluble CA II. They then applied 1.5% CO2/10 mM HCO3− and followed both surface and intracellular pH changes (5). Cytosolic CA II sped up the conversion of incoming CO2 to HCO3− and H+, thereby maintaining a low intracellular [CO2], and thus promoting CO2 influx. A key point in this paper is the difficulty of using intracellular pH changes to reach intuitive conclusions about the effect of intracellular CA II because the pH measurement and enzyme are “cis” to each other with respect to the cell membrane (i.e., both being cytosolic). In other words, cytosolic CA II accelerates cytosolic pH changes, regardless of effects on CO2 influx. However, because the surface-pH electrode is “trans” to the cytosolic CA II, enhancements of surface-pH changes provide direct evidence that cytosolic CA II accelerates CO2 influx. This effect was blocked by preincubating oocytes with cell-permeant ethoxzolamide. As anticipated for a response mostly determined by gradients and driving force, increasing the external CO2 concentration by rapid addition of 5% and 10% CO2, instead of 1.5%, significantly increased its influx. This was evidenced by greater rates of intracellular-pH change and greater amplitudes of surface-pH changes.
In the second paper, the authors examined the effect of expressing CA IV at the cell surface by targeting the enzyme to the extracellular leaflet of the membrane through linkage with glycosylphosphatidylinositol; CA IV accelerated CO2 influx (6). This effect was blocked by extracellular acetazolamide and substantially slowed by acetazolamide injected into the cytosol, implying that CA IV, as heterologously expressed in oocytes, is present on both the extracellular surface and intracellularly. This result confirms earlier work (7, 9). Because the surface electrode is “cis” to the surface CA IV, and the intracellular electrode is “cis” to the intracellular CA IV, how can one reach intuitive conclusions about the effect of the enzyme? A key point is that the authors found that CA IV accelerated both the decay of the surface-pH-spike and the exponential decrease of the CO2-induced intracellular-pH decline. In other words, because CA IV causes CO2 to equilibrate across the cell membrane over a shorter period of time, the CO2 flux must be greater in the presence of CA IV. This increased CO2 influx is explained by the extracellular CA replenishing CO2 as it diffuses across the plasma membrane into the oocyte, thereby enhancing the gradient or driving force of the gas. They also demonstrated, as originally shown by Gutknecht and Tosteson (3), that, in the presence of CA, addition of an organic pH buffer, such as HEPES, also accelerates CO2 Influx. Furthermore, as first shown by Gutknecht and Tosteson, addition of CA II in the bulk extracellular solution also sped up CO2 influx of both water- and CA IV-injected oocytes (6). This can also be easily explained: addition of a buffer constitutes a novel source of H+, which in the presence of CA drives the reaction towards CO2 production. In addition to extending the earlier work (3) to a living cell, the authors greatly expand the matrix of experimental conditions and make the new observation that, even in the absence of added CA, HEPES can increase the CO2 flux, provided that [CO2] is sufficiently high.
This brief description does not properly credit the vast amount of work and new insights in the two experimental papers. For example, the authors also are the first investigators to provide an explanation for why the addition of CO2 to a cell and CO2 removal from the same cell produce asymmetrical effects on pH transients. They also address the effects of microelectrode placement and provide critical controls by their simultaneous assessment of amplitudes, rates, and time delays of pH changes on both sides of the membrane. Even an oocyte is such a complex system that intuition is of limited value in providing insight into the experimental results. A particular strength of this series of papers is the use in the third paper of a sophisticated mathematical model to interpret the wealth of experimental data and to help provide insights (8) and adapts their previously published, three-dimensional reaction-diffusion model (11).
This new model adds a number of features, which includes a reduced oocyte water content [estimated to be around 40% of oocyte volume based on swelling behavior of oocytes (12) and is in good agreement with the water content of oocytes we recently measured using the traditional wet weight/dry weight technique (1)]; a barrier to diffusion inside the cytosol due to viscosity and a layer of vesicles; native expression of CA in both the cytosol and at the membrane; reduced permeability of the oocyte plasma membrane to CO2; and the presence of microvilli and a vitelline membrane. By setting all these parameters in the proposed three-dimensional model, simulations could be run and compared with experimental data. With few exceptions, the model reproduces the data very well. Moreover, the model predicts the novel concept that the effects of simultaneously providing intracellular and extracellular-surface CA are not just additive but supra-additive.
The statistician George E. P. Box has noted that “all models are wrong, but some are useful.” Boron and his colleagues point out that their current model, although useful, does not take into account the special environment between the surface-pH electrode and the cell membrane, nor the asymmetrical [CO2] in the extracellular fluid flowing around the oocyte in a chamber. This special environment, they predict, amplifies surface-pH changes by accentuating reaction over diffusion, whereas the relatively low extracellular [CO2] near the surface of the oocyte that is downstream from the direction of flow will slow the equilibration of CO2 throughout the system. Future work on the oocyte model will have to include these missing features with more sophisticated mathematical approaches. One can imagine extending the physiological experiments to mammalian cells and tissues, and increasing the scale of the models to include multicellular systems such as kidney tubules or pancreatic ducts.
In summary, the studies presented in the three papers provide important new insights into how CA II and CA IV expression in the cytosol and at the membrane (outside) enhances transmembrane CO2 fluxes by maximizing CO2 gradients across the plasma membrane.
No conflicts of interest, financial or otherwise, are declared by the author(s).
E.D. prepared figure; drafted manuscript; edited and revised manuscript; approved final version of manuscript.
- Copyright © 2014 the American Physiological Society