Stable potassium isotopes (41K/39K) track transcellular and paracellular potassium transport in biological systems

Abstract

As the most abundant cation in archaeal, bacterial, and eukaryotic cells, potassium (K⁺) is an essential element for life. While much is known about the machinery of transcellular and paracellular K transport-channels, pumps, co-transporters, and tight-junction proteins-many quantitative aspects of K homeostasis in biological systems remain poorly constrained. Here we present measurements of the stable isotope ratios of potassium (⁴¹K/³⁹K) in three biological systems (algae, fish, and mammals). When considered in the context of our current understanding of plausible mechanisms of K isotope fractionation and K⁺ transport in these biological systems, our results provide evidence that the fractionation of K isotopes depends on transport pathway and transmembrane transport machinery. Specifically, we find that passive transport of K⁺ down its electrochemical potential through channels and pores in tight-junctions at favors ³⁹K, a result which we attribute to a kinetic isotope effect associated with dehydration and/or size selectivity at the channel/pore entrance. In contrast, we find that transport of K⁺ against its electrochemical gradient via pumps and co-transporters is associated with less/no isotopic fractionation, a result that we attribute to small equilibrium isotope effects that are expressed in pumps/co-transporters due to their slower turnover rate and the relatively long residence time of K⁺ in the ion pocket. These results indicate that stable K isotopes may be able to provide quantitative constraints on transporter-specific K⁺ fluxes (e.g., the fraction of K efflux from a tissue by channels vs. co-transporters) and how these fluxes change in different physiological states. In addition, precise determination of K isotope effects associated with K⁺ transport via channels, pumps, and co-transporters may provide unique constraints on the mechanisms of K transport that could be tested with steered molecular dynamic simulations.

Keywords: Na-K ATPase; homeostasis; physiology; potassium; potassium channel; stable isotope.

FIGURE 1

The difference in δ⁴¹K values between external and intracellular K⁺ in growth experiments of C. reinhardtii and a diagram of K⁺ transport in single­celled algae grown under optimal conditions (See Methods) including major fluxes (F_in), transport machinery, and K isotope effects (ε_in). p = 4.5 × 10⁻⁵ by one-way ANOVA in Matlab.

FIGURE 2

Whole-body K isotope mass balance in R. norvegicus on a controlled diet. Measured δ⁴¹K values of urine (p = 0.19) and feces (p = 0.037) relative to diet indicate a preferential uptake of ³⁹K in the gut. The total range in δ⁴¹K_diet values in the tissues of R. norvegicus is ∼1‰ with an estimated whole-body average δ⁴¹K_diet value of approx. +0.5‰, consistent with preferential loss of ³⁹K in urine.

FIGURE 3

(A) K isotopic composition of tissues/fluids R. norvegicus (casein diet) normalized to plasma (δ⁴¹K_plasma = 0‰). δ⁴¹K_plasma values that are positive indicate that net K⁺ transport out of the tissue/cell/fluid is enriched in ³⁹K relative to K⁺ transport into the tissue/cell/fluid. δ⁴¹K_plasma values that are indistinguishable from 0 indicate that the K isotope fractionation associated with K⁺ transport into and out of the cell/tissue/fluid are either 0 or equal in magnitude (and thus cancel). δ⁴¹K_plasma values that are negative indicate that net K⁺ transport into the cell/tissue/fluid is enriched in ³⁹K relative to K⁺ transport out of the cell/tissue/fluid. (B) A diagram of K isotope mass balance including transport by (1) K-channels, (2) K-pumps/co-transporters, (3) through pores in tight-junctions, and (4) glomerular filtration. Eq. 3 solves for the δ⁴¹K value of a tissue/cell relative to blood plasma assuming steady-state K isotope mass balance with bi-directional K⁺ transport by both channels and pumps/co-transporters. Variables include f^out/in _ch , the fraction of total K⁺ transport into/out of the tissue/cell that occurs through K⁺ channels, Δf_ch , the difference in the fraction of K transport via channels (f^out _ch -fⁱⁿ _ch), ε_ch , K isotope fractionation associated with K channels, and ε_p/co K isotope fractionation associated with pumps/co-transporters. K isotope fractionation associated with K⁺ uptake in the gut (ε_gut) and K⁺ loss in urine (ε_kidney) reflect the flux-weighted average of both paracellular and transcelluar K⁺ transport. For example, K isotope fractionation in the kidney will be associated with both re-absorption and secretion of K⁺ in the proximal tubule, the thick ascending limb (TAL), and cells in the collecting ducts of the kidney. (C) Correlation between intercellular [K⁺] and δ⁴¹K_plasma values for a subset of tissues from R. norvegicus. Also shown are the measured activities of NaKATPase (µmol P h⁻¹ g⁻¹ from Gick et al., (1993) and values of Δf_ch from Eq. 3.

FIGURE 4

K isotopic composition of cerebrospinal fluid from R. norvegicus raised on a controlled diet (vivarium chow) normalized to blood plasma (δ⁴¹K_plasma = 0‰) and brain tissues normalized to cerebrospinal fluid (δ⁴¹K_CSF = 0‰) together with a diagram of K homeostasis across the choroid plexus and blood brain barrier (BBB) from Hladky and Barrand (2016) including transport by (1) K-channels, (2) K-pumps/co-transporters, and (3) transcellularly through pores in tight-junctions. δ⁴¹K_CSF values that are positive indicate that net K transport out of brain tissues (spinal cord, cerebellum, and cerebrum) and into CSF is enriched in ³⁹K relative to K transport from CSF into these tissues. The negative δ⁴¹K_plasma values for CSF indicate that net K transport into CSF from blood plasma is enriched in ³⁹K relative to K transport from CSF to blood plasma and reflects the flux­ weighted average of the transporters that dominate K⁺ exchange across both the choroid plexus and BBB.

FIGURE 5

(A) The difference in δ⁴¹K values between seawater and dorsal white muscle of various stenohaline and euryhaline marine fish. Muscle [K⁺] and δ⁴¹K_seawater values are listed in Supplementary Table S1. p = 1.4 × 10⁻¹⁰ for the difference in δ⁴¹K_seawater value between stenohaline and euryhaline teleosts. (B) A model of K isotope mass balance in a marine fish. K⁺ is supplied across the gill epithelium via transport through pores in tight-junction proteins and lost from mitochondria-rich cells (MRCs) through channels Eq. 5. K⁺ cycling within MRCs is modeled using Eq. 3. Combining Eq. 3 and Eq. 5a yields Eq. 6, the δ⁴¹K value of plasma/muscle relative to seawater as a funciton of ε_ch, ε_gill, Δf_ch of MRC’s and ε_p/co. As Δf_ch → 0, δ⁴¹K of fish muscle → ε_ch - ε_gill (≥ 0‰, euryhaline), whereas as Δf_ch→ 0, δ⁴¹K of fish muscle → ε_p/co-ε_{gill (<}0‰, stenohaline). fish illustrations from Lori Moore.