Quantitation of a neutral‐buoyancy assay (NBA) to estimate transmembrane N2 flux

Abstract

Several assays exist for CO2, O2 and NH3 permeability, but a major technical void has been the absence of an assay for the transmembrane movement of N2 and other relatively inert gases. Here, we report the quantitation of a novel and simple yet powerful assay based on the neutral buoyancy of a Xenopus oocyte injected with a bubble of nitrogen (N2) gas. Briefly, we inject a precise volume of N2 (200 nl)—with a known number of gas molecules (nGas)—creating a bubble that thereby lowers cell density (ρCell) sufficiently for the oocyte to float. After injection, we transfer the oocyte to a glass tube (rated to 10 ATM), nearly filled with saline. We then pressurize the atmosphere above the saline solution sufficiently for the bubble inside the oocyte to constrict, thereby raising ρCell (decreasing buoyancy) and causing the oocyte to descend to a neutrally buoyant depth of 5 cm. As N2 diffuses out of the bubble, dissolves in the surrounding cytoplasm, and eventually exits the oocyte, the tendency is for the bubble to collapse gradually (increasing ρCell) and for the oocyte to sink. Rather than allowing the oocyte to sink substantially, we developed a neutral buoyancy clamp system (high‐resolution camera, software developed in‐house, digitally controlled pressure regulator) to adjust and record continuously the pressure in the air phase (PNB) required to maintain neutral buoyancy at an oocyte depth of 5 cm. The pressure inside the bubble (PBubble > PNB by a fixed amount) falls approximately exponentially over time, and proportionally with the decreasing nGas. In order to calculate nGas from the recorded PNB, we performed a series of experiments in which—for each oocyte—we measured the spontaneous initial oocyte volume before bubble injection (Vi,Xo), injected a known volume of N2 (VGas), and then measured initial PNB for that combination of Vi,Xo and VGas. We repeated this procedure for 295 oocytes with Vi,Xo ranging from 812 to 1590 nl, and VGas ranging from 50 to 350 nl. For VGas from 100 and 300 nl, we observed a linear dependence of PNB on VGas/Vi,Xo, with the best‐fit equation: PNB = 11.28*(VGas/Vi,Xo) – 0.05 (Eq.1), R2 = 0.95. Here, the units of PNB, the slope, and the intercept are ATA. Recalling the ideal gas law and inserting the expression for VGas from Eq.1, we conclude that the apparent nGas = PB*Vi,Xo*(PNB + 0.05)/(11.28*RT) (Eq.2). Here, PB is room barometric pressure, and all units are consistent with an R of 0.0821 L atm K–1 mol–1. Because of unavoidable experimental error in Vi,Xo and VGas, and variability in oocyte density, the apparent nGas at time = 0 does not always match 8.26 nmol, the value computed for 200 nl of injected gas in our standard assays. We therefore rearranged Eq.2 to replace the best‐fit y‐intercept (i.e., 0.05 ATA) with an adjusted y‐intercept (b) that is characteristic for each oocyte. Using the revised relationship, nGas= PB*Vi,Xo*(PNB + b)/(11.28*RT) (Eq.3), we are now able to obtain the actual decay in nGas during our standard NBA and estimate absolute transmembrane N2 fluxes. While specifically developed for N2 fluxes, the NBA should be applicable to measuring transmembrane fluxes of other gasses.Support or Funding InformationOffice of Naval Research (N00014‐15‐1‐2060 & N00014‐16‐1‐2535)