The solution to bicarbonate

to the editor: A foundation of ex vivo experimental studies is the use of appropriately composed physiological salt solutions. Tight control of the chemical environment is one key strength of well-performed ex vivo investigations allowing for dissection of specific cellular contributions and molecular mechanisms under simulated physiological and pathophysiological conditions.

Experiments relying on rigorous evidence-based designs rather than variable laboratory traditions should improve reproducibility. In a recent issue of the American Journal of Physiology-Heart and Circulatory Physiology, we therefore commend the recent efforts by Wenceslau et al. (1) to provide scientific guidelines, including composition of salt solutions, for studies of isolated blood vessels. We particularly appreciate their argument that the composition of traditional salt solutions should be adjusted to reflect physiological in vivo levels, for instance, regarding ionized Ca²⁺ and glucose, which have classically been present in supraphysiological concentrations.

In line with this, we also recommend that the traditional supraphysiological levels of 5.4–6 mM K⁺ suggested by Wenceslau et al. (1) be reduced to a physiological level around 4 mM. This is particularly relevant for studies evaluating extracellular K⁺ accumulation as an activity-related signal for vasodilation.

Considering the intent of the guideline authors to reproduce in vivo concentrations of circulating ions, we were surprised by their suggestion to use 14.9 or 19 mM ${\text{HCO}}_3^{-}$, which is well below typical plasma concentrations in humans and rodents. We disagree with the authors that [${\text{HCO}}_3^{-}$] can be lowered without influencing contractile responses (1).

We would like to raise two important points in particular:

First, if the proposed salt solution is aerated with a gas mixture containing 5% CO₂—as suggested in the guidelines (1)—according to the Henderson–Hasselbalch equation, 14.9 mM ${\text{HCO}}_3^{-}$ produces a pH of 7.24. This level of acidity will affect tone in many vascular beds. If, alternatively, the salt solution at 5% CO₂ is adjusted with NaOH to pH 7.40—as described in the original paper cited for the absence of ${\text{HCO}}_3^{-}$-dependent vascular effects (2)—equilibration of the CO₂/${\text{HCO}}_3^{-}$ buffer system during titration will inevitably generate ${\text{HCO}}_3^{-}$ until, at pH 7.40, the solution contains 22 mM ${\text{HCO}}_3^{-}$, as predicted by the Henderson–Hasselbalch equation. The 5% CO₂ gas mixtures used by almost everyone and readily commercially available correspond in a humidified atmosphere to roughly 36 mmHg. If a more physiological gas mixture with 40 mmHg CO₂ is used, the corresponding [${\text{HCO}}_3^{-}$] is 24 mM at pH 7.40. In any case, the conclusion that low [${\text{HCO}}_3^{-}$] does not influence vascular tone is erroneous, because the final pH-adjusted salt solutions used for the cited control and “reduced ${\text{HCO}}_3^{-}$” experiments (2) in reality all contained 22 mM ${\text{HCO}}_3^{-}$. It is obviously no surprise that identical solutions resulted in identical vascular tone development.

Second, we strongly recommend using a physiological [${\text{HCO}}_3^{-}$] of 22–24 mM because ${\text{HCO}}_3^{-}$ plays at least three important roles that are concentration-dependent: it constitutes with CO₂ a mobile physiological buffer system that stabilizes pH and minimizes spatial pH gradients (3–5), it serves as a substrate for Na⁺, ${\text{HCO}}_3^{-}$ cotransporters, and Cl^–/${\text{HCO}}_3^{-}$ exchangers (6–11), and it can be sensed in both the intracellular (12) and surrounding extracellular (13, 14) space of endothelial cells to modify arterial function and structure.

Multiple studies show that H⁺-sensitive targets are important in most vascular beds, and more recent investigations reveal that ${\text{HCO}}_3^{-}$ also plays a direct signaling role in the control of vascular tone (15). ${\text{HCO}}_3^{-}$ transporters are critically important for setting the intracellular pH of vascular smooth muscle and endothelial cells (6–9), and omission of CO₂/${\text{HCO}}_3^{-}$ can modify, for instance, vascular smooth muscle Ca²⁺ sensitivity (7, 8), endothelium-dependent hyperpolarizations (6), and NO production (8). Intracellular ${\text{HCO}}_3^{-}$ sensing has been ascribed to the soluble adenylyl cyclase (12), whereas extracellular sensing requires the receptor protein tyrosine phosphatase (RPTP)γ (13, 14).

These abovementioned considerations are important for all types of studies on isolated blood vessels. Specifically, when evaluating physiological and pathophysiological roles of ${\text{HCO}}_3^{-}$, two experimental approaches have been employed:

First, due to the chemical equilibrium between CO₂, ${\text{HCO}}_3^{-}$, and H⁺, it is not possible under standard experimental conditions to vary only one of the parameters Pco₂, [${\text{HCO}}_3^{-}$], and pH. However, because of the relatively slow spontaneous hydration of CO₂—requiring several hundred milliseconds for equilibration—separate control of Pco₂, [${\text{HCO}}_3^{-}$], and pH can be achieved using out-of-equilibrium CO₂/${\text{HCO}}_3^{-}$ solutions based on rapid mixing and laminar propulsion of solutions with dissimilar CO₂/${\text{HCO}}_3^{-}$/H⁺ composition (16). This technique applied to cerebral and mesenteric arteries from rats and mice has provided new insights into the physiology of ${\text{HCO}}_3^{-}$-dependent proteins and their vascular consequences (5, 13).

Second, as a simpler approach, investigators have been comparing vascular responses in the presence and absence of CO₂/${\text{HCO}}_3^{-}$. To ensure appropriate buffering capacity, it is common in these cases to replace NaHCO₃ with an artificial buffer such as HEPES. In these types of experiments, we suggest adding a similar (osmotically compensated) concentration of HEPES to the CO₂/${\text{HCO}}_3^{-}$ containing solutions to distinguish the influence of removing CO₂/${\text{HCO}}_3^{-}$ from the potential effect of adding HEPES.

In conclusion, we strongly recommend physiological concentrations of 22–24 mM ${\text{HCO}}_3^{-}$ in experimental salt solutions for ex vivo tissue studies. In combination with a gas mixture containing 5% or (better) 40 mmHg CO₂, this establishes a pH of 7.40, which is appropriate for studies on arteries under physiological conditions. Obviously, other combinations of Pco₂, [${\text{HCO}}_3^{-}$], and pH can be relevant if evaluating veins, disease circumstances, or blood vessels from other animal species (e.g., marine organisms).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

E.B. drafted manuscript; E.B. and C.A. edited and revised manuscript; E.B. and C.A. approved final version of manuscript.