The development of acidosis during intense exercise has traditionally been explained by the increased production of lactic acid, causing the release of a proton and the formation of the acid salt sodium lactate. On the basis of this explanation, if the rate of lactate production is high enough, the cellular proton buffering capacity can be exceeded, resulting in a decrease in cellular pH. These biochemical events have been termed lactic acidosis. The lactic acidosis of exercise has been a classic explanation of the biochemistry of acidosis for more than 80 years. This belief has led to the interpretation that lactate production causes acidosis and, in turn, that increased lactate production is one of the several causes of muscle fatigue during intense exercise. This review presents clear evidence that there is no biochemical support for lactate production causing acidosis. Lactate production retards, not causes, acidosis. Similarly, there is a wealth of research evidence to show that acidosis is caused by reactions other than lactate production. Every time ATP is broken down to ADP and Pi, a proton is released. When the ATP demand of muscle contraction is met by mitochondrial respiration, there is no proton accumulation in the cell, as protons are used by the mitochondria for oxidative phosphorylation and to maintain the proton gradient in the intermembranous space. It is only when the exercise intensity increases beyond steady state that there is a need for greater reliance on ATP regeneration from glycolysis and the phosphagen system. The ATP that is supplied from these nonmitochondrial sources and is eventually used to fuel muscle contraction increases proton release and causes the acidosis of intense exercise. Lactate production increases under these cellular conditions to prevent pyruvate accumulation and supply the NAD+ needed for phase 2 of glycolysis. Thus increased lactate production coincides with cellular acidosis and remains a good indirect marker for cell metabolic conditions that induce metabolic acidosis. If muscle did not produce lactate, acidosis and muscle fatigue would occur more quickly and exercise performance would be severely impaired.
The following is the abstract of the article discussed in the subsequent letter:
Explaining pH Change in Exercising Muscle: Lactic acid, Proton Consumption, and Buffering vs. Strong Ion Difference
The interpretation of acid-base homeostasis has been debated for years. In a recent review, Robergs et al. (15) argue for a view of pH change in contracting skeletal muscle opposed to what we would call the traditional approach (1, 2, 7, 8, 10, 18). We thank them for revisiting their analysis in the light of our critical comments ( 3, 6), but their reply (16) misunderstands or ignores many of our criticisms. Rather than repeat these, we wish to point out how terminological ambiguity has confused the issues, after briefly reinforcing our argument by showing its quantitative equivalence to the alternative “physicochemical” approach advocated by Lindinger et al. (13) in their Editorial Focus.
“Traditional” and Physicochemical Approaches Agree
The physicochemical approach is based on indisputable principles of electroneutrality and ionization equilibrium (13), but the debate is about how to interpret practicable data. In exercising muscle, it has been applied (5, 9, 12, 13) by incorporating the increase in lactate concentration (Δ[La]) and, where measured, the fall in phosphocreatine concentration (Δ[PCr]) into the change in strong ion difference (ΔSID), and letting an increase in the concentration and average dissociation constant of nonbicarbonate weak acid account for the near-stoichiometric increase in inorganic phosphate concentration (Δ[Pi]) which results from the action of creatine kinase to buffer ATP (4) against temporary imbalance of supply and demand (Eqs. 4c and 6b in Ref. 6). However, by not taking explicit account of Pi, this obscures the fact that the predicted pH changes are identical to those in the traditional approach. To see this, consider ischemic exercise,1 where three physicochemical changes influence cell pH: 1) a change in strong ions ΔSID = 2Δ[PCr] − Δ[La], where the factor 2 is because PCr has charge ≈ −2 (for generality called g in Ref. 6); 2) a change Δ[A−] (negative, if pH rises) in the charge on weak acid ions, such as proteins, bicarbonate, some organic phosphates, and preexisting Pi; 3) a separate negative weak-ion contribution by the new Pi formed, equal to bΔ[Pi], where b is its pH-dependent negative charge (6). The physicochemical approach calculates absolute pH (13), but to compare with the traditional approach, we must consider pH change. Considering the three changes together, the electroneutrality condition (Eq. 3 of Ref. 13) can be written thus (1) in which all charges are given explicitly, and Kw is dissociation constant of water (13). In the traditional approach (6, 7), glycolysis with matching ATP hydrolysis generates one H+ per lactate, and the net H+ load (ΔN) to the cytosolic buffers is equal to the glycolytic H+ load (measured as Δ[La], although conceptually distinct from it) less the H+ “consumed” by PCr breakdown: ΔN = Δ[La] − (b-g)Δ[PCr], where (b-g) ≈ (b+2) is given in Table 1 of Ref. 6. Substituting in Eq. 1, neglecting the very small H+ terms and dividing by ΔpH, we obtain the traditional buffer capacity (β) (17) (2) which is just Eqs. 10 and 11 of Ref. 6. (Eq. 2 is correct for an approximately linear titration curve; otherwise, we must recast the expression in terms of differentials). The two approaches therefore agree during metabolic changes in vivo, as they do for simple buffer systems in vitro.
Lactate accumulation tends to lower SID, but the countervailing effect of PCr decrease, contrary to what is usually assumed (12, 13), is likely to be sufficient to raise SID. In the example used in Ref. 6, ΔSID is positive throughout, reaching ∼20 meq/l at the end of exercise. If it were not, it can be shown that [ADP] (which is a function of pH and [PCr], and therefore indirectly of Δ[La] and Δ[PCr]; Ref. 7) would remain at its resting value or even fall during exercise, which is never seen. Typical reported changes in e.g., K+ (9) are unlikely to outweigh this effect of PCr and make ΔSID negative in nonischemic exercise. Other things being equal, a positive ΔSID would decrease [H+] (12), and the reason the opposite can occur in exercise is the simultaneous increase in the weak acid Pi (12). Thus the distinction in the traditional approach between metabolic “H+ consumption” and “buffering” sensu stricto (6, 7) parallels a physicochemical distinction between the charge contributions of metabolically generated Pi (always negative, tending to raise [H+]) and of the existing cytosolic weak acids (whose negative contribution decreases when [H+] rises, tending to mitigate acidification).
A revealing special case is zero lactate production, e.g., in normal muscle at the start of exercise or in McArdle's disease (1, 2, 6–8), where the traditional approach sees pH increase due to net H+ consumption by PCr breakdown, moderated by cytosolic buffers (Eq. 8 in Ref. 6). In the physicochemical approach, the fall in [H+] is due to the positive ΔSID caused by PCr decrease, partly countered by the combined negative charge contributions of the increased Pi and of other weak acids (whose negative contribution increases when [H+] falls, tending to mitigate alkalinization). It is a striking, but unacknowledged, feature of the analysis of Robergs et al. (16) that their total H+ production (15) is positive in this case (Eq. 13 in Ref. 6), despite the increase in pH (3, 6).
Some of the Debate Is About Terminology
The Lohmann reaction.
Over a large range of ATP turnover, PCr decrease is matched by Pi increase (Eqs. 4c, 6b, and 7 in Ref. 6). This net process, which is not confined to “the initial reactions of repeated muscle contractions” (16), is sometimes called “the Lohmann reaction” (6, 7, 10), a term which Robergs et al. (15, 16) regard as inappropriate for a pair of reactions taken together. Because the biochemical facts are not in dispute (1, 3, 6, 7, 10, 18), the disagreement is terminological. To avoid argument, and because this name is ambiguous [being sometimes used for the creatine kinase reaction alone (Eq. 6a in Ref. 6)], we can dispense with it.
Robergs et al. (16) also object to this term. ATP synthesis and hydrolysis are clearly not coupled like e.g., NADH oxidation and pyruvate reduction, but they are kept very nearly equal by the creatine kinase equilibrium (4). The physicochemical approach ignores concentrations that do not change significantly, and the ionization state of ATP cancels out in the overall H+/lactate stoichiometry (6). Additionally, we can neglect ADP and AMP, because their concentrations are very low, making irrelevant Robergs et al.'s complaint (16) that we ignore adenylate kinase. We do not claim that “all ATP, ADP, and Pi from catabolism cycles 100% between muscle contraction and glycolysis” or “that all the ADP, Pi, and H+ from ATP hydrolysis are only used to reform ATP from glycolysis”, nor do we deny that there is “no equality or known [i.e., fixed] stoichiometry between ATP hydrolysis and each of glycolysis, the creatine kinase reaction, and the AMP deaminase reaction” (16). On the contrary, the analysis explicitly separates ATP hydrolysis into the two components balanced by ATP synthesis due to glycolysis and to creatine kinase (Eqs. 4, b and c in Ref. 6). Of course, in high-intensity exercise, ATP concentration can decrease (16), but this is due to loss of the whole adenine nucleotide pool via AMP deaminase (14), which complicates the calculations (19) but does not invalidate these principles.
“Lactic acid production.”
Lindinger et al. (13) point out that Robergs et al.'s (15, 16) neglect of the pH-dependence of stoichiometry has caused confusion. Their recent lengthy analysis (Eqs. 14 and 29 of Ref. 16) and their original exposition (Eqs. 5 and 6 of Ref. 15) both yield exactly the 1:1 lactate/H+ stoichiometry that underlies the traditional approach (3, 6, 12), and we cannot understand their stated reasons why they did not “focus on” or “emphasize” this (16). Full analysis (6) shows that this overall stoichiometry is independent of pH, although near resting pH the H+ derives mainly from ATP hydrolysis, of which Robergs et al. (15) make much, whereas at low pH, the altered ionization of ATP, ADP, and Pi (Eq. 5b in Ref. 6) makes glycogenolysis itself the predominant H+ source, which Robergs et al. (16) at different points deny and deem irrelevant. Nevertheless, because one H+ accompanies each lactate, it does not seem unreasonable to describe this as production of lactic acid. To see this, consider the hypothetical case of no PCr change. Here, Eq. 2 becomes Δ[La]/ΔpH ≈ β, which is conceptually equivalent to the addition of (literal) lactic acid to an aqueous buffer in vitro, the H+ load (measured as Δ[La]) being equal to the H+ buffered (βΔpH, or more exactly ∫βδpH). The physicochemical approach (13) describes this as the addition of lactate without accompanying strong cations, but we have seen that the predicted pH change is the same. The total H+ production of Robergs et al. (15), adding the H+ from ATP hydrolysis and pyruvate synthesis, would be 2Δ[La], whereas their total H+ consumption (15), adding “structural” H+ buffering to H+ consumption by pyruvate reduction, would be βΔpH+Δ[La]; thus Δ[La] appears, redundantly, on both the production and consumption sides of their analysis.2
“Buffering” of cell pH.
As we have seen, the physicochemical and traditional approaches predict the same pH change. In the traditional approach, each lactate is evidence of a metabolically generated H+ to be dealt with, and the relevant in vivo buffer capacity (β) is that measured in vitro by titration (1, 7, 17), not involving metabolism or ion fluxes. We have seen that this buffering is distinct from the effect of PCr breakdown to “consume” H+ (6, 7) and that the same distinction exists in the physicochemical approach. It also has metabolic-control relevance; the balance between lactate and PCr changes is at least partly due to the Pi-dependence of phosphorylase activity (7, 11), and it is directly related to the control of [ADP] (7). By contrast, the consumption of H+ by lactate formation from pyruvate (Eq. 2 in Ref. 6), on which Robergs et al. (15) place much emphasis, is stoichiometric and uninfluenced by metabolic control. There seems little point in calling it a metabolic contribution to “buffering” (15, 16). Adding it to all the other H+-consuming processes yields a “true buffer capacity” (16) with no nonmetabolic relationship to cytosolic buffer capacity measured in vitro (Eq. 15b in Ref. 6), no reason to expect invariance between exercise conditions and no value in predicting pH change. In short, it has no clear conceptual or operational meaning. Given the quantitative equivalence between traditional and physicochemical approaches, it is to be doubted that “a combination of a stoichiometric approach [i.e., that of Robergs et al.] fine tuned by the SID method” (16) is possible.
Robergs et al. do not address the argument that calculations of differences in H+ and lactate ion transport from muscle to blood are based on base excess changes, which are biased by exchange of bicarbonate with the interstitial fluid (3).
It is clearly not illegitimate to “take products of totally different reactions, couple those reactions together, and combine the products into new molecules” (16), as biochemistry is full of examples. If the objection is, instead, to the idea of combining reactions in a full accounting, this is, on the contrary, essential (13).
We see no point to the avowedly fictitious reactions in their final section or its provocative title (16). Their first illustration describes glycolysis to lactate accompanied by the obligatory H+, plus buffering by bicarbonate, and clearly does not imply “that glucose oxidation outside the mitochondrion produces CO2” (16). In their second illustration (16), the stoichiometry of gluconeogenesis from lactate is both incorrect and irrelevant.
The numerous quotations excavated from our earlier publications (16) simply show that we have for some while held the views we defend here. Readers will judge for themselves whether ad hominem comments (16) are appropriate to scholarly debate.
↵1 It is not “only when presenting the complete metabolic milieu in working muscle that the true cause of acidosis is revealed” (16); this metabolically simple, but not unphysiological, example serves to show the defects of the analysis proposed by Robergs et al. (16).
- Copyright © 2006 the American Physiological Society