Am J Physiol Regul Integr Comp Physiol 289: R904 –R910, 2005;

Letters to the Editor doi:10.1152/ajpregu.00117.2005.

REPLY then argue that when experiencing extreme acidosis, muscle

seems to produce a small fraction of protons during ATP
Lingering construct of lactic acidosis
hydrolysis and one proton per lactate? Does this imply that
We would like to thank Dr. Kemp (6) and Böning et al. (4) for lactate production does produce protons during acidic condi-
their letters to the editor and Dr. Lindinger et al. (8) for their tions, even if Dr. Kemp previously agreed that lactate produc-
editorial contributions to the topic of the biochemistry of tion and proton release are unrelated biochemical events? Is
metabolic acidosis in contracting skeletal muscle. As we stated Dr. Kemp saying that the decreased ionization of ATP during
in our original manuscript (9), there have been far too few a developing acidosis means that ATP hydrolysis does not
academic discussions and reviews on this topic, resulting in the cause acidosis? What does Dr. Kemp view to be the proton
incorrect acceptance of the lactic acidosis construct as fact. releasing and proton consuming reactions of muscle metabo-
To specifically address the comments from each letter, we lism? Is the main fundamental criticism that we should have
provide answers to each query separately, below. We provide tallied true proton release numbers based on partial ionization
a short response to the editorial of Dr Lindinger et al. after our models?
responses to the two letters. On the basis of the absence of a clear purpose for Dr.
Nevertheless, before responding to specific contents of the Kemp’s criticisms, we read some of his past research (5), and
two letters, we would like to reinforce the clear and simple the research of his colleagues (1, 7). This process was ex-
evidence that we feel is being overlooked in the two letters to tremely helpful, for we found the following quotes from these
the editor. To argue for a lactic acidosis means that both of the studies.
phosphoglycerate kinase and lactate dehydrogenase reactions “. . . ‘proton balance’ only involves glycolytic lactate/H⫹
are incorrect. generation and net H⫹ consumption by PCr splitting.” (Ref. 5;
The fact is that the biochemistry of these two reactions (Figs. 1 p. 901, Abstract point 1)
and 2) is factual and, as such, irrefutable and not open for “Total H⫹ load (lactate load less H⫹ consumption) was used
debate or discussion. These reactions clearly reveal that the to estimate cytosolic buffer capacity.” (Ref. 5; p. 901, Abstract
lactic acidosis construct is a fictional concept that serves to point 3)
reinforce misunderstood muscle biochemistry and preserve A schematic biochemical summary shows H⫹ production
decades of errors and misinterpretation. To argue the accep- linked to lactate production (Ref. 5; p. 904, Fig. 1).
tance of this fiction is unscientific. Dr. Kemp balances the substrates and products of glycolysis
We entertain this series of letters to the editor because they (commencing with glycogen) in an ischemic condition (no
once again allow us to state the truths about metabolic acidosis, mitochondrial respiration), then couples glycolytic ATP pro-
highlight continued misinterpretations, and hopefully eradicate duction to ATP hydrolysis (e.g., muscle contraction) and re-
any thought of a lactic acidosis from academia and scientific veals the following.
research. Glycogen-n ⫹ H2O 3 Glycogen-n-1 ⫹ 2 Lactate ⫹ 2 H⫹
Response to Letter to the Editor: Kemp GJ. Lactate accu- (Ref. 5; p. 905)
mulation, proton buffering, and pH change in ischemically “In normal cellular functions the creatine kinase reaction is
exercising muscle. coupled to ATP utilization and resynthesis; the sum of both
reactions is classically known as the Lohmann reaction.” (Ref.
WHAT WAS THE PURPOSE OF OUR MANUSCRIPT? 7; p. C1741)
We were very clear in what we wanted to accomplish with “During exercise, the net H⫹ load that results in pH changes
our manuscript. Our purpose was to show that it is a biochem- is the difference between glycolytic H⫹ production (⫽change
ical impossibility for lactic acid to be produced in skeletal in lactate) and H⫹ consumption by PCr splitting.” (Ref. 1; p.
muscle. In addition, we presented evidence based on simple 2392)
computations and past research that revealed that there is no We will address the errors or misleading nature of all of the
stoichiometry to lactate and proton production in contracting aforementioned quotes in the sections to follow. Clearly, the
skeletal muscle, with proton production far in excess of lactate view of Dr. Kemp and his associates is that lactate production
production. Finally, we stated and provided research evidence causes H⫹ production, and subsequently, that total proton
in support of the cause of metabolic acidosis to be an increased release (proton load) can be indirectly calculated from the
dependence on nonmitochondrial ATP turnover. change in lactate accumulation when adjusted for the protons
It was not our intention to provide definitive numbers for consumed in the creatine kinase reaction. Of course, these
total proton release from muscle metabolism during exercise, beliefs and interpretations of metabolic biochemistry in skele-
and we did not attempt to provide definitive values for a muscle tal muscle are incorrect, as we further explain in this response.
buffer capacity. We also identified that our numbers would not
be totally accurate based on the pH dependence of proton IS IT VALID TO COUPLE THE ATPASE AND CREATINE
release from ATP hydrolysis and proton binding to metabolites KINASE REACTIONS TOGETHER INTO THE
such as ADP and inorganic phosphate (Pi) (9; see p. R513, left LOHMANN REACTION?
column, paragraph 1).
Bendahan et al. (1), Kemp (5, 6), and Kushmerick (7) based
WHAT IS THE PURPOSE OF THE LETTER OF DR. KEMP?
their whole presentations of muscle metabolism on the assump-
tion that ATP turnover is tightly coupled to the creatine kinase
We found it difficult to comprehend what the real purpose of reaction during intense muscle contractions, or in Kemp’s
the letter of Dr. Kemp is. Is Dr. Kemp attempting to say that he example, ischemic exercise. Kushmerick (7) presents the cou-
agrees that lactate production does not produce protons, but pled balance of the Lohmann reaction (it is not a reaction in
R904 0363-6119/05 $8.00 Copyright © 2005 the American Physiological Society http://www.ajpregu.org
 Letters to the Editor
R905

Fig. 1. The substrates and products of the phosphoglycerate kinase reaction. The product 3-phosphoglycerate is the first “carboxylic acid” formed in glycolysis.
The phosphate transfer of this reaction reveals that a proton was never present to be released to the cytosol and alter cellular proton exchange and pH. As such,
3-phosphoglycerate and all of the remaining glycolytic carboxylic acid intermediates do not function as acids as they never have a proton that can be released
into solution. Arrow pointing away from a bond represents bond/group removal. Arrow pointing to a bond represents addition of an atom/group.

biochemical terms, nor is it biochemically coupled) as follows. piration. It is only when presenting the complete metabolic
⫹ milieu in working muscle that the true cause of acidosis is
ATPase ATP 3 ADP ⫹ Pi ⫹ ␣H
revealed. Falsely coupling separate reactions is not a valid
⫹ CK PCr ⫹ ADP 3 ATP ⫹ Cr ⫹ ␤H⫹ presentation of metabolic biochemistry and leads to misunder-
⫽ Lohmann reaction: PCr 3 Cr ⫹ Pi ⫹ ␥H⫹ standings of proton balance.
where ␣, ␤, and ␥ represent ionization coefficients and CK is
creatine kinase. DOES THE PH DEPENDENCE OF THE PROTON YIELD FROM
At best, such an approach may only be accurate for the ATP HYDROLYSIS CHANGE THE BIOCHEMICAL CAUSE
initial reactions of repeated muscle contractions. At worst, the OF ACIDOSIS?
Lohmann reaction concept is an oversimplification of the
contribution of the phosphagen system to the biochemistry of We agree that the proton yield of ATP hydrolysis is pH
skeletal muscle metabolism. In an anoxic environment (no dependent. However, this fact is largely irrelevant to the
muscle mitochondrial respiration), the ADP produced from understanding of the cause of metabolic acidosis. Skeletal
ATP hydrolysis is not solely consumed by the creatine kinase muscle fatigue during intense exercise is rapid in onset and
reaction, but is also used in glycolysis and the adenylate kinase unpreventable during “all out” efforts. As we depict in Fig. 3,
(myokinase) reaction (ADP ⫹ ADP 3 ATP ⫹ AMP). The Pi the ATP turnover of skeletal muscle during such conditions
produced from ATP hydrolysis is also used as a substrate for mirrors the muscle force production profile, as researched
glycolysis. AMP is converted to IMP and ammonia (AMP by Spriet et al. (10). Thus the period of greatest ATP
deaminase) during acidosis, revealing that the sum of AMP, turnover occurs early, during the initial phase of muscle
IMP (or ammonia), and glycolytic flux equals the inaccuracy of fatigue, and decreases rapidly, resulting in relatively small
the Lohmann reaction in accounting for all ATP regeneration. ATP turnover during what has now become an extreme acidic
Because there is no equality or known stoichiometry between condition.
ATP hydrolysis and each of glycolysis, the creatine kinase In short, once muscle is acidic, it is also severely compro-
reaction, and the AMP deaminase reaction (they can all be mised in its ability to generate force and thus demand ATP
quantified, but their relative proportions can vary depending turnover. These are not necessarily cause-and-effect responses.
on the exercise condition) in working muscle, summary Clearly, the metabolic period of relevance to the development
reactions of part or all of the phosphagen system are of metabolic acidosis is the condition change from neutral cell
misleading and inaccurate. pH to moderate acidosis (pH decrease from 7.0 to ⬃6.6),
Our Eqs. 3 and 4 (9; p. R507), Eqs. 5 and 6 (9; p. R508), Fig. conditions where Dr. Kemp seems to agree that decreases in
11 (9; p. R510) and Fig. 14 (9; p. R511) provide a detailed pH are best explained by ATP hydrolysis. Applying our pre-
account of the origin and fate of ADP, Pi, and H⫹ during sentation of metabolic biochemistry and applying the decreas-
muscle metabolism. We included mitochondrial respiration, for ing proton release from ATP during acidosis reveal that during
almost all exercise conditions have functioning mitochondria, severe acidosis, there are major constraints to added proton
oxygen provision—and hence substrate consumption (O2, release (ATP proton gain and decreased ATP demand). The
ADP, Pi, H⫹)—and ATP production from mitochondrial res- important issue here is not what happens to proton balance

Fig. 2. The substrates and products of the lactate dehy-

drogenase (LDH) reaction. Two electrons and a proton
are removed from NADH, and a proton is consumed
from solution, to reduce pyruvate to lactate. Arrow
pointing away from a bond represents bond/group re-
moval. Arrows pointing to a bond represent addition of
an atom/group.

AJP-Regul Integr Comp Physiol • VOL 289 • SEPTEMBER 2005 • www.ajpregu.org

Letters to the Editor
R906
ration. As such, these scientists use lactate production to
estimate the proton load (adjusted for the creatine kinase
reaction). This is incorrect for all the reasons presented in our
original manuscript and which require no further repetition
here. You cannot estimate a muscle buffer capacity using
lactate production to account for proton release. Lactate pro-
duction consumes a proton, and thus the estimates of muscle
buffer capacity used by Dr. Kemp and Dr. Bendahan must
therefore grossly underestimate the true buffer capacity, as
they underestimate both proton load (they should use their own
data for nonmitochondrial ATP turnover) and metabolic proton
removal (they need to add lactate production to the reactions
comprising metabolic buffering). It does not matter what sen-
sitive methods are used to quantify ATP, ADP, creatine phos-
phate and muscle pH, the incorrect estimation of proton loads
condemns these data to major error.
Response to Letter to the Editor: Böning D, Benele R, and
Fig. 3. Relative change in the force production and estimated rates of ATP Maassen N. Lactic acid still remains the real cause of exercise-
turnover and proton release during 64 repeated contractions of the human induced metabolic acidosis.
quadriceps muscles for 1.6 s at 20 Hz electrical stimulation, with occluded
blood flow. The darker the intensity of the shading represents greater rates of Title
estimated proton release. Adapted from Spriet et al. (10).
We are obviously totally opposed to the wording of this title.
when the cell pH approaches 6, but how cell pH got that low Source of Increasing Pi in Exercising Muscle
to begin with.
Finally, this issue reveals the inconsistency in Dr. Kemp’s As stated in our manuscript, and in every textbook of metabolic
arguments within his letter (6). Dr. Kemp commences his letter biochemistry, the creatine kinase reaction is as follows.
with recognition to the proton release from ATP hydrolysis and
that this is likely to cause initial reductions in muscle pH Creatine Kinase
(1)
during repeated intense (as well as hypoxic or anoxic) contrac- CrP ⫹ ADP ⫹ H⫹ N Cr ⫹ ATP
tions. Dr. Kemp states, “I will argue that while this analysis
[that ATP hydrolysis disconnected from mitochondrial respi- Böning and colleagues argue that the creatine kinase reaction is
ration causes acidosis] of proton generation is broadly correct also coupled to ATP hydrolysis. However, where is the evi-
at resting pH, it is much less so at low pH values found in dence for this assumption? Böning et al. (4) present none. We
exercising muscle.” Dr. Kemp then presents admirable evi- clearly argued against this coupling in our reply to Dr. Kemp
dence for the decreased proton release from ATP hydrolysis as on the issue of the fictitious “Lohmann reaction” (see section:
IS IT VALID TO COUPLE THE ATPASE AND CREATINE KINASE REACTIONS
pH declines, but then states the following, “. . .at pH 6, most of
TOGETHER INTO THE ‘LOHMANN REACTION’?). To comprehend the
the ‘glycolytic’ protons actually do come from glycolysis to
lactate.” Dr Kemp bases this last statement in part on his Eq. error of this view and all that it implies, read the later section
5b, revealing the decreased ionization of ATP at pH 6, as well of this response: A BÖNING-KEMP VIEW OF METABOLIC BIOCHEMISTRY.
as summary reactions of ATP hydrolysis, glycolysis, and Amount of Proton Release During Glycolysis
lactate production (6; see Table 1). However, what Dr. Kemp
fails to realize is that this summary reaction that we detailed in It is incorrect to label our presentation of glycolysis as
Eqs. 1– 6 reveals that the proton product accompanying lactate wrong or incomplete. To reinforce this issue, we present the
does not come from lactate production, but from ATP hydro- main and summary reactions of intermediary metabolism in
lysis (9; Fig. 11, p. R510). Thus, at a pH near 6, there is really contracting skeletal muscle to clearly show that our presenta-
minimal proton release, and his summary reaction (our Eq. 5) tion was valid. We also present these reactions, as the deriva-
should be written with proton release adjusted by an ionization tion of the main summary reactions of carbohydrate and lipid
constant close to zero. catabolism is a useful exercise for the student, scientist, and
academic alike and, as such, is a great test of the mastery of
ARE ESTIMATES OF MUSCLE BUFFER CAPACITY metabolic biochemistry.
INACCURATE IF THEY ASSUME LACTATE PRODUCTION
CONTRIBUTES TO THE MUSCLE PROTON LOAD? Glycolysis
We do not understand the effort of Dr. Kemp to argue Glucose ⫹ 2 Pi ⫹ 2 ADP ⫹ 2 NAD⫹ 3 2 Pyruvate
(2)
against the logic and factual content of our manuscript on the ⫹ 2 ATP ⫹ 2 NADH ⫹ 2 H⫹ ⫹ 2 H2O
grounds of muscle buffer capacity. Earlier, we revealed in the
quotes from past research of Bendahan et al. (1), Dr. Kemp (5), Pyruvate Dehydrogenase Complex
and Kushmerick (7) that these scientists accept the notion that
2 Pyruvate ⫹ 2 CoA ⫹ 2 NAD⫹ 3 2 Acetyl-CoA
lactate production is stoichiometric to proton release in con- (3)
tracting skeletal muscle in the absence of mitochondrial respi- ⫹ 2 NADH ⫹ 2 CO2

AJP-Regul Integr Comp Physiol • VOL 289 • SEPTEMBER 2005 • www.ajpregu.org

 Letters to the Editor
R907

TCA Cycle Glucose-6-phosphate ⫹ 2 Pi ⫹ 3 ADP

2 Acetyl-CoA ⫹ 6NAD⫹ ⫹ 2 FAD⫹ ⫹ 2 GDP ⫹ 2 NAD⫹ 3 2 Pyruvate ⫹ 3 ATP (16)
⫹ 2 Pi ⫹ 4 H2O 3 4 CO2 ⫹ 6 NADH ⫹ 2 FADH2 (4) ⫹ 2 NADH ⫹ 1H⫹ ⫹ 2H2O
⫹ 2 GTP ⫹ 4 H⫹ ⫹ 2 CoA Pyruvate Dehydrogenase Complex
Electron Transport Chain 2 Pyruvate ⫹ 2 CoA ⫹ 2 NAD⫹ 3 2 Acetyl-CoA
(17)
10 NADH ⫹ 2 FADH2 3 10 NAD⫹ ⫹ 2 FAD⫹ ⫹ 2 NADH ⫹ 2 CO2
(5)
⫹ 24 e⫺ ⫹ 14 H⫹6 O2 ⫹ 24 e⫺ ⫹ 24 H⫹ 3 12 H2O TCA Cycle
Phosphorylation 2 Acetyl-CoA ⫹ 6 NAD ⫹ 2 FAD⫹ ⫹ 2 GDP
⫹

34 ADP ⫹ 34 Pi ⫹ 34 H⫹ 3 34 ATP ⫹ 34 H2O (6) ⫹ 2 Pi ⫹ 4 H2O 3 4 CO2 ⫹ 6 NADH ⫹ 2 FADH2 (18)

Net Balance of Complete CHO Oxidation ⫹ 2 GTP ⫹ 4 H⫹ ⫹ 2 CoA

Electron Transport Chain
Glucose ⫹ 6 O2 ⫹ 38 ADP ⫹ 38 Pi ⫹ 38 H⫹ 3 38 ATP
(7) 10 NADH ⫹ 2 FADH2 3 10 NAD⫹ ⫹ 2 FAD⫹
⫹ 6 CO2 ⫹ 44 H2O (19)
⫹ 24 e⫺ ⫹ 14 H⫹6 O2 ⫹ 24 e⫺ ⫹ 24 H⫹ 3 12 H2O
Muscle Contraction
Phosphorylation
38 ATP ⫹ 38 H2O 3 38 ADP ⫹ 38 Pi ⫹ 38 H⫹ (8)
34 ADP ⫹ 34 Pi ⫹ 34 H⫹ 3 34 ATP ⫹ 34 H2O (20)
Net Balance of Glucose Oxidation and Muscle Contraction
Net Balance of Complete Glycogen Source Glucose Oxidation
Glucose ⫹ 6O2 3 6CO2 ⫹ 6H2O (9)
Glycogen-n ⫹ 6 O2 ⫹ 39 ADP ⫹ 39 Pi ⫹ 39 H⫹ 3
(21)
For Glucose Ending in Lactate Glycogen-n⫺1 ⫹ 39 ATP ⫹ 6 CO2 ⫹ 44 H2O
Glycolysis Muscle Contraction
⫹
Glucose ⫹ 2 Pi ⫹ 2 ADP ⫹ 2 NAD 3 2 Pyruvate 39 ATP ⫹ 39 H2O 3 39 ADP ⫹ 39 Pi ⫹ 39 H⫹ (22)
⫹
(10)
⫹ 2 ATP ⫹ 2 NADH ⫹ 2 H ⫹ 2 H2O Net Balance of Glycogen Oxidation and Muscle Contraction
Lactate Dehydrogenase Reaction Glycogen-n ⫹ 6 O2 3 Glycogen-n⫺1 ⫹ 6 CO2 ⫹ 6 H2O (23)
⫹ ⫹
2 Pyruvate ⫹ 2 NADH ⫹ 2 H 3 2 Lactate ⫹ 2 NAD (11) RQ ⫽ 6/6 ⫽ 1.0
Glucose to Lactate For Glycogen Ending in Lactate
Glucose ⫹ 2 Pi ⫹ 2 ADP 3 2 Lactate Glycogenolysis
(12)
⫹ 2 ATP ⫹ 2 H2O Glycogen-n ⫹ Pi 3 Glucose-6-phosphate
(24)
Muscle Contraction ⫹ Glycogen-n⫺1
2 ATP ⫹ 2 H2O 3 2 ADP ⫹ 2 Pi ⫹ 2 H⫹ (13) Glycolysis
Net Balance of Glucose to Lactate and Muscle Contraction Glucose-6-phosphate ⫹ 2 Pi ⫹ 3 ADP

Glucose 3 2 Lactate ⫹ 2 H⫹ (14) ⫹ 2 NAD⫹ 3 2 Pyruvate ⫹ 3 ATP (25)

⫹
⫹ 2 NADH ⫹ 1 H ⫹ 2 H2O
Note, as explained in the text, we disagree with the coupling of
the ATP hydrolysis of muscle contraction to glycolysis as not Lactate Dehydrogenase Reaction
all ATP, ADP, and Pi from catabolism cycles 100% between
2 Pyruvate ⫹ 2 NADH ⫹ 2 H⫹ 3 2 Lactate ⫹ 2 NAD⫹ (26)
muscle contraction and glycolysis.
Glycogen to Lactate
When Glycogen Is The Substrate Glycogen-n ⫹ 3 Pi ⫹ 3 ADP ⫹ 1 H⫹ 3 Glycogen-n⫺1
(27)
Glycogenolyis ⫹ 2 Lactate ⫹ 3 ATP ⫹ 2 H2O
Glycogen-n ⫹ Pi 3 Glucose-6-phosphate Muscle Contraction
(15)
⫹ Glycogen-n⫺1 3 ATP ⫹ 3 H2O 3 3 ADP ⫹ 3 Pi ⫹ 3 H⫹ (28)
Glycolysis Net Balance of Glycogen to Lactate and Muscle Contraction

AJP-Regul Integr Comp Physiol • VOL 289 • SEPTEMBER 2005 • www.ajpregu.org

Letters to the Editor
R908

Glycogen-n ⫹ H2O 3 Glycogen-n⫺1 ⫹ 2 Lactate ⫹ 2 H⫹ (29) We know of no research evidence or biochemical principles
that validate the direct coupling of ATP hydrolysis to the
Note, as explained in the text, we disagree with the coupling of creatine kinase reaction or glycolysis under any muscle con-
the ATP hydrolysis of muscle contraction to glycolysis, as not tractile condition. As we explained in our response to Dr.
all ATP, ADP, and Pi from catabolism cycles 100% between Kemp, our Figs. 11 and 14 present the reality of the complexity
muscle contraction and glycolysis. and interactions of substrates and products across multiple
What is “most problematic” with our presentation of these metabolic pathways and isolated reactions and in interaction
reactions in the minds of Böning et al. is that we did not focus with the compartmentalization of mitochondria within con-
on our Eq. 5 and 6 (9; p. R508), which show glucose or tracting skeletal muscle.
glycogen, “coupled” to ATP hydrolysis. Yes, when you “cou-
ple” glycolysis to ATP hydrolysis, you do end up with the Exercise-Induced Metabolic Acidosis Without Production of
production of 2 lactate and 2 H⫹ (Eqs. 14 and 29). However, Lactic Acid
the equality of the glycogen vs. glucose source proton release
depends on whether you keep ATP turnover the same for both We do not find this argument to be logical or clearly in
substrates. Our Eq. 6 was based on the fact that when muscle support of your view. Individuals who suffer from McArdle’s
glycogen is used to fuel glycolysis, there is no involvement of disease are not only unable to produce lactate, but they are
the hexokinase reaction (which releases a proton and consumes unable to rely on glycolysis for ATP turnover during muscle
an ATP). Thus, for every 2 lactate produced and if the 2 ATP contraction. As such, McArdle’s disease causes a severe re-
gain from glycolysis remains your reference, there is one less duction in the capacity of nonmitochondrial ATP turnover.
proton released when catabolizing glycogen vs. glucose. In Furthermore, the remaining metabolic source of nonmitochon-
other words, for a given rate of ATP turnover, there are less drial ATP turnover is the creatine kinase reaction (Eq. 1),
protons released from glycolysis when starting with glycogen which consumes a proton and the adenylate kinase reaction.
(protons:ATP ⫽ 2:3) vs. glucose (2:2). This important contrast Thus McArdle’s disease does not disprove the nonmitochon-
is overseen when interpreting the resulting balanced reaction of drial ATP turnover cause of metabolic acidosis. If you severely
glycolysis coupled to ATP hydrolysis. diminish the capacity for nonmitochondrial ATP turnover, you
Unlike Böning et al., who state that the summary reaction prevent the ability to sustain intense muscle contractions and
products of 2 lactate ⫹ 2 H⫹ is lactic acid, we view this consequently prevent metabolic acidosis.
interpretation to be incorrect. This interpretation is not based
on biochemical fact, scientific principles, or research-based Lactate and Proton Release From Muscle to Blood
evidence, but rather on a convenient explanation construed
over the decades since the work of Hill and Meyerhoff (see We addressed this issue in our response to Dr. Kemp in the
Ref. 9) to provide a simple explanation of metabolic acidosis. section ARE ESTIMATES OF MUSCLE BUFFER CAPACITY INACCURATE IF
This simple and convenient explanation is based on two dis- THEY ASSUME LACTATE PRODUCTION CONTRIBUTES TO THE MUSCLE
turbing assumptions: 1) that all the ADP, Pi, and H⫹ from ATP PROTON LOAD? However, it needs to be stressed yet again that if
hydrolysis are only used to reform ATP from glycolysis and you compute a nonbicarbonate buffer capacity based on the
most disturbingly, 2) that you can take products from com- assumption that lactate production is the source of protons, you
pletely different reactions, combine them together to form a will obviously get a stoichiometry between proton release and
new end product, and interpret this end product to be causal to lactate release. As we have shown in an earlier quote from Dr.
altered cellular and system conditions of acid-base balance. It Böning’s work (2, 3), he assumes that quantifying lactate
is for these reasons that we chose not to emphasize Eqs. 14 and quantifies metabolic proton release and that the change in
29 in our manuscript. blood lactate quantifies blood proton gain. These assumptions
Why do Dr. Kemp and Böning et al. believe in a lactic are obviously wrong. Is this a harsh assessment of this work?
acidosis? We do not know for sure but speculate that is what Maybe. But we once made the same errors but have recognized
they were taught, that is what they presumably teach their the errors of our past assumptions and changed the way we
students, and that is the foundation upon which they conduct teach and conduct research on the topic of acidosis. Many other
and publish research. Such a publication track record is also academics and scientists have done the same. Since the publi-
clear for Dr. Böning, as revealed below. cation of our manuscript (6), there is no longer the excuse of
“During intense exercise a large amount of lactic acid (La) “but that is how we have always interpreted acidosis.”
is produced in the body and may disturb homeostasis, leading
to an increase in hydrogen ion (H⫹) activity.” (2) Muscular Buffers
“. . . lactic acid is practically completely dissociated into
La⫺ and H⫹ because of its low pK (3.0). ” (3) What is your point here? Yes, the research on muscle buffer
“The extracellular pH defense against the lactic acidosis capacity is fraught with inconsistencies, and as we explained in
resulting from exercise can be estimated from the ratios our response to Dr. Kemp, fundamental flaws. As most re-
⌬[La]䡠⌬pH⫺1 (where ⌬[La] is the change in lactic acid con- search estimating total or metabolic buffer capacities has used
centration and ⌬pH is the change in pH) and ⌬[HCO⫺ 3 ]䡠⌬pH
⫺1
lactate to estimate muscle proton release, which includes re-
⫺
(where ⌬[HCO3 ] is the change in bicarbonate concentration) search from Dr. Kemp (5) and Dr. Böning (2), this research
in blood plasma. The difference between ⌬[La]䡠⌬pH⫺1 and needs to either be redone or reassessed with revised computa-
⌬[HCO⫺ 3 ]䡠⌬pH
⫺1
yields the capacity of nonbicarbonate buff- tions that treat lactate production as a proton consumer and
ers (mainly hemoglobin).” (3) nonmitochondrial ATP turnover as the source of protons.

AJP-Regul Integr Comp Physiol • VOL 289 • SEPTEMBER 2005 • www.ajpregu.org

 Letters to the Editor
R909
Calculation of Nonmitochondrial ATP Production lactic acid in spite of the last step for proton release being ATP
hydrolysis,” (4) into correct perspective— clear fiction.
Thank you for detecting this typographical error.
A BÖNING-KEMP VIEW OF METABOLIC BIOCHEMISTRY
Response to Editorial: Lindinger MI, Kowalchuk JM, Heigen-
hauser GJF. Applying the laws of physics to muscle acid-base
As previously stated, Dr. Kemp and Dr. Böning have made chemistry.
fundamental biochemical errors in their arguments in support We are indebted to the prior work of Lindinger et al. (8) for
of a lactic acidosis. To put such views into their true perspec- teaching us to recognize the importance of the strong ion
tive for metabolism, Dr. Kemp and Böning et al. believe that difference (SID) in understanding proton balance and activity.
you can take products of totally different reactions, couple Nevertheless, we feel that it is important to clarify an under-
these reactions together, and combine the products into new lying message contained in this editorial.
molecules to provide a valid assessment of muscle metabolism. The question crucial to the main purpose of our original
The question that we have is if you view this approach to be manuscript is “Where do the protons come from?” Clearly,
valid then why stop at the production of this new fictitious protons do not come from the lactate dehydrogenase reaction
molecule called lactic acid? We can continue this principle, and therefore the production of lactate. However, the Lindinger
seemingly unconstrained by the laws of bioenergetics and et al. (8) presentation of the physical chemistry pertaining to
Einstein’s theories, to bring together whatever end products the acid-base balance of aqueous solutions indirectly argues
from different reactions we like, simply to support whatever that there is no evidence for proton consumption and release
unscientifically based theory we can dream up. from chemical reactions and that changes in cellular or blood
For example, the proton from ATP hydrolysis can leave the pH are caused predominantly by changes in strong ions (their
muscle cell where it is buffered by bicarbonate (HCO⫺ 3 ). Eq. 7). We do not agree.
Thanks to the application of this “logic”, the incomplete Despite the solid basic science foundation of the SID concept,
oxidation of glucose to lactate can now be changed, as follows. we feel that it alone does not answer the question of where do the
protons come from. If it did, then the answer is water, and this
Glucose ⫹ H2O 3 2 Lactate ⫹ 2 H⫹
opposes what we know of the organic chemistry of the reactions
2 H ⫹ ⫹ 2 HCO3⫺ 3 2 H2 CO3 3 2 H2O ⫹ 2 CO2 of intermediary metabolism. Certainly, the SID approach can
explain the final aqueous solution pH when given component
In summary changes for a given proton release condition of cellular metabo-
Glucose ⫹ 2 HCO3⫺ 3 2 lactate ⫹ H2O ⫹ 2 CO2 lism. Thus we feel that to best comprehend the acid-base changes
associated with intense exercise, you need a combination of a
The interpretation of this reaction is that glucose oxidation stoichiometric approach fine tuned by the SID method. For ex-
outside of the mitochondria produces CO2! ample, given the metabolic proton release conditions of intense
However, we can continue this view of metabolic biochem- exercise, the SID computations would predict the final muscle
istry. Lactate leaves the muscle and can be circulated to the pH. Such a scenario is presented in our Fig. 17 (9; p. R513),
liver where it is reduced to pyruvate by a reversal of liver where the final pH resulting from this imbalance in proton
lactate dehydrogenase, and then converted to glucose via liver release and consumption would adhere to the SID concept.
gluconeogenesis. Thus, applying the views of Dr. Kemp and
Böning et al., we could also combine these reactions, as follows. What Is the Take-Home Message?
⫹ ⫹
2 lactate ⫹ 2 NAD 3 2 pyruvate ⫹ 2 NADH ⫹ 2 H We have learned that despite a wealth of e-mail messages
2 pyruvate ⫹ 4 ATP ⫹ 2 GTP ⫹ 6 H2O ⫹ 2 NADH from around the world that congratulated us on our recent
publication, there remain numerous physiologists and meta-
⫹ 2 dihydroxyacetone bolic biochemists consumed by the inertia of viewing muscle
acidosis as a lactate-caused condition.
phosphate 3 glucose ⫹ 4 ADP ⫹ 2 GDP ⫹ 6 Pi ⫹ 2 NAD ⫹
We encourage Dr. Kemp and Böning et al. to reassess their
In summary own data (2, 5), view proton production to be caused by ATP
2 lactate ⫹ 4 ATP ⫹ 2 GTP ⫹ 6 H2O hydrolysis, and recalculate muscle proton loads and buffer
capacities from their prior research. We thank Lindinger et al.
⫹ 2 dihydroxyacetone phosphate 3 glucose (8) for their insightful explanation of the SID, and we will
⫹ 4 ADP ⫹ 2 GDP ⫹ 6 Pi ⫹ 2 Hⴙ include the SID, and our interpretations of its merit and
limitations, in our future manuscripts on the topic of the
These new reactions reveal that the incomplete oxidation of biochemistry of metabolic acidosis.
glucose uses ATP and produces glucose. Not only that, but the It has been more than 80 years since the work of Hill and
real culprit in the cause of acidosis is not lactic acid at all, but Meyerhoff (9). We have known the organic and acid-base
the production of glucose acid (glucose ⫹ 2H⫹). chemistry of the lactate dehydrogenase reaction for more than
Obviously, these reactions are fictitious, but no more so than 50 years. It has been 28 years since the initial publications that
the production of lactic acid, or the existence of the Lohmann criticized the lactic acidosis construct. We are impatient for
reaction. We hope that this extension of the principles em- science to recognize that three is no such thing as a lactic
ployed by Dr. Kemp and Böning et al. places the following acidosis, thus allowing more valid research and learning of the
interpretation of Böning et al., “The resulting metabolite is intricacies of muscle metabolic regulation and proton balance.

AJP-Regul Integr Comp Physiol • VOL 289 • SEPTEMBER 2005 • www.ajpregu.org

Letters to the Editor
R910
We continue to hypothesize that an increased dependence on 8. Lindinger MI, Kowalchuk JM, and Heigenhauser GJF. Applying the
nonmitochondrial ATP turnover is currently the most valid expla- laws of physics to muscle acid-base chemistry. Am J Physiol Regul Integr
Comp Physiol 289: R000 –R000, 2005.
nation of the cause of exercise-induced metabolic acidosis. 9. Robergs RA, Ghiasvand F, and Parker D. Biochemistry of exercise-
REFERENCES induced metabolic acidosis. Am J Physiol Regul Integr Comp Physiol 287:
R502–R516, 2004.
1. Bendahan D, Kemp GJ, Roussel M, Fur YL, and Cozzone PJ. ATP 10. Spriet LL, Sodeland K, Bergstrom M, and Hultman E. Aerobic energy
synthesis and proton handling in muscle during short periods of exercise release in skeletal muscle during electrical stimulation in men. J Appl
and subsequent recovery. J Appl Physiol 94: 2391–2397, 2003. Physiol 62: 611– 615, 1987.
2. Böning D, Maassen N, Thomas A, and Steinacker JM. Extracellular pH
defense against lactic acid in normoxia and hypoxia before and after a
Himalayan expedition. Eur J Appl Physiol 84: 78 – 86, 2001.
3. Böning D. Improved blood buffering in high-altitude natives. J Appl Robert A. Robergs
Physiol 93: 2214 –2215, 2002. Farzenah Ghiasvand
4. Böning D, Benele R, and Maassen N. Lactic acid still remains the real Exercise Physiology Laboratories, Exercise Science
cause of exercise-induced metabolic acidosis. Am J Physiol Regul Integr Program
Comp Physiol 289: R902–R903, 2005.
5. Kemp GJ. Interrelations of ATP synthesis and proton handling in ische- Department of Physical Performance and Development
mically exercising human forearm muscle studied by 31P magnetic reso- The University of New Mexico, Albuquerque, NM
nance spectroscopy. J Physiol 535: 901–928, 2001.
6. Kemp G. Lactate accumulation, proton buffering, and pH change in Daryl Parker
ischemically exercising muscle. Am J Physiol Regul Integr Comp Physiol
289: R895–R901, 2005. Exercise Science Program
7. Kushmerick MJ. Multiple equilibria of cations with metabolites in muscle California State University–Sacramento
bioenergetics. Am J Physiol Cell Physiol 272: C1739 –C1747, 1997. Sacramento, CA

AJP-Regul Integr Comp Physiol • VOL 289 • SEPTEMBER 2005 • www.ajpregu.org