Just started learning about aerobic respiration today, specifically Glycolysis.

We were told that aerobic respiration produces 38 ATP, while anaerobic respiration produces only 2.

However, we're also told that:

  • Glycolysis produces 2 ATP
  • Link reaction produces 6 ATP
  • Krebs cycle produces 24 ATP (18 from NADH, 4 from FADH, and an extra 2)

This only adds up to 32 ATP, so where is this other 6 coming from? Or is this information incorrect.


2 Answers 2


In short, the difference stems from different values regarding the number of ATP attributed to the electron carriers in the electron transport chain (ETC).

My guess is that your class didn't go too far in depth on the subject, and also that your class or text is using mixed sources of information.

I have overviewed (general) glucose metabolism here, but feel free to skip this section if you don't need it.

Glucose metabolism is generally composed of 4* processes:

  1. Glycolysis: a 6-C glucose molecule is broken down to 2 pyruvate molecules producing a net gain of 2 ATP and 2 NADH (electron carrier) molecules. Note that this is a net gain.
  2. Pyruvate Dehydrogenase Complex (PDC): Here Pyruvate is made into Acetyl CoA, which enters the Krebs Cycle (step 3). This is the Link reaction step that you mentioned. This step produces a 1 NADH molecule per pyruvate.
  3. Krebs Cycle (TCA): The Krebs cycle is central to cell metabolism and produces a lot of electron carriers, netting 1 GTP (ATP energy "equivalent"), 3 NADH, and 1 "FADH2" (electron carrier) PER Acetyl CoA entering the TCA.
  4. Oxidative Phosphorylation (ETC): This step converts your electron carriers into ATP. It does this by using the energy released by electron transfers (e- are transferred from your electron carriers NADH, FADH2 to intermediate compounds to oxygen. This provides the energy to actively transport protons) to pump protons (H+ atoms) across a semi-permeable membrane to create a electro-chemical gradient. These H+ atoms will flow down this gradient through the channel of the ATP synthase, creating ATP. Currently, it is pretty widely accepted that an NADH molcule provides enough energy to net 2.5 ATP, and a FADH2 molecule provides enough energy to net 1.5 ATP.

As a result, we get the following:

  1. +2ATP, +2NADH
  2. +1NADH x2
  3. (+1GTP, +1FADH2, +3NADH) x2
  4. -10 NADH, +2.5ATP x10; -2FADH2, + 1.5ATP x2

This yields 2+2+25+3=32 ATP, which is what you obtained in your calculation.

However, a previously accepted value states that the 4th step nets 3 ATP/NADH molecule and 2 ATP/FADH2 molecule. Running the same calculation, we obtain 2+2+30+4, netting 38 ATP, causing the discrepancy you notice. I wouldn't say either set of values is universally accepted, but I would say that 32 ATP is generally recognized as the "right" answer.

In other words, 32 ATP is the result of the more recently accepted equivalences of 2.5ATP/NADH and 1.5ATP/FADH2 in oxidative phosphorylation, whereas 38 ATP is the result of the older pair accepted equivalences of 3ATP/NADH and 2ATP/FADH2.

Some notes and caveats:

Please note that in (most) Eukaryotes, the NADH produced in the cytosol (Glycolysis) is "reduced to FADH2" via the mitochondrial shuttle as the high energy electrions enter the ETC. This means that each NADH produced by glycolysis only nets 1.5 ATP, resulting in 30 ATP/glucose in majority of eukaryotes.

It is also important to note that the numbers that I have provided are mostly used in classroom calculations for general systems. Real organisms will often deviate for a multitude of reasons. Further, some organisms may not follow this rule at all.

*A few sources might classify the PDC as part of glycolysis or as part of the TCA.

  • 2
    $\begingroup$ Fine, but you misuse the word “convention”. You are takling about “accepted values” for the equivalence of NADH and FADH2 with ATP. At one time they were thought to be 3 and 2 (i.e. these values were generally accepted), but more recent work gives values nearer 2.5 and 1.5 (which are now generally accepted). It is now usual in eukaryotes to consider the mitochondrial shuttle, so a value of 30 (rather than 32) is cited, e.g. in Berg et. al. $\endgroup$
    – David
    Sep 8, 2018 at 22:13
  • $\begingroup$ @David I went back and edited out the word "convention." I'm usually pretty bad with choosing my words, so I'm not surprised to hear that I misused the word. However, I was wondering what made you say that "convention" was used wrong? $\endgroup$
    – SmallFish
    Sep 8, 2018 at 22:30
  • 2
    $\begingroup$ A "convention" refers to a normal or usual manner of proceeding, and in science usually refers to nomenclature or representation. The reason for not using it in the way you did originally is that it concerns the representation of an object rather than the object itself. Examples are 1. Writing nucleic acid sequences (e.g. GTTC) — 5′ to 3′ direction, phosphodiester linkages omitted; 2. Writing of species in papers — italicized, first mention in full with second part in lower case (Escherichia coli) subsequently first part abbreviated (E. coli). $\endgroup$
    – David
    Sep 9, 2018 at 9:14
  • 1
    $\begingroup$ — I would put it the other way round. The object is a tetranucleotide and its representation is GTTC according to the current convention. If you look at early books on nucleic acids it would have been written GpTpTpC — an earlier convention, but the object is just the same. However as an analogy for the values for NADH/ATP conversion, consider that this was a discrete molecule the structure of which had been wrongly determined and, say, the C was later found to be methylated. Then the object has changed. $\endgroup$
    – David
    Sep 9, 2018 at 21:56
  • 1
    $\begingroup$ @user237650 — SE would regard this as a side discussion, but, in brief, it relates to the way ATP is synthesised by the chemi-osmotic mechanism and the structure of the ATP synthase. If you have access to it and your background is appropriate this paper might help. Otherwise it is a valid query and you should consider asking it as a separate question. $\endgroup$
    – David
    Sep 10, 2018 at 11:59

I dont know how much do you know this metabolic paths but i am going to try to get that 32 ATP out of 38 ATP step by step.

First off all in glycolysis you have to activate sugars with phosphate (glucose;fructose). This process is done by hydrolysis of two ATP. So we have 36 ATP left.

I am going to move forward to the ratios of kofactors. Ratio ATP:NADH+H and ATP:FADH2 in oxidatice phosphorylation is not actually 3 and 2 ATP, but 2.5 and 1.5 ATP respectively.

ATP synthase produces 1 ATP from 3 H+. However the exchange of matrix ATP for cytosolic ADP and Pi (antiport with OH or symport with H+) mediated by ATP–ADP translocase and phosphate carrier consumes 1 H+ / 1 ATP due to regeneration of the transmembrane potential changed during this transfer, so the net ratio is 1 ATP / 4 H+-

The mitochondrial electron transport chain pump 10 H+/1 NADH+H+ and 6 H+/1 FADH2. In the end it is that 2.5 ATP for NADH+H and 1.5 ATP for FADH2.

Sum it. FOr 1 molecule of glucose comes from substrate-level phosphorylation 2 ATP and 2 ATP (GTP) from Krebs cycle dicectly. 2NADH+H from clycolysis (2x 1.5 ATP or 2 x 2.5 ATP from malete aspartate. 8x 2.5 ATP from dekarboxylation of pyruvate and last but not least 2x1.5 ATP for FADH2 from Krebs cycle.

4+3 (or 5)+20+3= 30/or 32 ATP.

Source: https://www.saylor.org/site/wp-content/uploads/2013/04/BIO101A-6.2.3-CellularRespiration.pdf


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .