January 2, 2015

Metabolic Theory of Cancer: Glycolysis Run Amok & Mutant Hexokinase

“The ability to sustain an enhanced glycolytic rate represents one of the most consistent and profound biochemical phenotypes of many cancer cells.
                                            -Mathupala et al, 1997.

Through the previous posts in this series, we have woven a path through an introduction to the metabolic origins theory of cancer, some basic facts about cancer cells, cellular energy generation, mitochondrial structure & function, and potential causes of mitochondrial dysfunction. We have also established that there is a large degree of mitochondrial abnormality in cancer cells. We ended things last time by saying that we would explore cancer's next two most striking calling cards. So here we go. More of the metabolic hallmarks of cancer cells. (And finally, you will start to see why all my blabbing on and on about glycolysis and the Krebs cycle were necessary. I promise!)

Before we get into things today, I’d like to thank anyone who’s still along for the ride. I genuinely believe the metabolic theory of cancer holds an incredible amount of promise, and I consider it a privilege to be able to share with you my understanding—however rudimentary—of the science involved. It has occurred to me that, for the purpose of this series, my blog has become a free course in biochem & physiology, rather than the popular stuff with sensationalist, attention-grabbing headlines about adrenal fatigue, digestion, and women’s hormones. I may be getting far fewer page hits than the big boys and various “gurus” out there, but I sincerely think this could be life-saving information. (And frankly, between you and me…just us friends here…I’m kinda tired of all that adrenal/thyroid/digestion stuff, even if I do find it fascinating most of the time.) Based on the number of page hits, not too many folks are interested in this deep dive into the metabolic origins of cancer. Oh well. Their loss. I like writing about this, and I know there are at least one or two of you out there who value it, so I'll keep going. (And I will try not to be discouraged that other people's posts about Paleo chocolate chip banana bread and such generate more buzz than this supremely critical information.)

So to anyone out there whos still with me, thank you. I hope you continue to find it worth your time.


Recall that we’ve discussed two primary pathways for the creation of cellular energy (ATP). The first, glycolysis, is a relatively inefficient pathway that generates a net of just 2 ATP per molecule of glucose metabolized. Glycolysis does not require oxygen, and it takes place in a cell’s cytoplasm. The second pathway, cellular respiration (also called oxidative phosphorylation, or OxPhos), generates around 36 ATP per molecule of glucose metabolized, making it a much more efficient pathway. (Efficient doesn't necessarily mean its better; it just means that we end up with way more energy from the very same amount of starting material. I guess maybe that does make it better, but really, both pathways are necessary for human life.)

It’s all about energy!
OxPhos requires oxygen, and it takes place inside mitochondria. Another way the pathways differ is that glycolysis is, by definition, limited to glucose. OxPhos, on the other hand, is a pathway that can harvest energy from glucose, fatty acids, amino acids, and ketones as starting materials. (And generally speaking, molecule-for-molecule, we harvest way more energy from fats than from glucose. See this post for details. If you don’t feel like reading that post, here’s a very simple way to think of it: carbohydrates provide 4 calories per gram, right? And fats provide 9 calories per gram, right? So right there, at the most basic level, we can see that, gram-for-gram, fats provide more than twice as much energy as carbohydrates do. We tend to think of calories as something bad, and something we want to limit, but really, “calorie” is a measurement of energy. Think of that next time someone on a very low-cal or low-fat diet complains that they’re tired all the time. But I digress...)

We’ve covered the critical connection between mitochondrial structure and function. Today, let’s look at what happens when—for whatever reasons—mitochondria fail to perform their critical OxPhos function. (And make no mistake, there are many reasons why this might happen. We’ve looked at one in detail—excessive production of reactive oxygen species [“free radicals”], and we took a more general look at other things that can potentially mess with mitochondrial function. We'll cover a few more down the line, when we get to potential causes of cancer.)


Recall from the first post on cellular energy generation that glycolysis converts glucose to pyruvate. When a cell has HEALTHY, FUNCTIONAL MITOCHONDRIA, the next step is that inside the mitochondria, the pyruvate will be converted to acetyl CoA, which will then enter the big, scary Krebs cycle, where it will end up generating eighteen times as much ATP as was generated through that measly, inefficient glycolysis. Okay, great. Awesome. There’s no debating that the vast majority of ATP produced in the body gets produced inside the mitochondria.

So what happens when a cell DOESN'T have healthy, functional mitochondria? What if, for some reason(s), there are too few mitochondria? Or maybe there are plenty, but they’re damaged somehow. Both of these situations would lead to a shortage of ATP, right? When your generators are on the fritz, or there aren’t enough of them, you don’t get enough power, right?
This is the best I could do for 
“hail Mary-type play.” 
(And we've gotta have some graphics, 
or this post will be BORING.)

This amounts to an “energy crisis” inside the cell. And nature—wily lil’ thing that it is—never has a single point of failure. There’s always a backup plan. Some kind of failsafe. A last-ditch, Hail Mary-type play that it saves for the very last seconds of the game before the buzzer sounds.

Instead of the cell dying from lack of ATP, the backup plan kicks in. And in the case of a cell that cannot generate any appreciable amount of energy through OxPhos, the backup plan is to revert to good ol glycolysis. But remember: glycolysis is inefficient. It generates very little ATP. So in order to make up for all the ATP that is not being generated by legions of mitochondria like it’s supposed to be, the cell has to engage in way more glycolysis than normal. 

Recall what Dr. Seyfried said:
  • “Mitochondrial function or OxPhos sufficiency cannot be normal in tumor cells that contain few if any mitochondria. Glycolysis and lactate fermentation would need to be upregulated in these tumor cells in order to compensate for the absence of OxPhos.”
  • “An upregulation of glycolytic genes would be needed to facilitate compensatory energy production through glycolysis when cellular respiration is deficient for protracted periods.”

And according to other researchers:
  • “One of the most prominent and universal metabolic alterations seen in cancer cells is an increase in the rate of glycolytic metabolism even in the presence of oxygen.”

And we know why, right? In order to keep up with its energy demands, and because its mitochondria are broken and can’t effectively metabolize fats or ketones, the cell requires enormous amounts of the only fuel substrate it can metabolize effectively: glucose. (*Ding, ding!* This is likely one of the reasons tumors create their own blood supply—in order to make sure they get all the glucose they need, they divert it to themselves, at the expense of other tissue in the body. Recall from a post a while back, this is angiogenesis.)


Mutant WHAT?

HEXOKINASE is an enzyme that converts glucose into something called glucose-6-phosphate. This is the very first step of glycolysis. The name “hexokinase” comes from the biochemical reaction this enzyme performs. It adds a phosphate group (PO3to glucose, which is a sugar molecule that has 6 carbon atoms. The general term for any 6-carbon sugar is “hexose”—hex for six, and we know things that end in “-ose” are sugars (for example, sucrose, lactose, and fructose). The “kinase” part of the word comes from the Greek kinesis, for movement, as in kinesthetics and kinesiology. So we can think of a kinase as moving a phosphate group from one molecule to another. You DO NOT need to know this to understand the metabolic theory of cancer. I’m just a word nerd and I like learning where the names come from. (Plus, knowing the etymology lets us know what something does even if we don’t know the chemistry. For example, a phosphatase enzyme removes a phosphate group. A hydrogenase adds hydrogen atoms.)

Here's what the reaction looks like: 

Notice that the phosphate group being added to glucose comes from ATP. (ATP has 3 phosphate groups. [ATP stands for adenosine TRI-phosphate], and the kinase enzyme breaks one of them off, leaving ADP [adenosine DI-phosphate -- 2 phosphates], and slapping that third phosphate group onto the glucose molecule. This is important. We will come back to it in a bit. Don't worry if you're not crystal clear on what's going on just yet. For now, just know that the chemical reactions of glycolysis are actually powered by ATP to start. They consume 2 ATP, so even though the process creates a total of 4 ATP, the net yield of glycolysis is 2 ATP. The key thing to know is that in order to end up with these 2 net ATP, we first need to use 2 ATP, and we need to get those starting ATPs from somewhere. From where? We'll get to that in a minute. (I wasn't kidding when I said glycolysis is inefficient. For every 4 steps forward, we take 2 back.)

So hexokinase turns glucose into glucose-6-phosphate (G-6-P) by adding a phosphate group to the number 6 carbon of a glucose molecule. As G-6-P starts to build up inside the cell, the increased concentration of G-6-P is supposed to shut off hexokinase. In biochem-speak, this is called negative feedback inhibition. Meaning, as the product of a reaction builds up, it puts the brakes on an earlier step of the reaction in order to slow it down or stop it. This makes sense, right? If we have G-6-P starting to accumulate, then we don’t need to keep making more of it, so we need some way to tell hexokinase to slow down.

One of the VERY INTERESTING things about some types of cancer cells is that they have an alternative form of hexokinasecalled hexokinase II or 2 (HK2). There are 4 different kinds of hexokinase. Different forms of the same enzyme are called “isoenzymes, and these different hexokinase isoenzymes have greater or less affinity for glucose. (This means certain forms are better at latching on to glucose than others.) Cancer cells express the HK2 enzyme way more than healthy cells do, and HK2 happens to have the highest affinity for glucose. 

Also, unlike the normal hexokinase found in healthy cells, this “mutant hexokinase” is NOT inhibited by G-6-P. The G-6-P builds up and builds up, but hexokinase 2 never gets the message to stop.
Are you shocked? I was! 
This stuff is NUTS!

First consequence: Glycolysis will keep going, and going, and going.

Second consequence: Lots and lots of pyruvate will build up. (Remember, the end product of glycolysis is 2 ATP and 2 pyruvate. We'll come back to this in the next post.)

Third consequence: The cell will seek more and more glucose to keep feeding the insatiable appetite of the mutant hexokinase.


If I just blew your mind, hang on. There’s more.

I mentioned that the hexokinase enzyme needs ATP to perform the first reaction of glycolysis. Well, this hexokinase 2, which cancer cells have a ton of, is seriously determined to get as much of that ATP as quickly as it can. See, the other forms of hexokinase kind of float around in the cytoplasm, waiting to do their thing. But HK2 anchors itself to the outside of the mitochondria, so it can suck up ATP right from the source, as soon as it’s generated. (Note: 100% of the mitochondria in cancer cells aren’t 100% broken. They are still generating some ATP, just not enough to successfully power the cell.)

As Dave Barry would say, “I am not making this up.” HK2 is an evil genius. Instead of hanging out in the cytoplasm like the other hexokinase isoenzymes, HK2 parks itself right at the outer mitochondrial membrane, geographically locating itself to strategically grab the ATP immediately, to keep stoking the flames of glycolysis. (Like I said, cancer cells have messed up mitochondria, but those mitochondria are still producing a little bit of ATP.)

The place where HK2 docks on the outer mitochondrial membrane is called the “voltage dependent anion channel,” or VDAC. (Dont worry about the scary name. I’m only specifying it so the quote below will make sense.)
  • “HK-2 bound to VDAC in the outer mitochondrial membrane has preferred access to ATP synthesized in the inner membrane and thus the rate at which glucose-6-P is synthesized by this ‘design’ is higher than that which would result if the newly synthesized ATP diffused into the cytosol and then to the active site of HK-2 bound to the outer membrane.” (Pedersen, 2007.)
  • “Binding of hexokinase II in this capacity almost completely prevents its product inhibition by glucose-6-phosphate, thus allowing the mitochondrial bound enzyme to initiate and then maintain a high glycolytic flux rate in tumors.” (Mathupala et al, 2010.)
  • “Mitochondrial bound hexokinase accounts for as much as 70% of the total cellular hexokinase in hepatoma [liver cancer] cells in contrast to the negligible hexokinase levels found on the mitochondria of normal liver cells. When the glucokinase (in normal liver) to hexokinase (in hepatoma) transition has taken place, a striking kinetic effect results with the isoenzyme in tumors exhibiting an approximately 100-fold higher affinity for glucose.” (Mathupala et al, 1997.)
  • “In contrast to other hexokinase isoforms, HK II harbors two active sites per enzyme moiety.” (Mathupala et al, 2006.) This is huge. To use a crude analogy, if we think of enzymes as garages (or "car park," or "car port" for you non-Americans), most hexokinase isoenzymes are like one-car garages for glucose; only one glucose molecule can park there at a time. But HK2 is like a two-car garage. It can handle twice the amount of glucose. HK2 has “the capacity to double the rate of formation of glucose-6-phosphate, relative to isoforms I and III. Therefore, it is not surprising that type II hexokinase is the isoform predominantly expressed in highly malignant tumors.” (Mathupala et al, 2010.)

Believe it or not, there’s even more going on with this special form of hexokinase. Stuff that will blow your mind many more times. But we’ve got to move on for now. We’ll address these other properties of HK2 a few posts down the line, when we explore the notion of cancer as an evolutionarily conserved protective mechanism. 



This is something I’d never heard of until I read Tripping Over the Truth, the book that inspired this series. There were some very nefarious political and academic shenanigans involved in suppressing this hexokinase stuff. Peter Pedersen, PhD, is one of the leading scientists who did an enormous amount of research in this area. Most of what we know today about HK2 came from his lab, his work, and that of his students and colleagues. One of these collaborators was tasked with finding a substance that could inhibit HK2. After all, if such a substance could be found, it might be possible to hit the "off switch" on this insatiable enzyme, and therefore, possibly take away a cancer cell's ability to metabolize its main fuel source...and therefore, starve it to death.

The scientist who dedicated herself to this and did the bulk of the work on it (at the expense of her personal life, according to the book) is Dr. Young-Hee Ko. Dr. Ko did find a compound that inhibits HK2. It’s called 3-bromo-pyruvate
  • “The pyruvate mimetic 3-bromopyruvate (3-BP) is generally presented as an inhibitor of glycolysis and has shown remarkable efficacy in not only preventing tumor growth, but even eradicating existant tumors in animal studies.”  
3-bromo-pyruvate has shown promise in killing cancer cells by putting the brakes on hexokinase II while sparing healthy cells:

This is important because hexokinase is the rate-limiting step of glycolysis. In biochemistry, we can think of a rate limiting step as kind of the on/off switch for a reaction. If the rate-limiting step doesn't "go," then the rest of the reaction doesn't go, either. (The same way statin drugs interfere with HMG CoA reductase, the rate-limiting step in the endogenous synthesis of cholesterol.) So if we have a compound that inhibits hexokinase 2, then glycolysis basically stops. And remember: we are talking about cells with broken mitochondria. They can't harness energy from fats or ketones effectively. They must use glucose. So if we have a compound that prevents these cells from using glucose (because we've inhibited the all-important rate-limiting enzyme), then these cells will starve (to death...we hope). 

If you are wondering what would happen to healthy cells under the influence of 3-bromopyruvate, good on ya! You should be wondering that right about now. We'll get to that in just a couple more posts. (As a little preview, though, keep in mind: those are healthy cells, with healthy mitochondria. Those cells can use fats and ketones. Hmmmm...)

Why have you never heard of 3-bromo-pyruvate? I dunno. Why had you never heard of the metabolic theory of cancer? Like I said: very nefarious shenanigansThat’s one of the nice things about Christofferson's book: the author covers some very basic biochemistry on how this all works, but he also gets into the forces that have guided the last century of cancer research—for better or worse—toward the somatic mutation theory, leaving the metabolic origins theory in the dustbin of history…until now. He gets into the personalities, the egos, the organizations, and their possible motivations. It’s a science book that reads as smoothly as a novel. Highly recommended. (Dr. Ko is hard to track down these days, but it looks like she's still in the field, and is now running her own company!)

The question you might be asking yourself now—and if you are, you definitely have your thinking cap on (and I want to hug you) is:  is hexokinase 2 a cause of cancer, or an effect? Does HK2 come first, and cause the rampant glycolysis, or is a cell's switching from other forms of hexokinase to predominantly HK2 (the nonstop glucose eating machine) a response to broken mitochondria that can’t metabolize fats and ketones, and therefore, in order for the cell to keep itself alive, it needs to engage in as much glycolysis as it can, ergo, it switches to HK2 as a protective mechanism? (I will share my theory with you a few posts down the line, but you can probably hazard a guess right now.)

So the predominance of HK2 is the second of the remaining cancer hallmarks we’ve been waiting to address. (The first was broken mitochondria.) Next time, we’ll look at the third one.

Preview: we now have tons of pyruvate building up inside the cell because of all the glycolysis. Our mitochondria are broken, and therefore, most of the pyruvate won't be converted to acetyl CoA. This pyruvate has one main fate: fermentation. Lots of it.

P.S. If you would like to read the full text of any of the articles I referenced in this post, and you cannot access them on PubMed, please contact me.

Remember: Amy Berger, M.S., NTP, is not a physician and Tuit Nutrition, LLC, is not a medical practice. The information contained on this site is not intended to diagnose, treat, cure, or prevent any medical condition.


  1. Enjoying your trip down the metabolic pathway.
    Good stuff. Keep it coming.

  2. Wow, nice work Amy! I need to make time to read Christofferson's book. And I look forward to your next post.

  3. Amy, this is great stuff. Please keep it coming. It is so rewarding to actually understand the processes going on in the body and to then relate that what I put in my mouth, how much of it goes in, and what I do with my body (movement & exercise). Very empowering information, and I am ever so grateful to you for presenting it the way you do. The evenings when your posts arrive in my in-box are the only time I break my rule of "no screens on after 8pm". Now I'll be late to bed!!

    1. :-) Well, I'm sorry I'm keeping you up late, but it's for a good cause, right? Thanks for reading!

  4. I've read Christofferson's book and your work expands on it nicely. Keep up the great posts. I have a newly diagnosed low grade lymphoma so am reading everything I can about the metabolic approach and ketogenic diets. At the least that should be a nice adjunct to standard treatment.

  5. Another fantastic post - please keep them coming. Thank you so much for doing this!

    1. Thanks so much, Marc! Really appreciate the positive feedback. Glad more people are starting to find this.

  6. Finding all your blogs tucked away is like being on an Easter egg hunt with plenty of eggs lying around all over

  7. Don't lose heart. You are doing the good thing. The truth about cancer (if it is the truth and I hope so) is better than almond cookies. This is refreshing Amy.

  8. So when OxPhos fails the cell goes back to fermentation. How does it know what to do?
    Where is the plan? The answer seems to be the cell encodes ALL information back to the year dot in its DNA.
    So the plans for fermentaion are still there, just waiting to be unlocked. This is beautiful. Nature builds on top of, does not destroy.

  9. Amy the great thing here is the ordinary person can interact with it, can think about it and truly understand, perhaps for the first time. The professionals use buzzwords and psycho-babble to keep the commoners out. You are like Robin Hood, you take from the rich and give to the poor.
    My hat is off.

    1. This might be one of the nicest comments I've ever received! Thank you, truly.

    2. I agree with Tim, Amy xx

  10. My guess would be HK2 or its progenitors is upregulated in accordance with the ancestral DNA encoding for it, kind of like rebooting after a disc error so you can pick an earlier version.
    The energy stress imposed on the cell would force selection for it and somehow unlock the encoding, otherwise the cell would die.

  11. Hi Amy,
    As soon as I read your description of HK2 as "mutant" I knew what the source of this was.
    I wrote to Travis Christofferson about this but didn't receive a reply.
    HK2 is only aberrant in HK4-dominant (glucokinase) tissues, and there are very few of these -- mainly the endocrine pancreas and the liver. That is because these organs functionally require the "sensing" of blood glucose levels.
    HK2 is the dominant hexokinase expressed in muscle tissue, and its behavior in attaching to the VDAC is NORMAL in myocytes for example. This should be unsurprising, since high power (i.e. rapid ATP generation) is often required in muscle, and this is accomplished by fermentation to lactate rather than respiration which is VO2-limited.
    Peter Pedersen did indeed first discover this behavior of HK2 in liver, which is the tissue he focuses upon in his cancer research. But it is only unusual primarily in liver -- not necessarily in other human tissues. It is a normal human hexokinase highly expressed in some tissues.
    HK4 has a much higher Km than the other human hexokinases, and is completely inappropriate for high-power requiring tissues such as muscle:
    So any type of human cell which is differentiated to have significant anaerobic ATP generation requirements normally (some of the time), and concurrently has no need for glucose sensing, will express HK1/2/3 predominantly.
    In a liver tumor, the malignant cell has abnormally developed this same requirement for its survival (due to damaged respiration/mitochondria), and adapted to mimic what is normal behavior in other organs. Since all of the cells are differentiated from the same genetic code (i.e. stem cells) this should not be surprising. It merely requires the turning on and turning off, or upregulation and downregulation, of various genes present in all of our cells, including hexokinase II.