It is always delightful to me when people look more closely at well-understood pathways and make new discoveries that change our understanding. This is one such occasion.
Basically, it appears that they discovered a feed-forward loop in the synthesis of phosphoenolpyruvate, the penultimate product of glycolysis. Though this wasn't mentioned in the news article, I wonder if this helps to explain the Warburg effect.
Anyways, this doesn't exactly jump out to me with applications since the research seems so early stage. True, discovery of novel pathways implies that inhibition of these pathways could be useful for treating cancer. But until some research is actually done in that vein, I'm hesitant to read too deeply. I'll review the manuscript itself, but I don't think we know for sure that this "cancer-only" pathway is not at play in important, rapidly-growing non-cancerous tissues, which would be my main concern.
OK, I've read the manuscript now. Very interesting work. I didn't realize this initially, but they are also saying that the transient phosphorylation of PGAM1 is a mechanism by which PEP is converted to pyruvate at a meaningful rate!
Reason for the exclamation point: the conversion of PEP -> pyruvate is is considered the last of 3 rate-limiting steps in glycolysis. They propose an end-run around this rate-limiting step. In their model, PEP donates its phosphate to PGAM1. Thus, PEP becomes pyruvate. Normally I wouldn't expect this to be a meaningful way to produce pyruvate--but their data appear to support the notion that PGAM1 becomes dephosphorylated at a high enough rate that this process can repeat itself fast enough to sustain pyruvate production. Oh, and phosphorylated PGAM1 isn't just a temporary phosphate holder--they hypothesize (relying on other studies) that it helps shunt carbons into biosynthetic pathways (which you need if you are a proliferating cancer).
It's not every day that someone figures out how living cells bypass a canonical rate-limiting step in the fundamental metabolic process shared by all living things.
It took me about 20 minutes to google all of the terms in your response and figure out why this is really important, but I feel like I might actually get some of this this now and feel the excitement.
Thanks for this as the paper itself was more or less unapproachable for me. Biohacking still seems to me like this great big unexplored space filled with new terms and exciting ideas that I'm just beginning to get acquainted with.
>Anyways, this doesn't exactly jump out to me with applications since the research seems so early stage.
The first thing I thought was that starving the cells of sugar could be an effective treatment. Maybe feeding patients something like the ketogenic diet for epileptics (ie lots and lots of heavy cream and not much else esp. not many carbs) would help. A quick pub med search turns up a little bit of work in this area but there might be room for more.
Having just spent 3 days at an annual NCI meeting this article sort of made sense to me. What's fascinating is that the state of the art is moving so fast in this field that many are predicting within a few years we'll be able to deliver individualized chemo-therapies. General internists and even oncologists can't keep up with the rate of progress.
As an aside, it's also a space where fast moving small startups could make a big difference
The key is to individualize to the level of the specific cancer, not necessarily to the patient with the cancer (Osler would punch me for saying this).
I say this because, for example, leukemia is not one cancer; it is a constellation of cancers with similar features. However, if you take one specific cancer from this group (chronic myelogenous leukemia of the Bcr/abl type), you can treat it with imatinib. The survival at 5 years with imatinib is 89%--an incredible success. I'm saying this to point out that we already have individualized treatments for (a vanishing minority of) cancers.
As we grow to understand the molecular and genetic basis for each individual cancer, we will, one by one, be able to target these and turn them into chronic diseases like hypertension or hypercholesterolemia. Perhaps, one day, we will even cure them. The current article being discussed actually hints at a more global phenomenon, and may produce tools that we can use as add-ons to the individualized cancer treatments that we will be giving patients.
I'll just repeat the two things that I always like to say in this type of discussion:
1: We need to target the molecular basis for disease.
2: We need to remember the HAART model when treating cancer, which involves targeting multiple enzymes necessary for tumor growth and proliferation (or at least target different parts of the same enzymes) in order to fight drug resistance.
Well as a programmer I won't dispute your assertion, but what I heard at a keynote from a very prestigious scientist is that we will actually start targeting the patient, not just the specific cancer. He cited some examples where certain known therapies that starve tumors are disastrous in a very small percentage of patients due to certain genetic differences. It was like hearing science fiction. This is certainly a great time to be in this field.
The scientist and I don't disagree; I think we just have a semantic difference. If the cancer is genetically different from the expectation, then it requires a different treatment. So if you have a disease that looks just like CML but doesn't have a Bcr/Abl fusion tyrosine kinase, imatinib is probably not what we'd want to use. I would call that molecule-centric, and s/he might call it patient-centric. Either way, the wording doesn't matter so long as you're treating the patient with the most appropriate therapy.
I see, well I'll certainly take your word for it. I really need to read more of this literature as I've been working with these folks quite a bit, and hardly know the difference between a gene and a protein.
Are you doing bioinformatics? If so, it's a great excuse to learn more about the field, and learning more about the field will make you care more about the work you're doing--a feed-forward loop. At least, when I started doing medical genetic research, that's how things worked for me.
sort of, I work in description logics, they are used to build medical terminologies. NCI has used them over the years in cancer genomics. A good amount has rubbed off just hanging out with these folks but recently I've become keen to learn more.
To really get the most you can out of the article, you might want to check out Lehninger Principles of Biochemistry, by Nelson and Cox. It's a wonderful book! Glycolysis is covered in chapter 14.
Only a layman here, but it wounds like this could inhibit tumor growth in a broad spectrum of cancers. I wonder, if a drug inhibiting this alternative metabolic pathway were developed, would cancers simply evolve a different pathway?
Maybe (I don't study this) but at least some of this pathway is also used in embryonic cells, so it's pretty fundamental. More often, cancer cells will evolve resistance to whatever inhibitor molecule is targeting them, i.e. mutate so the ligand doesn't bind as well.
Basically, it appears that they discovered a feed-forward loop in the synthesis of phosphoenolpyruvate, the penultimate product of glycolysis. Though this wasn't mentioned in the news article, I wonder if this helps to explain the Warburg effect.
Anyways, this doesn't exactly jump out to me with applications since the research seems so early stage. True, discovery of novel pathways implies that inhibition of these pathways could be useful for treating cancer. But until some research is actually done in that vein, I'm hesitant to read too deeply. I'll review the manuscript itself, but I don't think we know for sure that this "cancer-only" pathway is not at play in important, rapidly-growing non-cancerous tissues, which would be my main concern.