I have spent the last 12 years working with and around cancer data: visualizing it, integrating it, analyzing it. Until recently, I hadn’t questioned the way that biological processes in cancer are explained in text books, in scientific papers or in online videos. I’ve relied on a habit developed during high school (inspired by a brilliant biology teacher): that of visualizing biological interactions in the cell by role playing. For example, to learn the Krebs cycle (see picture below), I would start by pretending that I am Acetyl-CoA. When oxaloacetate approaches, we grab on to each other and become citrate. I need to let go of the CoA that I have in my hand in order to bind to oxaloacetate… Now that I am citrate, I need to find a NAD+ to give it a H+ and release a CO2… And so on…
It may not be very scientifically accurate to think about the Krebs cycle this way but at least it was more fun than trying to memorize it. There is a major flaw in this strategy — it assumes that the molecules know what they’re “supposed” to do next.
Biological Interactions as Constraints
Biology is not a set of disconnected processes in spite of how it is commonly taught, studied and represented. The way the Krebs cycle is normally drawn looks clean, well organized. But there’s a lot missing — the diagram creates the impression that the molecules are isolated from everything else that’s going on in the cell, that just the right number of NAD+ are available to help move the cycle along.
In reality, the inside of a mitochondria (where the Krebs cycle happens in Eukaryotes) is a mess of molecules. And in that chaos, the right molecules and proteins DO find each other, they DO react and they DO create the molecules for the next step the cycle. How? Because there are constraints, which evolved over time, which make it so.
The expression of enzymes and the flow of molecules, atoms and ions through membranes of organelles in the cell are the mechanism by which the necessary constraints are created. Their presence increases the probability that the cellular machinery will behave in the direction that increases the generation of ATP (chemically stored energy for the cell) and removes toxic molecules.
Lets look at the Krebs cycle from a constraint lens: to start, Acetyl-CoA is just going about its business — randomly wriggling due to its brownian motion. Until it comes across an electro-magnetic force that limits that motion — a constraint created by a random encounter with oxaloacetate and the enzyme citrate synthase (represented below). The Acetyl-Coa+Oxaloacetate+Citrate Synthetase complex forms because its total free energy is lower than the sum of the energy of its individual components (entropy goes up with energy transfered from high to low energy systems). Similarly, citrate forms because its free energy is lower than that of the enzyme ensemble.
That of course begs the question — how did just the right protein evolve to be in the right place at the right time with the right configuration to catalyze the right reaction? The honest answer is —we don’t know. But we can hypothesize that if a protein happened to evolve which created a constraint (no matter how weak) such as to catalyze that reaction faster than it would naturally occur, then it’s feasible to assume that cells with the genetic code to create such constraint would have had better evolutionary success than those that didn’t and could copy that information onto their descendents.
Generalizing is a dangerous thing to do but since we’re still in the spirit of thought experiments, lets generalize anyway. The generalization we’re going to make is about biological processes in cancer cells.
Cancer is a disease different from all others in the sense that it’s a disease caused by cells that replicate faster and consume more oxygen and glucose than the cells around them. These are cells that (at least temporarily) have more fitness than healthy cells (per modern evolutionary theory definition of fitness). We know a few things about how cells achieve this “fitness” level — and the constraints that are supposed to prevent it. In fact, chemotherapy works because it blocks cell’s access to sugar… It makes you feel nauseous and lose your hair because it does this with both cancer and normal cells in the hope cancer cells are killed faster since they consume so much more sugar than normal cells. This higher sugar consumption also explains why PET scans work.
We know that cancer is a genetic disease. That means that whatever drives a cancer cell, it is created by the mutations that a parent cell suffered which lead to the constitutive (=continuous) activation of processes inside
that cell which prompted it to make rapid copies of itself. For example, BRAF is a well known gene in melanoma because through genetic sequencing of melanoma cells from many patients, the same mutation in the BRAF protein (called V600E) keeps being discovered. In fact V600E can be found in about 50% of all cutaneous melanoma cases (src).
Genes like BRAF are called “oncogenes”, which is a misnomer because it gives the impression that the genes’ evolutionary path led them to be cancer causing. This is a distorted way to think about BRAF. Instead, normal copies of BRAF have a constraint in them: they won’t trigger normal cell growth until they receive a signal. That signal is external to the cell and that is a very important detail: extra-cellular signals, distributed via the circulatory system, evolved to create constraints on when and how many cells proliferate in multi-cellular organisms. Mutated BRAF copies do not have that constraint and therefore can proliferate without much regard for extra-cellular signalling.
Which leads us to the final question: can’t we cure cancer by re-creating the constraints that are eliminated by mutations? Well, yes we can! I chose the BRAF example intentionally because in 2011 the FDA approved a drug called Vemurafinib (whose name comes from comes from “V600E mutated BRAF inhibition”) which specifically blocks cells with BRAF V600E copies from proliferating.
BRAF, however, is just one of many different mutations that remove from cells the constraints necessary for whole-organism-controlled proliferation. Also, single cells and the mechanisms therein are difficult to control because medicine and our scientific instruments still operate at an organism-sized level, not at the molecular level. However, promising technologies are starting to give us the power to operate at the micro level. But that’s a topic for another post :-)