Some Thoughts on Protein Stability

For the fourth time in recent memory, I find myself in a class where I’m told that proteins are not very stable. I find this jarring every time I hear it because, in the same breath, the honored professor says most protein folding reactions have Gibbs energies ranging between -5 and -8 kCal/mol (~ 20-35 kJ/mol for European protein scientists, see BNID 107069). A metabolic reaction with ∆G < -7 kcal/mol (-30 kJ/mol) is usually considered irreversible, so how is that protein biologists consider that same ∆G “not very stable?” It took me a bit to wrap my head around why protein and metabolism folks don’t see eye to eye on J. Willard Gibbs, so I thought I’d share the insight. Special thanks to Andy Martin for directing me towards the light.

I’ll take a second to reflect on the numbers. The standard change in Gibbs due to a reaction is just a measure of the reaction’s equilibrium; i.e. if I let it run till the end of time in a closed vessel, how many reaction substrates and products will I have in there. The equilibrium ratio between the product and substrate concentrations is called the equilibrium constant Keq = [P] / [S]. ∆G = - RT ln(Keq) is just a log-scaled version of the equilibrium constant (R is the gas constant, T is the absolute temperature). At a given temperature, ∆G determines the equilibrium constant and vice-versa. 

We biologists like to use incubators to regulate the temperature of our experiments. Mammals like to use homeostatic mechanisms to regulate the temperature of their bodies. Lots of organisms are found in characteristic temperature ranges, so much so that they are called mesophiles, thermophiles or psychrophiles according to their preferred temperature range. So, for the time being, let’s fix the temperature of our thought experiment to 25 C (298 K), since it’s a temperature that is relevant to a lot of life. In this case, a ten-fold increase in Keq (ten-fold more reaction products than substrates in equilibrium) results in a ∆G that is RT ln(10) = 1.4 kcal/mol = 5.7 kJ/mol smaller. That is, 1.4 kcal/mol more negative, more favoring the reaction products, more stable. In other words, every ~1.4 kcal/mol favors the reaction products over substrates by a factor of 10. If we let a protein with 5.6 kCal/mol stability fold forever, we’d find only 1 unfolded protein for every 10,000 folded ones. This, to me, sounds very very stable, so you can understand my confusion.

Turns out that my analogy to metabolism may have been the source of my confusion. In the context of metabolism, many reactions are reversible in the sense that they run in different directions in different cellular conditions. For example, most of the reactions in glycolysis are reversed when glucose is synthesized through gluconeogenesis. Many isomerization reactions like the one catalyzed by phosphoglycerate mutase (2-phosphoglycerate <=> 3-phosphoglycerate) have very small standard ∆Gs and are, therefore, easily reversed with small changes in the concentrations of reactants (try playing with reactant concentrations here). On the other hand, some metabolic reactions are irreversible (i.e. not easily reversed by changes in the concentrations of reactants) and the “reverse” reaction is actually catalyzed by a different enzyme using different co-factors. For example, the hexokinase reaction phosphorylating glucose to glucose 6-phosphate with ATP as the phosphate donor is a nearly irreversible reaction - it has a ∆G ~ -6 kcal/mol and would therefore require a roughly 10,000-fold change in the concentration ratio [glucose 6-phosphate] / [glucose] to proceed in the direction of ATP synthesis (presuming that the ATP and ADP concentrations are controlled homeostatically). So it’s not surprising that no ATP is synthesized from glucose 6-phosphate during gluconeogenesis - inorganic phosphate is produced instead. Indeed, phosphorylating glucose with inorganic phosphate is considered a different reaction than if ATP is used as the phosphate donor. And this makes tons of sense - the hexokinase reaction in glycolysis (glucose + ATP -> glucose 6-phophate + ADP) strongly favors the direction of phosphorylation while the phosphatase reaction in gluconeogensis favors the dephosphorylation (glucose-6p + water => Pi + glucose). 

When we think about protein folding, however, we take for granted that the folding reaction will proceed in the direction of, well, folding (some fascinating proteins like subtilisin notwithstanding). So when protein scientists say that -7 kcal/mol is not very stable, they must mean something different than what metabolism researchers mean when they say a reaction is “reversible.” Indeed, it would be strange if small changes in the concentration of a folded protein could change the net direction of a folding reaction - unfolded proteins are generally not active and have a tendency to aggregate. So therein lies the difference between metabolism and protein folding. In metabolism, some enzymatic reactions are regulated (e.g. hexokinase) while others mostly flow with the prevailing winds, catalyzing in the direction determined by metabolite concentrations. In contrast, protein production and degradation are regulated as a rule, not as an exception. Transcriptional and translational regulation of protein production is ubiquitous and regulated degradation through ubiquitination and other mechanisms is also … ubiquitous. So, while it would be odd for protein folding reactions to be “reversible” in the sense used in metabolism, protein folding and unfolding must be catalyzable on the energy scales available to cells in order for production and degradation to be regulated. It is, therefore, no surprise that the average stability of proteins is on the same order as the stability of ATP (∆G ~ -9 kcal/mol). After all, folding chaperones like GroES/GroEL and unfolding proteases like the proteasome couple the energy of ATP to the folding and unfolding of proteins.

Woodworkers Explanation of the Diffusion Equation

via Pops. Really interesting and fun application of the heat/diffusion equation to woodworking. Learned a lot.

4 weeks ago - 1

Voracious Worm Evolves to Eat Biotech Corn Engineered to Kill It - Wired Science

Oops! Via D. Arlow.

1 month ago - 1


CERN | Anna Pantelia

CERN is the European organization for Nuclear research and it’s considered the biggest particle physics experiment. it’s located at geneva and scientists, engineers and students from 113 nationalities are hosted. 29 September of 1954 was the ratification of this organization by 12 countries in Europe. Several important achievements have been made during experiments at CERN with the most important the development of World Wide Web. 

SoP - Scale of Work

(via sagansense)


The Beginning of Everything — The Big Bang

Has the universe a beginning or was it here since forever? Well, evidence suggests that there was indeed a starting point to this universe we are part of right now. But how can this be? How can something come from nothing? And what about time? We don’t have all the answers yet so let’s talk about what we know.


(via sagansense)


Santiago Ramón y Cajal

(via scientificillustration)

San Francisco's Class War, By the Numbers - The Nib - Medium

I always like it when there are numbers. What do people think? Does this fairly represent trends in the Bay? I’ve only been here for a hot minute.

2 months ago

via Rosalie Estelle

Single molecule dynamics of FtsZ in a FtsZ–FtsA cytoskeletal network. A small fraction of FtsZ monomers are white while the rest are red (dye trickery). You can see that individuals (white) don’t move even though the shape of the protein network re-arranges. 

Supplementary video 6 from Loose, M., & Mitchison, T. J. (2014). The bacterial cell division proteins FtsA and FtsZ self-organize into dynamic cytoskeletal patterns. Nature cell biology, 16(1), 38–46. doi:10.1038/ncb2885



Bombardier Beetle when threatened, sprays the attacker with a boiling hot mixture of caustic chemicals reaching 212° F (100° C). Even more impressive, the bombardier beetle can aim the poisonous eruption in the direction of the harasser.

The beetle itself is not harmed by the fiery chemical reaction. Using two special chambers inside the abdomen, the bombardier beetle mixes potent chemicals and uses an enzymatic trigger to heat and release them.

The foul concoction does burn and stain the skin. This defense proves effective against everything from hungry spiders to curious humans.

Such an amazingly sophisticated defense mechanism!