Photosynthesis · evolution · constraint
Why is Rubisco so slow?
The enzyme that feeds carbon into life looks badly designed. Its kinetics tell a different story.
Rubisco performs one of the decisive reactions on Earth: it takes inorganic carbon dioxide and feeds it into the chemistry from which cells build sugars, biomass, and eventually much of the living world. It is also a notoriously slow enzyme. Worse, it sometimes captures oxygen instead of CO₂. Plants compensate by making Rubisco in extraordinary quantities. The apparent contradiction is hard to miss: why should such an important enzyme be so poor at its job?
The gate through which carbon enters
Rubisco—ribulose-1,5-bisphosphate carboxylase/oxygenase—acts in the Calvin cycle of photosynthesis. It binds the five-carbon sugar RuBP. When CO₂ enters the reaction, an unstable six-carbon intermediate forms and splits into two molecules of 3-phosphoglycerate. Carbon has crossed from the atmosphere into metabolism.
Oxygen creates the troublesome branch. The product 2-phosphoglycolate must be recycled through photorespiration, consuming energy and releasing some of the carbon that had already been fixed. Rubisco therefore sits at a competition between two gases: CO₂, which feeds photosynthesis, and O₂, which diverts it.
Nor is the enzyme fast. Typical plant Rubiscos turn over only a few CO₂ molecules per active site each second. The usual verdict is therefore severe: slow catalysis, imperfect discrimination, enormous abundance. Rubisco seems like a relic evolution never managed to repair.
But that verdict assumes that speed, affinity, and specificity can all be improved independently.
The enzyme has fewer degrees of freedom than it seems
To test that assumption, Yonatan Savir, Elad Noor, Ron Milo, and I compared Rubiscos from organisms living in very different carbon environments: photosynthetic bacteria, cyanobacteria, algae, and C₃ and C₄ plants. Each enzyme can be described by several kinetic parameters: its maximal rate of carboxylation, its apparent affinity for CO₂ and O₂, and its ability to discriminate between them.
If these traits could evolve independently, natural Rubiscos should scatter through a high-dimensional space of possible kinetics. They do not. The parameters are strongly correlated by simple power laws. In the complete four-parameter subset, one principal component accounted for about 91% of the measured variation.
One correlation is especially revealing. Faster Rubiscos tend to have poorer discrimination between CO₂ and O₂; more generally, catalytic velocity rises together with a loss of CO₂ affinity. The desirable corner—very fast, very selective, very high-affinity—appears difficult to reach.
This changes the evolutionary question. Natural selection is not choosing a point in an open space. It is moving along a narrow path cut out by molecular chemistry and structure.
Different environments choose different points
A constraint alone does not tell us where an enzyme should sit on the path. The environment does. A Rubisco working at low CO₂ benefits strongly from affinity. One surrounded by concentrated CO₂ can surrender affinity and move toward higher catalytic speed.
C₃ plants expose Rubisco to relatively modest CO₂. C₄ plants first concentrate carbon around the enzyme and therefore operate in a very different local atmosphere. Cyanobacteria use carbon-concentrating mechanisms as well. These organisms should not carry the same Rubisco. They should occupy different positions on the same kinetic landscape.
That is what the comparison found. When net photosynthesis is optimized along the measured kinetic constraint, the predicted optimum shifts with intracellular CO₂. Rubiscos from the different physiological groups lie close to the positions expected for their own carbon environments.
The result also changes the usual emphasis on oxygen. In this constrained landscape, the dominant control is the tradeoff between carboxylation velocity and CO₂ affinity. Oxygen matters, but the local CO₂ concentration largely determines where the optimum lies.
Slow does not mean badly adapted. A slow enzyme may be near the best point available in its environment.
Rubisco and the invention of C₄ photosynthesis
There are two ways to improve a constrained machine: change the machine, or change the world immediately around it. C₄ photosynthesis takes the second route. Its biochemical machinery raises CO₂ near Rubisco, suppressing the competing oxygenation reaction and allowing a faster, lower-affinity enzyme to become advantageous.
The enzyme and its environment therefore form one evolutionary system. Carbon-concentrating mechanisms alter the optimal Rubisco; Rubisco kinetics alter the value of concentrating carbon. What looks like an imperfection of a molecule may make sense only when the surrounding machinery is included.
Why is Rubisco difficult to improve?
Rubisco has long been an obvious target for crop engineering. If carbon fixation limits growth, why not make the enzyme faster or stop it from reacting with oxygen? The kinetic landscape gives a caution: changing one trait tends to move the others. A mutation that improves speed may pay in affinity or specificity.
This does not mean that Rubisco is unimprovable. It means that improvement requires escaping, bending, or exploiting the constraint rather than turning a single independent knob. Changing carbon concentration, assembly, regulation, or the architecture of the enzyme may matter as much as changing an active-site rate.
The interesting object is not the optimum alone. It is the shape of the possible.
Constraint comes before optimality
Rubisco is useful beyond photosynthesis because its kinetic diversity can be measured across species and compared with a physical fitness function. It makes visible a general fact that is easy to forget when speaking loosely about biological design.
Evolution cannot choose among all imaginable proteins. It explores proteins that can fold, assemble, catalyze, and survive within the coupled restrictions of molecular physics. Selection then acts inside that accessible set.
Evolution optimizes within the possible.