C3 Photosynthesis

Why C3 photosynthesis matters

• It feeds most of the world. Wheat, rice, and soybean are all C3 — together they supply roughly 50 percent of human caloric intake. Improving C3 efficiency is the central engineering target for closing the projected 2050 food gap; engineering RuBisCO and reducing photorespiration could lift global yields by an estimated 15 to 30 percent (RIPE Project, Long et al.).
• RuBisCO is the most abundant protein on Earth. Conservative estimates put global RuBisCO mass around 0.7 gigatons, with each m² of leaf containing 1 to 5 grams. Plants make so much because each enzyme catalyzes only ~3 carboxylations per second — they brute-force throughput with bulk.
• Carbon flow constrains every terrestrial ecosystem. C3 photosynthesis fixes about 56 gigatons of carbon per year on land — roughly 7 times annual fossil-fuel emissions. Net primary productivity in temperate forests, boreal regions, and most agricultural land is set by C3 efficiency under prevailing CO2, water, and light.
• Photorespiration is the largest single loss in plant carbon balance. Sharkey's measurements show photorespiration consumes 25 to 50 percent of gross carbon assimilation in C3 leaves at 30 degrees Celsius and current atmospheric O2. Engineering bacterial bypasses (Maurino, Kebeish 2007; South, Cavanagh, Ort 2019) raised tobacco yield by ~40 percent in field trials by short-circuiting the C2 cycle.
• Light reactions and Calvin cycle are tightly coupled but separable. The light reactions in thylakoid membranes produce ATP and NADPH; the Calvin cycle in the stroma consumes them. Decoupling the two — running the Calvin cycle on chemical reductant alone — is a long-standing biotechnology goal for cell-free biomanufacturing of carbon-based chemicals.
• RuBisCO activase consumes ATP just to keep RuBisCO active. RuBisCO is inactivated by tight binding of its own substrate RuBP and inhibitors like 2-carboxyarabinitol-1-phosphate; the chaperone RuBisCO activase uses ATP hydrolysis to release them. Activase is heat-labile and starts failing above 35 degrees Celsius, which is why C3 yields collapse in heat waves before stomatal closure dominates.
• Atmospheric CO2 history is tracked by C3 vs C4 isotope discrimination. C3 plants discriminate against 13C by ~20 per mil; C4 by only ~4 per mil. Soil carbonate and tooth-enamel delta-13C records reveal global expansion of C4 grasslands ~7 million years ago as CO2 dropped — a planetary signal of photosynthetic mode switching.

Common misconceptions

• The Calvin cycle requires light directly. The cycle itself is not light-driven — it consumes ATP and NADPH produced by the light reactions, and most cycle enzymes are pH- and redox-regulated by ferredoxin/thioredoxin so they shut down in the dark. Older textbooks called it the "dark reactions"; modern usage avoids this because the cycle effectively halts at night.
• RuBisCO is a single enzyme everywhere. Form I RuBisCO (most plants, cyanobacteria) is a hexadecamer of 8 large + 8 small subunits. Form II (some bacteria) is just dimers of large subunits and tolerates higher O2. Form III (archaea) is hexamers and is involved in nucleotide salvage rather than carbon fixation. Form IV is RuBisCO-like but does not fix CO2 at all. Engineering attempts target Form I in plants while drawing on Form II for higher specificity factors.
• C3 plants cannot survive in hot climates. They can — wheat grows in Sudan, rice in tropical Asia — but their water-use efficiency is much lower than C4. C3 plants must transpire ~500 to 1000 g water per g carbon fixed; C4 manages with ~250 to 350. In water-limited tropics, C4 dominates the canopy understory or grasslands; C3 trees still dominate moist tropical forests.
• Photorespiration is purely wasteful. It is energetically expensive, but it also dissipates excess reducing power under high light, scavenges damaging hydrogen peroxide, and generates glycine and serine. Mutants with photorespiration knocked out (Arabidopsis glycolate oxidase mutants) die in air but grow fine at 1 percent O2 — proving the C2 cycle is essential when O2 is around.
• RuBisCO has been optimized by evolution as well as it can be. The carboxylation-versus-oxygenation tradeoff appears to define a Pareto frontier — improving k_cat for CO2 lowers specificity for CO2 over O2, and vice versa. Tcherkez and Farquhar (2006) modeled this as a chemistry-imposed limit. Recent directed-evolution work (Wilson, Mueller-Cajar, Whitney 2018) has pushed past it modestly in cyanobacteria, but not in plant Form I.
• All C3 plants respond identically to elevated CO2. They do not. FACE (Free-Air CO2 Enrichment) experiments at Duke and SoyFACE show wheat yield gains of ~15 percent at 550 ppm but soybean closer to 20 percent and rice closer to 10 percent. Genetic variation in stomatal density, leaf area index, and source-sink balance produces species- and even cultivar-level differences in CO2 fertilization response.

How the Calvin-Benson cycle works

The cycle has three phases: carbon fixation, reduction, and regeneration. Phase 1 is the RuBisCO reaction itself: ribulose-1,5-bisphosphate + CO2 yields two 3-phosphoglycerate. The enzyme first uses a Mg2+-coordinated active site to enolize RuBP, exposing a carbanion at C2 that attacks CO2, then hydrolyzes the unstable 6-carbon intermediate. Phase 2 phosphorylates each 3-PGA to 1,3-bisphosphoglycerate using ATP, then reduces it to glyceraldehyde-3-phosphate (G3P) using NADPH. Per turn of the cycle, 3 CO2 enter and one G3P (3 carbons) exits as net product — the rest must be recycled. Phase 3 regenerates RuBP through a remarkable carbon-shuffling mosaic involving aldolase, transketolase, sedoheptulose-1,7-bisphosphatase, and ribose-5-phosphate isomerase, ultimately phosphorylating ribulose-5-phosphate to RuBP using ATP.

Six turns of the cycle consume 18 ATP and 12 NADPH and produce one glucose (after G3P is shuttled out and combined). Regulation is tight: thioredoxin reduced by ferredoxin from the light reactions activates four cycle enzymes (RuBisCO activase, fructose-1,6-bisphosphatase, sedoheptulose-1,7-bisphosphatase, phosphoribulokinase, and a regulatory subunit of GAPDH) only when the chloroplast is illuminated. Stromal pH rises from ~7 in the dark to ~8 in light, which favors RuBisCO activity. Magnesium efflux from the thylakoid lumen accompanies proton pumping and activates Calvin cycle enzymes that require Mg2+. The system is therefore a light-on switch implemented through redox state, pH, and Mg2+, not through gene expression — fully reversible within seconds.

C3 vs C4 vs CAM photosynthesis

Feature	C3	C4	CAM
First fixed product	3-PGA (3 carbons)	Oxaloacetate (4 carbons)	Oxaloacetate (4 carbons), at night
Initial fixing enzyme	RuBisCO (mesophyll)	PEP carboxylase (mesophyll)	PEP carboxylase (nighttime)
Spatial separation	None — all in mesophyll	Mesophyll then bundle sheath (Kranz anatomy)	None — same cell
Temporal separation	Day only	Day only	CO2 fixed at night, sugars made by day
ATP per CO2	3 ATP + 2 NADPH	5 ATP + 2 NADPH	5-6 ATP + 2 NADPH
Photorespiration	~25% reactions at 30 °C	Suppressed (CO2 ~10x at RuBisCO)	Suppressed
Water use efficiency	~500-1000 g H2O / g C	~250-350 g H2O / g C	~50-100 g H2O / g C
Optimal climate	Cool, moist, moderate light	Hot, sunny, moderate water	Arid, hot deserts
Examples	Wheat, rice, soybean, oak	Maize, sugarcane, sorghum, switchgrass	Pineapple, agave, cacti, orchids
Share of biomass	~95% globally	~3-5%	< 1%

Famous experiments and case studies

• Calvin and Benson 1948 — the lollipop apparatus. A flat glass disc of Chlorella illuminated steadily, then injected with 14CO2 for pulses of 5, 15, 30, or 60 seconds before the cells dropped into hot methanol to halt metabolism. Two-dimensional paper chromatography revealed the carbon-labeling sequence — 3-PGA first, then sugar phosphates radiating outward. The full cycle was published by 1957.
• Andrews and Lorimer 1973 — RuBisCO oxygenase activity. Showed RuBisCO directly catalyzes the reaction of RuBP with O2 to give one 3-PGA + one 2-phosphoglycolate, providing the molecular basis for photorespiration. Reframed the enzyme as a partial failure mode rather than just a carboxylase.
• FACE experiments at Duke 1996 and SoyFACE 2001. Open-air CO2 fumigation rings raising local CO2 to 550 ppm in field crops, providing the most realistic measurement of CO2 fertilization. Wheat yields increased ~15 percent, but the protein content dropped ~6 percent — illuminating a fertilization-versus-quality tradeoff.
• South, Cavanagh, Ort 2019 — synthetic photorespiration bypass. Engineered tobacco with bacterial glycolate dehydrogenases and malate synthase from Chlamydomonas, redirecting 2-phosphoglycolate metabolism into the chloroplast. Field trial yield increase of ~40 percent compared to wild-type — the strongest demonstration that photorespiration is a real yield ceiling for C3 crops.
• Long, Marshall-Colon, Zhu 2015 — meta-analysis of photosynthesis improvement targets. Modeled engineering RuBisCO, optimizing canopy architecture, accelerating photoprotection recovery (NPQ relaxation), and tuning the antenna size. Estimated combined potential yield gain of ~30 to 60 percent in C3 crops by mid-century.