Photosynthesis involves capturing light energy and, most often, converting it to chemical energy stored as reduced carbon. It is the source of food, fuel, and fiber and there is a resurgent interest in basic research on photosynthesis. Plants make excellent use of visible light energy; leaves are ideally suited to optimize light use by having a large area per amount of material invested and also having leaf angles to optimize light utilization. It is thought that plants do not use green light but in fact they use green light better than blue light under some conditions. Leaves also have mechanisms to protect against excess light and how these work in a stochastic light environment is currently a very active area of current research. The speed at which photosynthesis can begin when leaves are first exposed to light and the speed of induction of protective mechanisms, as well as the speed at which protective mechanisms dissipate when light levels decline, have recently been explored. Research is also focused on reducing wasteful processes such as photorespiration, when oxygen instead of carbon dioxide is used. Some success has been reported in altering the path of carbon in photorespiration but on closer inspection there appears to be unforeseen effects contributing to the good news. The stoichiometry of interaction of light reactions with carbon metabolism is rigid and the time constants vary tremendously presenting large challenges to regulatory mechanisms. Regulatory mechanisms will be the topic of photosynthesis research for some time to come.
Photosynthesis describes a broad array of processes by which light energy is captured and converted to chemical energy that is used in biological organisms. The energy is often stored on carbon as a change in oxidation state from +4 (carbon dioxide) to 0 (sugars). Bacteria, algae, and plants can carry out this process. However, in algae and plants the capacity for photosynthesis comes from a bacterial endosymbionts (chloroplasts) and so photosynthesis is primarily a bacterial process, although some archaea can also carry out photosynthesis. Photosynthesis is the source of nearly all carbon in the biosphere. Photosynthesis research has enjoyed a resurgence with government support supplemented by private foundations, especially for funding of the international RIPE project (Realizing Increased Photosynthetic Efficiency, https://ripe.illinois.edu/). Additional sources about emerging research in photosynthesis can be found in a recent special issue of The Plant Journal , which has a number of reviews on diffusion of CO2 through stomata and the mesophyll of leaves, among other topics. Also, more information can be had in any volume of the book series Advances in Photosynthesis and Respiration Including Bioenergy and Related Processes (https://www.springer.com/series/5599 See for example the latest volume, 45, covering algal research, ). Here I will offer descriptions and opinions on a subset of the current ‘hot topics' in plant photosynthesis research. There are many important emerging areas of research on photosynthesis in cyanobacteria not touched on here. One thread that runs through the topics in this paper is regulation of the many processes that constitute photosynthesis.
Generally, photosynthesis makes use of what is called visible light, between 380 and 740 nm, between ultraviolet and infrared in the electromagnetic spectrum. The definition of photosynthetically active radiation (PAR) is more restricted, just 400 to 700 nm. Several pigments can be used for photosynthesis but by far the most common is chlorophyll. Chlorophyll extracted into a solvent has a relatively sharp peak of absorption in red light and a broader peak in blue light. Between is what is called the green gap although it turns out the gap is not that deep.
Plants use two different photosystems (PS I and PS II). The photosystems likely arose in different bacterial lineages and came together in cyanobacteria. The combined electron transport from PS II through a cytochrome b6/f complex to PS I allows conversion of light energy to reducing power (typically NADPH) and ATP simultaneously. Because there are different photosystems monochromatic light in the range of 680 to 700 nm is used inefficiently, but if light of wavelength less than 680 nm is combined with light of 700 nm then 700 nm light is more effective. This phenomenon is called the Emerson enhancement effect. Now, a similar argument has been made that light of wavelength longer than 700 nm can be used providing there is shorter wavelength light available simultaneously .
At the blue end of the spectrum it is known that light of wavelength shorter than 400 nm can drive photosynthesis. In Figure 5 of McCree , light of 350 nm was still 30% as effective as the maximum red light effectiveness. Most instruments designed to measure light available for photosynthesis cut on abruptly at 400 nm and off at 700 nm, both of which are oversimplifications of the true nature of PAR.
Most organisms use light throughout the visible spectrum much more uniformly than the strong red and blue absorption spectra of chlorophyll in a solvent would suggest. One reason is that chlorophylls are attached to proteins and exist in different environments, altering their absorption spectra. This makes both the red and blue peaks much broader, so much so that plants can use green light absorbed by chlorophyll. Cyanobacteria can fill in this green gap using specialized pigments called phycobilins (phycocyanin, allophycocyanin, and phycoerythrin). These pigments act as antennas and are arranged in protein structures called phycobilisomes that funnel green light to the photosystems.
Plants and algae do not have phycobilisomes even though, it is believed, chloroplasts were once free-living cyanobacteria. Nevertheless, plants make excellent use of green light, contrary to common thinking. In addition to the absorption band broadening described above, green light can have a much longer pathlength through a leaf than red or blue light. While red light and blue light will be absorbed quickly, sometimes only in the top layers of a leaf , green light can penetrate deeper into the leaf. Green light scattering will increase its pathlength and the chance for absorption by chlorophyll. This also causes backscattered light from a leaf to appear green. Under very bright light, red and blue light will be absorbed in the top layers of a leaf and excess red and blue photons will be dissipated there while green light penetrates to lower layers, where the chloroplasts are not saturated with light. This is why, under bright light, green light can be more effective than either red or blue light .
One way that plants do not fill the green gap is by carotenoid absorption acting as antennas for plant chlorophylls the way the phycobilisomes do for cyanobacteria. This common explanation, found in most biology textbooks, is simply wrong. Although blue light is more strongly absorbed than either green or red, it is less efficient per photon than either green or red light [4,7] (not just because of the extra energy per photon of blue light). This is because, rather than acting as antennas, the blue-absorbing carotenoids act as shades, absorbing light but not passing it to chlorophyll. Carotenoids are important for regulating energy flow and quenching dangerous side products formed by light absorption. Light energy can be efficiently passed from some carotenoids to chlorophylls in isolated reaction centers  but it has long been known that many carotenoids do not . What is more, the absorption spectrum of carotenoids does not extend into the green very much further than chlorophyll absorption (Figure 1). Many carotenoids are not positioned to pass light energy to chlorophyll and so depress the efficiency of blue light. Even allowing for the stronger absorption of blue light, green light (e.g. 550 nm) can be more efficiently used for photosynthesis than blue light (e.g. 450 nm) in many situations (Figure 1).
Absorption of light by chlorophyll a and b, adsorption of light by leaves, and relative quantum efficiency as a function of wavelength.
Managing light energy in leaves
Light energy drives photosynthesis so the ability to capture light is crucial to the success of photosynthetic organisms. For plants this means efficient display of chlorophylls in leaves. Many leaves are very thin so that a maximal leaf area can be displayed for a given investment of resources. The specific leaf area (SLA, the area of leaves for a given leaf dry weight, or its inverse, leaf mass per area, LMA) is a major predictor of plant growth rate [10–13], often a better predictor than the rate of CO2 uptake per unit leaf area.
Specific leaf area will depend on leaf density, volume, and thickness . Leaves can have variable volume of air space [15,16], too little airspace could reduce the diffusion of carbon dioxide while too much airspace could indicate an inefficient cellular architecture. There have been significant advances in understanding molecular signals that affect leaf cell morphology, providing a road map for improving leaf architecture to maximize photosynthesis [17,18].
Managing light energy in canopies
The canopy of trees can display a large number of leaves, but many lower leaves will be in the shade of higher leaves. This can also happen in crops. The amount of leaf area for a given ground area, called the leaf area index, can be as much as ten although three to five is more common. On the other hand, photosynthesis typically only makes use of about one quarter of full sunlight, i.e. photosynthetic rates are often  75% of maximum by 500 µmol m−2 s−1 as has been published for field-grown soybeans  or leaves of oak trees exposed to full sun . Many plants (for example Arabidopsis ) can respond to increasing light up to 500 µmol m−2 s−1 even when grown in just 100 µmol m−2 s−1. Light absorbed beyond 500 µmol m−2 s−1 is dissipated, essentially wasted. This can be useful if the light would otherwise go to competitors or deleterious if the light would otherwise go to lower leaves on the same plant. There have been attempts to improve crop yields by reducing the amount of chlorophyll in leaves so that upper leaves will be more transparent, allowing light to get to the lower leaves [20,23,24]. When chlorophyll is reduced uniformly through the canopy, the loss of photosynthesis in the top leaves was not fully compensated by the increased photosynthesis by the now better illuminated lower leaves. More work will be needed to distribute resources optimally through the canopy. Many plants have upper leaves displayed at many angles. The amount of light hitting a leaf will depend on the cosine of the angle between the rays of the sun and the surface of the leaf. Leaf angles at the tops of trees often approach a spherical distribution, that is the leaves could cover the various angles of a sphere. This allows light to penetrate into the canopy. This reduces the light intensity on the sun-exposed leaves but increases it for understory leaves. On the other hand, lower leaves can be arranged parallel to the ground so that whatever light does make it through the upper leaves will be maximally captured ([25,26] and unpublished data in Figure 2).
Leaf orientation and inclination.
Light: enough is enough
Most photosynthesis in leaves happens when there is considerably less light than is available in direct sunlight. This allows less investment in the enzymes needed for very rapid photosynthesis that could make use of full sunlight, allowing reduced SLA, which is associated with increased plant growth. However, this means that there will be occasions when there is more light than can be used in photosynthetic reactions. This can lead to photodamage. There are a number of mechanisms that allow photosynthetic organisms to protect themselves against excess light damage. Recent advances have been made in understanding the structure and function of the orange carotenoid protein (OCP) in cyanobacteria . These proteins can move a carotenoid from one location to another within the protein and get attached to the phycobilisomes that harvest light. This causes the light energy to be harmlessly dissipated when it is in excess of what can be used by the photosystems [28–31].
In plants, which lack phycobilisomes, several methods are available for dissipating light energy, some safer than others. Many of these are assessed by analyzing chlorophyll fluorescence yields, especially in response to perturbations such as a flash of light five times brighter than full sunlight. The safest method for dissipating light energy is called energy-dependent quenching (qE) although the quenching here refers to the effect on chlorophyll fluorescence, not quenching of the incoming light energy. Following advice from W.W. Adams III I will reserve ‘quenching' for fluorescence and use the term dissipation for the fate of excess photons.
Energy-dependent quenching depends on energy in two ways [32,33]. First, light-driven electron transport causes an accumulation of protons in the lumen of the thylakoid. The lowered pH stimulates violaxanthin deepoxidase, which converts violaxanthin into zeaxanthin. The interconversion of violaxanthin and zeaxanthin is called the xanthophyll cycle. The presence of zeaxanthin potentiates qE. The second thing that has to happen is the low luminal pH that results in protonation of a PS II protein called PsbS . Protonated PsbS in the presence of zeaxanthin causes changes in the light harvesting apparatus so that the energy of absorbed photons is dissipated as heat (radiationless decay) before it gets to PS II.
Current research focuses on the dynamics of regulation of photosynthetic electron and proton transport in a highly variable environment [35–37]. When the photon flux falling on a leaf increases abruptly, photosynthetic processes take a finite time to take full advantage . In one study it was estimated that over 20% of potential photosynthesis was lost because of the need for induction of photosynthetic processes upon a shade to sun transition . Mechanisms that dissipate excess light in bright light can be slow to reverse when a leaf is shaded. Speeding the relaxation of qE can increase overall plant yield [40,41].
Dissipation of light energy by qE develops and goes away more slowly than light can first exceed, and then become limiting again for photosynthesis. During the time between when light is in excess and when qE can safely dissipate the excess energy, photosynthesis can be damaged by the uncontrolled energy. Davis et al.  showed that the initial response to an abrupt increase in light is formation of the electrical component of the proton motive force and that this can lead to photodamage as a result of high speed light changes such caused by sunflecks .
If excess energy gets to PS II it can cause one of the proteins, D1, to be damaged. Fortunately, plants have an active repair cycle . The PS II repair cycle and qE can both help plants cope with fluctuating light environments . Photosynthetic electron flow from PS II to PS I can be slowed because of the energetic requirements for translocating protons at the cytochrome b6/f complex . When electron transport is limited in this way electrons can build up on PS II, leading to damage. The buildup of electrons at PS II can be assessed by the chlorophyll fluorescence quenching parameter qL . High qE and low qL will combine to protect PS I from overreduction. PS I overreduction may be more damaging than PS II overreduction because there is not a PS I repair mechanism like the PS II repair mechanism. Therefore, it is common to see evidence of excess light damage at PS II but rare to see damage at PS I, but when PS I is damaged it has more consequence for the plant than does PS II damage [48,49]. Plants lacking the proton gradient regulation 5 (PGR5) gene show no phenotype when grown under constant light but die under strongly fluctuating light because of damage to PS I .
Rubisco: a complicated enzyme
Carbon dioxide is first fixed by the enzyme rubisco (C4 plants like corn and Crassulacean Acid Metabolism plants have chemical preconcentrating mechanisms but carbon dioxide first fixed by that mechanism is released for fixation by rubisco). There are many non-rubisco pathways for photosynthesis [51,52], especially among anaerobic bacteria; in terms of amount of carbon these are very minor but there have been many attempts to find more efficient engineered pathways of carbon fixation . Improvements in rubisco have been attempted. In plants these are limited by the fact that the large subunit of rubisco is coded for in the chloroplast genome and transformation of the chloroplast genome is much more difficult than transformation of the nuclear genome. Nevertheless, significant advances in genetic engineering of plastids have been made [53–55].
Other ways to find improved rubisco relies on expressing it in bacteria  but this has been exceedingly difficult for the plant enzyme because it has not been possible to express it and have it fold properly in bacteria . However, as a result of progress in understanding the chaperone proteins it has now been possible to express rubisco and all of the necessary chaperonins so that functional rubisco can be expressed in E. coli . This opens the door to engineering to improve the catalytic properties of rubisco. Among the properties that need improvement is the kcat, which is between 2 and 10 s−1 per site . This may be average when compared with all enzymes  but it is very slow for an enzyme in primary metabolism and as a result plants must invest vast amounts of nitrogen in the rubisco protein, significantly increasing the fertilizer costs of agriculture.
Another property of rubisco under current study is the need for post translational modification. A CO2 molecule (not the CO2 that will be fixed) is covalently bound to a lysine as a carbamate followed by addition of Mg2+ or Mn2+ (Figure 3) . This is referred to as activation of rubisco and depends on an ATP-requiring AAA+ protein called rubisco activase , which removes molecules that bind slowly but tightly to rubisco [62,63]. There is some debate whether this is regulatory or represents a deficiency that could be corrected to engineer more efficient photosynthesis. Combining these two ideas, it could be that trapping of inactive rubisco by tight binding of inhibitors to the active site was a problem solved by activase but then activase evolved into a regulatory role. Cyanobacterial rubisco does not suffer as much from inactivation by tight binding inhibitors and yet many cyanobacteria have activase-like proteins that have a different function, possibly helping rubisco aggregate for inclusion in carboxysomes .
Rubisco (E) post-translational activation and the role of rubisco activase (Rca).
If regulation of rubisco activity by activase is not ideal for agricultural plants, or is no longer ideal in today's high CO2 atmosphere, then engineering activase could be an effective method for increasing crop yields. Rubisco activase is regulated by alternative splicing during gene expression , availability of ATP , redox , and phosphorylation . Measurements show that rubisco is inactivated in low light , high CO2 , and high temperature [71,72]. While deactivation at low light or high CO2 can easily be ascribed to a regulatory adjustment of rubisco activity when it is in excess, the case for rubisco deactivation at high temperature being regulatory is less settled [71,73]. Significant advances have been made in finding [74,75] or engineering  thermostable forms of rubisco activase.
Rubisco makes mistakes
Rubisco catalysis is not completely specific for either substrates or products. For example, in addition to its primary product 3- phosphoglyceric acid , rubisco also makes a small amount of pyruvate . This is slightly less efficient but provides a ready source of pyruvate for fatty acid synthesis and isoprenoid synthesis in chloroplasts. Rubisco will also use an alternative substrate, O2 instead of CO2 . This reaction is not quite so harmless. One of the resulting molecules, 2-phosphoglycolate (2-PG), is a very potent inhibitor of the essential enzyme triose phosphate isomerase [80–82]. The reactions in plants that metabolize 2-PG are called photorespiration because they involve O2 uptake and CO2 release although instead of producing energetic molecules for the plant it consumes both ATP and reducing power, even more than when CO2 is fixed. Roughly speaking, for every three steps forward in CO2 fixation during photosynthesis, plants take one step backward in photorespiration [83,84]. Increasing CO2 decreases oxygenation and photorespiration but it will continue to be a major inefficiency of photosynthesis for a long time. The costs of photorespiration to agriculture is quite large .
Recent research aimed at developing methods to overcome the inefficiency of photorespiration has focused on alternative pathways for metabolizing 2-PG . A major issue in photorespiratory metabolism is the release and refixation of ammonia. Ammonia fixation in the glutamine synthetase — glutamate oxoglutarate aminotransferase (GS-GOGAT) cycle for photorespiration greatly exceeds ammonia fixation needed for de novo protein synthesis. The nitrogen is used to make glycine and serine which are then metabolized eventually back to 3-phosphoglycerate to reenter the Calvin–Benson cycle. However, the glycine and serine can also be used directly for protein synthesis, in which case they can be seen as products of photosynthesis [87,88]. There are several ways in which photorespiratory metabolism has become intertwined with other metabolic pathways . Bloom and Lancaster  propose that this provides an evolutionary pressure to regulate photorespiration at a significant rate, but the contrary view is that photorespiration is a ‘lemon' even if lemonade (improved nitrogen availability) was made. The frequent evolution of C4 plants  like corn, which have much reduced rates of photorespiration is typically invoked to indicate that plants can easily find mechanisms for nitrogen acquisition that do not depend on photorespiration.
A variety of alternative photorespiratory pathways have been engineered into plants. However, sometimes the pathway as proposed may not be what causes improved plant performance. An early, exciting report was made by Kebeish et al. . They engineered a cyanobacterial pathway for glycolate metabolism consisting of a glycolate dehydrogenase (to make glyoxylate inside the chloroplast) followed by a glycolate carboligase (to make tartronic semialdehyde) and then tartronic semialdehyde reductase (to make glycerate). However, introducing the glycolate dehydrogenase alone caused a significant increase in photosynthetic rate [92,93]. In a similar vein, South et al.  reported that engineering glycolate dehydrogenase and malate synthase into tobacco (Figure 4) resulted in a stable transformant that performed better than the untransformed plant including under field conditions. This very encouraging finding suggests it will be possible to engineer more efficient photosynthesis. However, the mechanism is likely not precisely the pathway as drawn.
Alternative pathway number 3 to photorespiration of South et al. [
Pathway 3 of South et al. (Figure 4) is energetically much more efficient than photorespiration (Table 1). The ATP requirement is reduced nearly to one half while the reducing power changes from a net loss to a net gain because of the loss of 2 CO2’s. These CO2’s would need to be refixed at a cost of NADPH and ATP. One way to analyze the energetics is a calculation of the ATP and NADPH that would be needed for the two pathways plus the energy needed to refix the lost CO2. This is shown in Table 1 as starting and ending with ribulose 5-phosphate. South's pathway 3 is more efficient in terms of reducing power than photorespiration if the reducing power at glycolate dehydrogenase can be recaptured, but photorespiration is more efficient than South's pathway 3 in terms of ATP. However, Pathway 3 requires many fewer steps and no diffusion from organelle to organelle. It is difficult to know the benefit of such a simplified system.
|Pathway .||CO2 .||ATP .||NADPH eq .|
|South pathway 3||−2||−2||+2|
|Ru5P to Ru5P|
|South pathway 3||0||−8||−2|
|Pathway .||CO2 .||ATP .||NADPH eq .|
|South pathway 3||−2||−2||+2|
|Ru5P to Ru5P|
|South pathway 3||0||−8||−2|
Costs are calculated per oxygenation or assuming additional CO2 fixation to compensate the carbon loss of each pathway. Negative sign indicates consumption or loss from the plant. The NADPH eq could be NADPH, NADH, or other unknown electron acceptors. The lower half presents estimates of costs that include refixation of released CO2, starting and finishing at ribulose 5-phosphate (Ru5P).
Finally, it is worth noting that the pathway shown in Figure 4 does not accommodate a role of photorespiration in N metabolism. The greater growth of plants with this pathway is evidence against the hypothesis that photorespiration is beneficial because of its role in nitrogen metabolism 
Determining why the genetic changes identified by South et al. increased photosynthesis and yield may lead to a better understanding of the regulation of photosynthesis and yield and allow even more targeted approaches to dealing with photorespiration. 2-PG was not measured but it may have been decreased by the introduction of the glycolate dehydrogenase, just as found by Nölke et al. . Engineering photorespiration is likely to be of significant interest for the foreseeable future.
Carbon: active uptake
Most plants rely on diffusion for CO2 transport to rubisco. This saves energy but this kind of photosynthesis, called C3 photosynthesis, leaves plants vulnerable to photorespiration and would be difficult in aquatic situations where CO2 diffusion is very slow. Cyanobacteria have protein bounded microcompartments , of which the best known is the carboxysome, which contains all of the cell's rubisco. The carboxysome is part of a mechanism for active uptake of bicarbonate and conversion to CO2 at rubisco. Some researchers are trying to put carboxysomes into plants to overcome photorespiration and reduce the amount of rubisco needed for a given rate of photosynthesis . There has also been significant interest in putting the C4 preconcentrating mechanism into C3 crops, especially into rice .
Underlying many of the issues above is regulation. How are the various components of photosynthesis regulated so that everything works together harmoniously? This is a very important emerging research area in photosynthesis. A surprising finding was that photosynthesis could be divided between a state in which the capacity for regenerating the CO2 acceptor ribulose 1,5-bisphosphate (RuBP) is either (1) in excess and thus RuBP-saturated rubisco kinetics determined the CO2 response of photosynthesis or (2) limiting and so RuBP regeneration capacity determines the shape of the CO2 response of photosynthesis . This either-or limitation is not intuitive for biologists but in fact is known to occur in many situations . On the other hand, RuBP regeneration can be limited by either light availability and use or by the operation of the Calvin–Benson cycle. Increased capacity for Calvin–Benson cycle enzyme activity can lower the ΔG for ATP synthesis. This allows the lumen of the thylakoid to be regulated at a higher pH, reducing qE. This increases the efficiency of light use  so that when RuBP regeneration limits the photosynthetic rate, either light or Calvin–Benson cycle enzyme activity, or both simultaneously, may be rate setting factors.
Two other factors related to regulation of photosynthesis are important: (1) there is very little capacitance in the system, most carbon metabolite intermediates have pool sizes much <1 s (ie, would be consumed in under 1 s if production were abruptly halted, determined as flux divided by pool size), and (2) the stoichiometry is fixed, for example, the products of electron transport must be used in strict ratios. Excess ATP cannot compensate for a lack of NADPH (to a first approximation).
The stoichiometry issue is a very important area of current research into the regulation of photosynthetic processes. For example, C4 metabolism requires extra ATP but not NADPH (5 ATP/2 NADPH required) while the alternative, photorespiration, requires extra energy with only a slightly higher ATP/NADPH requirement (3.5 ATP/2 NADPH compared with 3 ATP/2 NADPH required in C3 photosynthesis). Current thinking is that normal linear electron flow produces 2.57 ATP per 2 NADPH because 14 protons are required for three ATPs in the ATP synthase . This leaves an ATP deficit for the Calvin–Benson cycle and even more so when photorespiration is present. For example, if oxygenation happens at a rate or 0.3 times the rate of carboxylation then the required ratio would be 3.11 ATP per 2 NADPH for a deficit of 0.27 ATPs or 1.26 protons per one NADPH (or pair of electrons in linear electron flow). This ATP deficit can be made up by a number of mechanisms but the one most frequently discussed is cyclic electron flow involving PS I. Two pathways are normally discussed, one that involves NDH (NAD(P)H dehydrogenase) [103,104], which was thought to take electrons from NAD(P)H and recycle them back to the cytochrome b6/f complex while transporting as many as four protons per electron , which can then be used to make ATP. Recent evidence summarized by Peltier et al.  indicates that the true electron donor may be ferredoxin. The total activity of NDH may be limited and another cyclic electron flow pathway has been demonstrated. It has been thought to involve the PROTON GRADIENT REGULATION 5 protein (PGR5) and PGR5-like protein (PGRL1) and is sensitive to antimycin A . The relative importance of these two pathways has been controversial, at least in part because the methods for measuring cyclic electron flow, especially in intact systems, are controversial. A recent report with C4 plants indicates that the NDH pathway may supply the cyclic electron flow for ATP synthesis while the PGR5 pathway contributes in some way to a sink for electrons downstream of PS I but does not contribute ATP . Cyclic electron flow is also important in heat stress [108,109] and in a glucose-6-phosphate shunt . Regulation of cyclic electron flow could involve sensing of ATP status .
Other methods for increasing ATP synthesis relative to NADPH include the water-water, or Asada, cycle in which electrons from water are eventually donated back to O2 to reform water. This pathway was also called pseudocyclic photophosphorylation. It is now mostly considered a method for detoxifying reactive oxygen species formed at PS I [112,113]. A second method involves export of reducing power from the chloroplast to the mitochondrion and making use of the mitochondrial electron transport chain to make ATP. This can be accomplished by the malate valve, export of malate and import of oxaloacetate can transfer NADH, or the non-phosphorylating glyceraldehyde 3-phosphate enzyme  that would allow GAP export and PGA import to the chloroplast transferring NADPH from the chloroplast to the cytosol [115,116].
Will increased photosynthesis increase crop yield?
As the source of reduced carbon for the plant it is almost axiomatic that improved photosynthesis will increase crop yield. The increased growth and yield of plants in elevated CO2 confirms the importance of photosynthesis. Nevertheless, some argue that improved photosynthesis alone is not the best way to improve crop yields . To use a car analogy, it is important to pay attention to steering, tires, and even brakes on a race car, but a strong engine will be essential to a winning race team. It is important to understand the regulation of photosynthesis that integrates the chloroplast into the host metabolism. Research into regulation of photosynthesis to fit with the needs of the rest of the plant is in early stages and holds great promise for improving plant performance.
Fundamental pathways for electron and proton transport and carbon flow have been established; current research focuses on how these pathways are regulated.
Leaves take advantage of most of the visible light spectrum. Green light, rather than being mostly unused, is sometimes better used that blue light.
Regulation of light use is critical to take best advantage of available light while keeping deleterious light damage under control.
The speed of processes that adjust light use and CO2 use are currently being studied to optimize photosynthesis in highly variable environments.
The fixed stoichiometry of the various photosynthetic processes and extreme range of time constants make regulation essential. The importance of regulatory processes means that it can be hard to see past the regulation to observe the underlying constraints.
The author declares that there are no competing interests associated with this manuscript.
Currently my research on photosynthesis comes from the US Department of Energy Award DE-FG02-91ER20021 and partial salary support is provided by Michigan AgBioResearch.
T.D.S. wrote this review.