Respiration in Plants

Every living cell needs a steady supply of energy. Building proteins, pumping ions across membranes, moving materials from one cell to another and simply staying alive at rest all cost energy. That energy is stored in food molecules such as glucose, but a cell cannot spend a glucose molecule directly the way it spends ATP. Respiration is the process by which the energy locked inside food is released in small, controlled steps and repackaged into ATP, the ready cash of the cell.

The compound that a cell breaks down to release this energy is called the respiratory substrate. Carbohydrates, usually glucose, are the most common respiratory substrate. Fats, proteins and organic acids can also be used, and cells switch to them when carbohydrates run short, for example in a germinating oilseed or in a leaf that has used up its sugar during the night. In plants the glucose used in respiration often comes from sucrose, which is first split into glucose and fructose by the enzyme invertase, and from stored starch, which is broken down by starch-hydrolysing enzymes.

Plants have no lungs, no gills and no circulating blood to move gases around, and yet every living cell in a plant respires. This works for four reasons. Each part of a plant looks after its own gas exchange, so there is little need to transport gases from one organ to another. The overall demand is low, because a great deal of a woody plant is dead or dormant tissue and only the living cells respire actively. The living cells lie close to the surface, so the distance a gas has to diffuse is short. And inside a photosynthesising cell the O₂ released by photosynthesis is used up by respiration in the same cell, while the CO₂ released by respiration is fed straight back into photosynthesis. Stomata in leaves, lenticels in older stems, and the loosely packed parenchyma with its interconnected air spaces are enough to handle whatever exchange is left.

One more idea separates respiration from burning. If you set fire to sugar, the same total amount of energy comes out, but it comes out in one rush as heat and light and the cell can trap none of it. Respiration is a stepwise oxidation controlled by a long series of enzymes, so the energy leaves the substrate in small packets. Part of it is captured in ATP and in reduced coenzymes such as NADH, and the rest escapes as heat. This stepwise, controlled release is the whole design of the pathway, and it is the reason respiration takes so many steps to do something a flame does in one.

Glycolysis is the first stage of respiration and it happens in every living cell, plant, animal, fungus and bacterium alike. It takes place in the cytoplasm, needs no oxygen and no mitochondrion, and ends with one six-carbon glucose molecule split into two three-carbon molecules of pyruvic acid. It is also called the EMP pathway, after Embden, Meyerhof and Parnas, who worked out the scheme.

The pathway runs through ten enzyme-controlled steps, and it is easiest to follow as a story in four parts. First, glucose is trapped and primed. Glucose is phosphorylated by the enzyme hexokinase, using one ATP, to give glucose-6-phosphate. This is isomerised to fructose-6-phosphate, which is phosphorylated a second time by phosphofructokinase, using a second ATP, to give fructose 1,6-bisphosphate. Notice that the cell has spent two ATP before it has earned anything at all. Second, the six-carbon sugar is split. Fructose 1,6-bisphosphate breaks into two three-carbon compounds, dihydroxyacetone phosphate and 3-phosphoglyceraldehyde, usually shortened to PGAL. The dihydroxyacetone phosphate is then converted into PGAL as well, so from this point on every remaining step happens twice, once for each PGAL.

Third, the pay-off begins. PGAL is oxidised and at the same time phosphorylated, using inorganic phosphate taken from the cytoplasm rather than from ATP, to form 1,3-bisphosphoglyceric acid, shortened to BPGA. The hydrogen removed during this oxidation is picked up by NAD⁺, which becomes NADH + H⁺. BPGA then hands one of its phosphate groups directly to ADP to make ATP, leaving 3-phosphoglyceric acid behind. Because the phosphate comes straight off the substrate and not from a proton gradient, this is called substrate-level phosphorylation. Fourth, the finish. 3-phosphoglyceric acid is rearranged to 2-phosphoglyceric acid, a molecule of water is removed to give phosphoenolpyruvate or PEP, and PEP transfers its phosphate to ADP, again by substrate-level phosphorylation, forming pyruvic acid.

Now do the accounting for one glucose. Two ATP were spent in the priming steps, at hexokinase and at phosphofructokinase. Four ATP were made in the pay-off phase, two from the two BPGA molecules and two from the two PEP molecules. The net gain is two ATP. Two molecules of NADH + H⁺ are also formed, one from each PGAL. The carbon end products are two molecules of pyruvic acid, which still hold most of the energy that was in the original glucose. That is why glycolysis is described as a partial oxidation: no CO₂ is released, no oxygen is used, and the real harvest is still to come.

Pyruvic acid stands at a fork in the road. What happens to it next depends on whether oxygen is available and on the kind of cell it is in. If oxygen is absent, the cell takes one of the fermentation routes. Fermentation is the incomplete breakdown of glucose, and its real job is not to make ATP but to keep glycolysis from stalling.

Here is the problem fermentation solves. Glycolysis needs a supply of NAD⁺ to oxidise PGAL. Every time the pathway runs, it converts NAD⁺ into NADH. A cell holds only a small pool of NAD⁺, so if nothing regenerates it, that pool is used up within seconds, glycolysis stops, and with it stops the only ATP the cell can still make. Fermentation regenerates NAD⁺ by dumping the hydrogen from NADH onto pyruvic acid, or onto something made from pyruvic acid.

Alcoholic fermentation happens in yeast and in plant tissues starved of oxygen, such as roots sitting in waterlogged soil. Pyruvic acid is first decarboxylated by the enzyme pyruvic acid decarboxylase, which removes a carbon as CO₂ and leaves acetaldehyde. Acetaldehyde is then reduced to ethanol by alcohol dehydrogenase, and it is this second step that turns NADH back into NAD⁺. The end products are ethanol and CO₂.

Lactic acid fermentation happens in some bacteria, including the ones that set curd, and in our own skeletal muscle during hard exercise when the blood cannot deliver oxygen fast enough. Here pyruvic acid itself is reduced straight to lactic acid by the enzyme lactate dehydrogenase, again converting NADH back to NAD⁺. No CO₂ is released along this route, which is the quickest way to tell the two fermentations apart.

Both routes are poor at releasing energy. Less than seven per cent of the energy stored in glucose is released, and not even all of that little is trapped in the high-energy bonds of ATP. The net ATP from fermentation is still only the two ATP earned back in glycolysis. Fermentation is also risky for the cell, because ethanol and lactic acid are toxic if they accumulate. Yeasts poison themselves once the alcohol concentration in their medium reaches about thirteen per cent, which is why naturally fermented drinks cannot become stronger than that without distillation. In plants, roots that stay waterlogged and anaerobic for long enough build up ethanol and are eventually damaged.

If oxygen is available, pyruvic acid does not stay in the cytoplasm. It moves into the mitochondrion, and the rest of respiration happens there. Aerobic respiration finishes the job glycolysis left half done: it strips every remaining hydrogen off the carbon skeleton, releases the carbon as CO₂, and eventually hands those hydrogens and their electrons to oxygen while trapping the energy released as ATP.

The first event inside the mitochondrion is the link reaction, an oxidative decarboxylation that takes place in the matrix. Pyruvic acid meets a large enzyme complex called pyruvate dehydrogenase, which needs Mg²⁺ as a cofactor along with NAD⁺ and coenzyme A. One carbon leaves as CO₂, hydrogen is removed and picked up by NAD⁺ to give NADH + H⁺, and the two-carbon fragment that remains is attached to coenzyme A as acetyl CoA. Remember that one glucose gave two pyruvic acid molecules, so this reaction runs twice, giving two CO₂, two NADH and two acetyl CoA per glucose. The link reaction is a bridge and not a cycle, because nothing is regenerated in it.

Acetyl CoA now enters the tricarboxylic acid cycle, also called the Krebs cycle or the citric acid cycle, which also runs in the matrix. The two-carbon acetyl group condenses with the four-carbon oxaloacetic acid, along with a molecule of water, to form the six-carbon citric acid; the enzyme is citrate synthase and coenzyme A is released to go and fetch another acetyl group. Citric acid is rearranged to isocitric acid. Isocitric acid then undergoes oxidative decarboxylation to the five-carbon alpha-ketoglutaric acid, losing one CO₂ and giving one NADH. Alpha-ketoglutaric acid undergoes a second oxidative decarboxylation to the four-carbon succinyl CoA, losing a second CO₂ and giving a second NADH. Succinyl CoA is converted to succinic acid, and this is the one step in the cycle where a high-energy phosphate bond is made directly: GTP is formed and then passes its phosphate to ADP to give ATP, which is substrate-level phosphorylation again. Succinic acid is oxidised to fumaric acid, and here the hydrogen acceptor is not NAD⁺ but FAD, so FADH₂ is produced. Fumaric acid takes up a molecule of water to become malic acid, and malic acid is oxidised to oxaloacetic acid, giving the third NADH. Oxaloacetic acid is now free to accept another acetyl group and the cycle turns again.

Count one turn carefully, because this is the most common place to slip. One turn of the TCA cycle releases two CO₂ and produces three NADH, one FADH₂ and one ATP. Water is taken in when acetyl CoA and oxaloacetic acid condense to form citric acid, and again when fumaric acid is hydrated to malic acid, and oxaloacetic acid is regenerated at the end. One glucose sends two acetyl CoA molecules into the cycle, so the cycle turns twice per glucose. Add the link reaction to the cycle and you can see the carbon balance: each pyruvic acid gives off three CO₂ in all, one in the link reaction and two in the cycle, so a whole glucose gives off six CO₂.

All the NADH and FADH₂ collected so far are only carriers; they hold energy but the cell cannot spend them. That energy is cashed in on the inner mitochondrial membrane, which is folded into cristae to increase its surface area. Embedded in this membrane is the electron transport system, or ETS, a chain of four protein complexes together with two mobile carriers, arranged in order of increasing attraction for electrons so that electrons flow down the chain in one direction.

Electrons from NADH enter the chain at complex I, NADH dehydrogenase. Electrons from FADH₂ enter lower down, at complex II, succinate dehydrogenase, which is the one enzyme of the TCA cycle built into the inner membrane instead of floating free in the matrix. Both complexes pass their electrons to ubiquinone, a small lipid-soluble carrier that moves about within the membrane. Ubiquinone hands them on to complex III, the cytochrome bc₁ complex. From there a small protein called cytochrome c, which sits loosely on the outer surface of the inner membrane, ferries electrons one at a time across to complex IV, the cytochrome c oxidase complex, which contains cytochromes a and a₃ along with two copper centres. Complex IV finally passes the electrons to oxygen, the terminal electron acceptor, which takes up protons at the same time and becomes water. This is exactly why the whole pathway stops without oxygen: with nothing to accept the electrons at the end, the chain backs up, NADH cannot be reoxidised, and the TCA cycle grinds to a halt for want of NAD⁺.

Because FADH₂ feeds its electrons in after complex I, those electrons drive fewer proton-pumping steps than electrons from NADH do, and that single fact is the reason FADH₂ yields less ATP than NADH.

As electrons travel down the chain, the complexes use the energy released to pump protons out of the matrix into the intermembrane space. This builds up a proton gradient across the inner membrane, so the intermembrane space becomes more acidic and more positively charged than the matrix. The gradient is a store of potential energy, like water held behind a dam. According to the chemiosmotic hypothesis put forward by Peter Mitchell, it is this gradient, and not any direct chemical link, that drives ATP synthesis.

The protons return to the matrix through ATP synthase, also called complex V. It has two parts. F₀ is embedded in the membrane and forms the channel through which protons cross. F₁ is the knob-like head that projects into the matrix and carries the catalytic site that joins ADP to inorganic phosphate. For each ATP made, two protons pass through F₀ down the electrochemical gradient. Because this ATP synthesis is coupled to the oxidation of NADH and FADH₂ in the chain, it is called oxidative phosphorylation. Set that against substrate-level phosphorylation in glycolysis and the TCA cycle, where a phosphate group is simply handed over from a substrate to ADP with no membrane and no gradient involved at all.

Now add everything up for one molecule of glucose oxidised completely by aerobic respiration. Glycolysis gives 2 ATP and 2 NADH. The link reaction, running twice, gives 2 NADH. The TCA cycle, turning twice, gives 2 ATP, 6 NADH and 2 FADH₂. So the running totals are 4 ATP made directly by substrate-level phosphorylation, 10 NADH and 2 FADH₂.

Now cash the coenzymes in, taking one NADH oxidised inside the mitochondrion as worth 3 ATP and one FADH₂ as worth 2 ATP. Eight of the ten NADH are made inside the mitochondrion, two in the link reaction and six in the two turns of the cycle, and they give 24 ATP. The remaining two NADH were made out in the cytoplasm during glycolysis, and the inner mitochondrial membrane will not let NADH itself through; its reducing power has to be ferried in by a shuttle, and that ferrying costs something, so each of these two is counted as 2 ATP, giving 4 more. The 2 FADH₂ give 4 ATP. Add the 4 ATP made directly and the total comes to 36 ATP per molecule of glucose. Set that beside fermentation, which nets only 2 ATP from the very same glucose, and it is clear why aerobic respiration is worth all the machinery it needs.

You will sometimes meet the figure 38 instead. It comes from valuing the two glycolytic NADH at 3 ATP each, as though they had been made inside the mitochondrion in the first place, which gives 30 ATP from the ten NADH and 38 in total. The chemistry behind the two numbers is identical; they differ only over what the shuttle costs. Work with 36 and know where 38 comes from, so neither number surprises you.

In any case the total is a theoretical maximum, not a measurement. It holds only if a list of assumptions is true, and in a living cell not one of them really is. The calculation assumes that the pathways run one after another in a tidy, orderly sequence, with nothing bypassed. It assumes that the NADH made in the cytoplasm during glycolysis is carried into the mitochondrion and oxidised there rather than being used up in the cytoplasm for something else. It assumes that none of the intermediates of the pathway is pulled out to build another compound, when in fact intermediates are constantly being withdrawn. And it assumes that glucose is the only substrate being respired and that nothing else joins the pathway partway along. A real cell breaks all four of these assumptions at every moment.

There is a second reason the number is soft. How much ATP a cell actually gets from one NADH depends on how many protons the chain pumps and on how many of them leak back across the membrane without passing through ATP synthase. The whole system is coupled but not perfectly efficient. So treat 36 as the figure to quote and to calculate with, while understanding what it is: an upper limit worked out under ideal conditions, not a count of ATP molecules in a working mitochondrion.

It also helps to keep track of what happens to the atoms, because that makes the balance sheet make sense without any diagram. Glucose has six carbons. None leaves during glycolysis. Two leave in the link reaction, one from each pyruvic acid. Four leave during the two turns of the TCA cycle. That is six CO₂ out for six carbons in. Six molecules of O₂ are taken in, and every one of them ends up in water after accepting electrons at complex IV. The hydrogen that came off the substrate at all those dehydrogenation steps is the hydrogen you find in that water.

It is tempting to picture respiration as a one-way demolition line, but that picture is wrong. The respiratory pathway is amphibolic, meaning it acts as a catabolic pathway and an anabolic one at the same time. It is catabolic because it breaks substrates down to CO₂ and water. It is anabolic because its intermediates are the raw material from which the cell builds other molecules.

Look first at how different substrates get in. Fats are broken down into glycerol and fatty acids. The fatty acids are converted step by step into acetyl CoA and enter the pathway at the TCA cycle, while glycerol is converted into PGAL and joins glycolysis partway along. Proteins are hydrolysed into amino acids, and each amino acid is deaminated, that is, its amino group is removed, before the carbon skeleton left behind enters the pathway at whichever point matches its structure. Some enter as pyruvic acid, some as acetyl CoA, and some as intermediates of the TCA cycle such as alpha-ketoglutaric acid or oxaloacetic acid.

Now run those same arrows backwards. When the cell needs to make a fatty acid, it withdraws acetyl CoA from the respiratory pathway rather than oxidising it. When it needs certain amino acids, it withdraws alpha-ketoglutaric acid or oxaloacetic acid. So the very intermediates that are being broken down are also being borrowed for synthesis, which is precisely what amphibolic means. This is also the practical reason the neat 36 ATP figure can never be reached: carbon keeps leaving the pathway sideways.

The respiratory quotient, written RQ, tells you which substrate a tissue is respiring. It is the ratio of the volume of CO₂ given out to the volume of O₂ taken in over the same period of time. For a carbohydrate such as glucose the equation is neatly balanced, six O₂ taken in and six CO₂ given out, so RQ is 1. Fats hold very little oxygen relative to the carbon and hydrogen they contain, so oxidising them takes far more O₂ than the CO₂ they release, and RQ falls below 1; for the fat tripalmitin the ratio works out at about 0.7. Proteins give an RQ of about 0.9. Organic acids are already rich in oxygen, so they need comparatively little extra O₂ and give an RQ above 1.

In practice pure substrates are rare and most cells respire a mixture, so the RQ you measure is an average. That makes RQ genuinely useful as a clue rather than a proof. A germinating oilseed such as castor or groundnut, living on stored fat, shows an RQ well below 1, while a germinating cereal grain living on stored starch shows an RQ close to 1. Notice also that RQ is a ratio of volumes, so it has no unit.