Pentose phosphate pathway: function, products, and regulation

The pentose phosphate pathway, also called the phosphogluconate pathway, is a metabolic pathway, common to all living organisms, for the oxidation of glucose alternative to glycolysis, from which it branches downstream of glucose 6-phosphate synthesis, and whose main function is the production, in variable ratios, of NADPH, a reduced coenzyme, and ribose 5-phosphate, a five-carbon phosphorylated sugar, namely, a pentose phosphate, hence the name pentose phosphate pathway.^[14]
The phosphogluconate pathway, branching from glycolysis, is also called the hexose monophosphate shunt.
Conceptually, this pathway can be viewed as consisting of two phases: the oxidative phase, in which NADPH is produced, and the non-oxidative phase, which produces ribose 5-phosphate and other phosphorylated carbohydrates.^[16]

It has been estimated that more than 10 percent of glucose is shuttled through this metabolic pathway that, noteworthy, although it oxidizes the monosaccharide, does not involve any direct production or consumption of ATP.^[7][13]

Contents

• Hystory
• Main function
  • Additional functions
• Where does the pentose phosphate pathway occur?
• Oxidative phase
  • Oxidation of glucose 6-phosphate to 6-phosphoglucono-delta-lactone
    • Catalytic mechanism of glucose 6-phosphate dehydrogenase
    • Regulation of glucose 6-phosphate dehydrogenase
  • Hydrolysis of 6-phosphoglucono-delta-lactone to 6-phosphogluconate
  • Oxidative decarboxylation of 6-phosphogluconate to ribulose 5-phosphate
    • Catalytic mechanism of 6-phosphogluconate dehydrogenase
• Non-oxidative phase
  • Isomerization of ribulose 5-phosphate to ribose 5-phosphate
    • Catalytic mechanism of phosphopentose isomerase
  • Epimerization of ribulose 5-phosphate to xylulose 5-phosphate
    • Catalytic mechanism of phosphopentose epimerase
  • Transketolase
    • Transketolase and thiamine pyrophosphate
    • Catalytic mechanism of transketolase
  • Transaldolase
    • Catalytic mechanism of transaldolase
• Cell’s need for NADPH, ribose 5-phosphate and ATP
  • The need for NADPH is much greater than that for ribose 5-phosphate and ATP
  • The need for NADPH and ATP is much greater than that for ribose 5-phosphate
  • The need for ribose 5-phosphate is much greater than that for NADPH
  • The needs for ribose 5-phosphate and NADPH are balanced
• References

Hystory

The first evidence of the existence of the pentose phosphate pathway was obtained in the 1930s by the studies of Otto Warburg, Nobel Prize in Physiology or Medicine in 1931, who discovered NADP during studies on the oxidation of glucose 6-phosphate to 6-phosphogluconate.^[22]
Further indications came from the observation that glucose continued to be metabolized in tissues even in the presence of glycolysis inhibitors, such as fluoride and iodoacetate ions, inhibitors of enolase (EC 4.2.1.11) and glyceraldehyde 3-phosphate dehydrogenase (EC 1.2.1.12), respectively.
However, the pathway was fully elucidated only in 1950s thanks to the work of several researchers and primarily of Efraim Racker, Fritz Lipmann, Nobel Prize in Physiology or Medicine in 1953 thanks to the discovery of coenzyme A, Bernard Horecker and Frank Dickens.^[8]

Main function

The main function of the pentose phosphate pathway is the production of NADPH and ribose 5-phosphate.^[22]
NADPH is needed for reductive biosynthesis, such as the synthesis of fatty acids, cholesterol, steroid hormones and of two non-essential amino acids, proline and tyrosine, from glutamate and phenylalanine, respectively, as well as for the reduction of oxidized glutathione.^[14] In such reactions the reduced coenzyme acts as an electron donor, or rather as a donor of a hydride ion (:H^–), namely, a proton and two electrons.

Note: In vertebrates, about half of the NADPH necessary for the reductive steps of fatty acid synthesis derives from the pentose phosphate pathway, and the rest from the malic enzyme (EC 1.1.1.40) reaction.^[13]

Malate + NADP⁺ ↔ Pyruvate + NADPH + H⁺ + HCO₃^–

Ribose 5-phosphate is used for the synthesis of nucleotides and nucleic acids, DNA and RNA, of ATP, coenzymes such as coenzyme A, NAD, NADP and FAD, and of the essential amino acids tryptophan and histidine.^[5] This five-carbon phosphorylated is not used as such; it is activated to 5-phosphoribosyl 1-pyrophosphate (PRPP), in the reaction catalyzed by ribose phosphate pyrophosphokinase or PRPP synthase (EC 2.7.6.1).^[13]

Ribose 5-phosphate + ATP → 5-Phosphoribosyl 1-pyrophosphate + AMP

Additional functions

In addition to the production of NADPH and ribose 5-phosphate, the pentose phosphate pathway has other functions, both anabolic and catabolic.^[18]

• In yeasts and many bacteria it is involved in the catabolism of the five carbon sugars ribose, xylose and arabinose.
  In humans too, it is involved in catabolism of the aforementioned pentoses and of the less common sugars with three, four and seven carbon atoms derived from diet, as well as of:

pentoses derived from the catabolism of structural carbohydrates;
ribose 5-phosphate derived from nucleotide catabolism.

• In photosynthetic organisms it contributes to carbon dioxide (CO₂) fixation during the Calvin cycle.^[18]
• In addition to ribose 5-phosphate, it also provides other intermediates for various biosynthetic processes, such as:

erythrose 4-phosphate, used for the synthesis of phenylalanine, tryptophan, and tyrosine, the three aromatic amino acids;
ribulose 5-phosphate, used for riboflavin synthesis;
sedoheptulose 7-phosphate which, in Gram-negative bacteria, is used for the synthesis of heptose units in the lipopolysaccharide layer of the outer membrane.^[19]

Where does the pentose phosphate pathway occur?

In animal cells and bacteria, the hexose monophosphate shunt, as well as glycolysis, fatty acid synthesis, and most of the reactions of gluconeogenesis, occurs in the cytosol.^[5] And, considering glycolysis, gluconeogenesis and the pentose phosphate pathway we can state that these three metabolic pathways are interconnected through several shared enzymes and/or intermediates.
In plant cells the pentose phosphate pathway occurs in plastids, and its intermediates can reach the cytosol through membrane pores of these organelles.^[13]
In humans, the level of expression of the enzymes of the pathway varies widely from tissue to tissue. Relatively high levels are found in the liver, adrenal cortex, testicles and ovaries, thyroid, mammary glands during lactation, and in red blood cells.^[14] In all these sites, constant supply of NADPH is required to support reductive biosynthesis and/or to counteract the effects of reactive oxygen species (ROS) on sensitive cellular structures, such as DNA, membrane lipids, and proteins by the reduction of oxidized glutathione (GSSG) to reduced glutathione (GSH), which is the major intracellular antioxidant in erythrocytes, as in most other cells, in the reaction catalyzed by glutathione reductase (EC 1.8.1.7).^[1]

GSSG + NADPH + H⁺ → 2 GSH + NADP⁺

High levels of the of the phosphogluconate pathway enzymes are also present in rapidly dividing cells such as enterocytes, skin cells, bone marrow cells, those of the early embryo and, in pathological conditions, cancer cells. Indeed, these cell types require a constant supply of ribose 5-phosphate for nucleic acid synthesis.^[16]
Conversely, these enzymes are present in very low levels in skeletal muscle, in which the pentose phosphate pathway is virtually absent and glucose 6-phosphate is primarily used for energy production via glycolysis and the citric acid cycle.^[5]

Oxidative phase

The oxidative phase of the pentose phosphate pathway consists of two irreversible oxidations, the first and third reactions, and a hydrolysis.
In this phase, glucose 6-phosphate is converted to ribulose 5-phosphate, a five-carbon phosphorylated sugar and the starting substrate for the reactions of the non-oxidative phase, with the concomitant formation of two molecules of NADPH and the release of C-1 of glucose as CO₂.^[7] The overall equation is:

3 Glucose 6-phosphate + 6 NADP⁺ + H₂O → 6 NADPH + 6 H⁺ + 3 CO₂ + 3 Ribulose 5-phosphate

Oxidation of glucose 6-phosphate to 6-phosphoglucono-delta-lactone

In the first step of the oxidative phase, glucose 6-phosphate dehydrogenase catalyzes the oxidation of glucose 6-phosphate to 6-phosphoglucono-δ-lactone, an intramolecular ester, via the transfer of a hydride ion from carbon 1 of glucose 6-phosphate to NADP⁺, that acts as oxidizing agent.

Glucose 6-phosphate + NADP⁺ → 6-Phosphoglucono-δ-lactone + NADPH + H⁺

Note: This reaction yields the first molecule of NADPH of the pentose phosphate pathway.
The reaction catalyzed by glucose 6-phosphate dehydrogenase or G6PD (EC 1.1.1.49) is unique to the pathway.^[1] And, similarly to what happens in most metabolic pathways, also in this case the first reaction unique to the pathway, generally known as a committed step, is an essentially irreversible step, with a ΔG in the liver of -17.6 kJ/mol (-4.21 kcal/mol), and is highly allosterically regulated.^[22] And the enzyme is indeed the major control point for the flow of metabolites through the pathway.
In humans, the highest levels of G6PD are found in neutrophils and macrophages, phagocytic cells in which, during inflammation, NADPH is used for to produce superoxide radicals (O₂^-.) from molecular oxygen in the reaction catalyzed by NADPH oxidase (EC 1.6.3.1).^[2][16]

2 O₂ + NADPH → 2 O₂^-. + NADP⁺ + H⁺

In turn, superoxide radicals can be used for the synthesis for defensive purposes, namely, to kill phagocytized microorganisms, of other ROS but also of reactive nitrogen species (RNS), such as:

• hydrogen peroxide (H₂O₂), in the reaction catalyzed by superoxide dismutase or SOD (EC 1.15.1.1)

2 O₂^-. + 2 H⁺ → H₂O₂ + O₂

• peroxynitrite (O=N–O–O), in the reaction with nitric oxide (•NO)

O₂^–. + •NO → O=N–O–O^–

• hydroperoxide radical (HOO•)

O₂^-. + H⁺ → HOO•

Catalytic mechanism of glucose 6-phosphate dehydrogenase

The catalytic mechanism of the enzyme has been studied in great detail in the microorganism Leuconostoc mesenteroides, whose glucose 6-phosphate dehydrogenase has the peculiar characteristic of being able to use NAD⁺ and/or NADP⁺ as coenzyme.^[12]

The enzyme does not require metal ions for its activity; one of the amino acids in the active site acts as a general base being able to abstract a hydride ion from the hydroxyl group bound to C1 of glucose 6-phosphate.^[4]
In the bacterial enzyme this is carried out by the atom Nɛ2 of the imidazole ring of a histidine side chain. This nitrogen atom has a lone pair of electrons able to make a nucleophilic attack. This causes glucose 6-phosphate, a cyclic hemiacetal with carbon 1 in the aldehyde oxidation state, to be oxidized to a cyclic ester, namely, a lactone, and the transfer of an hydride ion from C1 of glucose to C4 of the nicotinamide ring of NADP⁺ to form NADPH.
Because such histidine is conserved in many of the glucose 6-phosphate dehydrogenases sequenced, it is likely that this catalytic mechanism can be generalized to all glucose 6-phosphate dehydrogenases.

Regulation of glucose 6-phosphate dehydrogenase

Glucose 6-phosphate dehydrogenase is the major control point of carbon flow through the pentose phosphate pathway, and then the major control point for the rate of NADPH synthesis.^[2]
In humans, the enzyme exists in two forms: the inactive monomeric form, and the active form that exists in a dimer-tetramer equilibrium.^[1]
One of the main modulators of its activity is the cytosolic NADP⁺/NADPH ratio.^[15] High levels of NADPH inhibit enzyme activity, because NADPH is a potent competitive inhibitor of G6PD, whereas NADP⁺ is required for the catalytic activity and for the maintenance of the active conformation. In fact, the binding of the oxidized coenzyme to a specific site close to the dimer interface, but distant from the active site, is required to maintain its dimeric conformation.

Under most metabolic conditions the NADP⁺/NADPH ratio is low, less NADP⁺ is available to bind to the enzyme, and hence enzyme activity is reduced, regardless of gene expression levels. Under these conditions the oxidative phase is virtually inactive.
Conversely, in cells in which metabolic pathways and/or reactions using NADPH are particularly active, the reduction of cytosolic NADPH concentration, and hence the increase in NADP⁺ concentration occurs. This leads to an increase in glucose 6-phosphate dehydrogenase activity, and to the activation of the oxidative phase of the hexose monophosphate pathway.
Therefore it is possible to state that the fate of glucose 6-phosphate also depends on the current needs for NADPH.^[15]
A second mechanism for the regulation of G6PD activity calls into question the accumulation of acyl-CoAs, intermediates in fatty acid synthesis.^[5] These molecules, by binding to the dimeric form of the enzyme, lead to dissociation into the constitutive monomers, and then to the loss of the catalytic activity.
Insulin up-regulates the expression of the genes for glucose 6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase. Therefore, in the well-fed state, the hormone increases carbon flow through the pentose phosphate pathway and then the production of NADPH.^[7]
Note: Insulin also promotes the synthesis of fatty acids.

Hydrolysis of 6-phosphoglucono-delta-lactone to 6-phosphogluconate

In the second step of the oxidative phase 6-phosphoglucono-δ-lactone is hydrolyzed to 6-phosphogluconate, a linear molecule.
6-Phosphoglucono-δ-lactone is hydrolytically unstable and undergoes a nonenzymatic ring-opening, a reaction that occurs at a significant rate.^[22] However, in the cell this ring-opening reaction, an hydrolysis, is accelerated by the catalytic action of 6-phosphogluconolactonase (EC 3.1.1.31).^[3]

6-Phosphoglucono-δ-lactone + H₂O → 6-Phosphogluconate + H⁺

Oxidative decarboxylation of 6-phosphogluconate to ribulose 5-phosphate

In the last step of the oxidative phase, 6-phosphogluconate undergoes an oxidative decarboxylation to form ribulose 5-phosphate, a keto pentose, CO₂, and a molecule of NADPH. The reaction is catalyzed by 6-phosphogluconate dehydrogenase (EC 1.1.1.44), enzyme that requires the presence of magnesium ions, Mg²⁺.

 6-Phosphogluconate + NADP⁺ → Ribulose 5-phosphate + NADPH + CO₂

Note: This reaction yields the second molecule of NADPH of the pentose phosphate pathway.^[15]

Catalytic mechanism of 6-phosphogluconate dehydrogenase

The catalytic mechanism of the enzyme is similar to that of isocitrate dehydrogenase (EC 1.1.1.41), an enzyme of the citric acid cycle.^[22] It consists of an acid-base catalysis proceeding through a three step mechanism in which two strictly conserved residues, a lysine (Lys), and a glutamate (Glu), are involved; in humans, Lys¹⁸⁵ and the Glu¹⁹². Lysine acts as acid/base group, whereas glutamate as an acid.^[6][17]

In the first step, the oxidative step, 6-phosphogluconate is oxidized to a beta-keto acid, the 3-keto-6-phosphogluconate.
In this step the ε-amino group of the aforementioned lysine acts as a general base, as a nucleophile, abstracting a proton from the hydroxyl group bound to C-3. Then, the transfer of a hydride ion from C-3 to C-4 of the nicotinamide ring of the NADP⁺ occurs. This leads to the formation of the 3-keto intermediate and a molecule of NADPH that leaves the active site.
In the second step, the decarboxylation step, 3-keto-6-phosphogluconate, that is very susceptible to decarboxylation, is converted to the cis-1,2-enediol of ribulose 5-phosphate, a high energy intermediate. In this step the aforementioned lysine acts as a general acid donating an H⁺ at the C-3 carbonyl oxygen, and the C-1 of glucose 6-phosphate is lost as CO₂.
Finally, 6-phosphogluconate dehydrogenase catalyzes a stereospecific keto-enol conversion leading to the formation of ribulose 5-phosphate. In this step, the aforementioned glutamic acid residue acts as a general acid donating an H⁺ to the C-1 of cis-1,2-enediol intermediate, while the ε-amino group of the lysine accepts a proton from the hydroxyl group bound to the C-2. The result is the formation of ribulose 5-phosphate.

Non-oxidative phase

In the non-oxidative phase of the pentose phosphate pathway, several phosphorylated carbohydrates are produced, whose fate depends on the relative needs for NADPH, ribose 5-phosphate, and ATP of the cell.^[7]
This phase consists of five steps, all freely reversible, in which a series of interconversions of phosphorylated sugars occurs. It begins with two reactions: the isomerization and epimerization of ribulose 5-phosphate to form ribose 5-phosphate and xylulose 5-phosphate, respectively.^[14]
Note: enzymatic isomerizations and epimerizations play an important role in carbohydrate metabolism.
Epimerases (EC 5.1), a subclass of Isomerases (EC 5.), catalyze the configurational reversal at an asymmetric carbon atom, usually by a deprotonation/protonation mechanism.
In isomerization reactions, the interchange of chemical groups occurs between carbon atoms.

Isomerization of ribulose 5-phosphate to ribose 5-phosphate

In the first step of the nonoxidative phase of the pentose phosphate pathway, an isomerization reaction, ribulose 5-phosphate, a ketose, is converted to the corresponding aldose, ribose 5-phosphate. The reaction is catalyzed by phosphopentose isomerase or ribose 5-phosphate isomerase (EC 5.3.1.6).

Ribulose 5-phosphate ⇄ Ribose 5-phosphate

These molecules are an example of functional group isomerism.

Catalytic mechanism of phosphopentose isomerase

The catalytic mechanism of phosphopentose isomerase is similar to that of phosphohexose isomerase (EC 5.3.1.9), a glycolytic enzyme, and leads to the formation of the high energy intermediate cis-1,2-enediol of ribulose 5-phosphate via a proton-transfer mechanism common to the aldose-ketose isomerizations.
The proposed catalytic mechanism for phosphopentose isomerase from E. coli, in the direction of ribulose 5-phosphate formation from ribose 5-phosphate, as in the Calvin cycle of photosynthesis, is described below.^[23]

In the first step, the furanose ring of the substrate is opened, opening induced by the interaction with an aspartic acid residue (Asp⁸¹) that accepts a proton from the hydroxyl group bound to C-1, whereas it is likely that water is the proton donor.
Note: The opening of the furanose ring is quite rare in solution (<0.5 percent).
Once the chain is opened, a glutamic acid residue (Glu¹⁰³) acts as a general base, as a nucleophile, abstracting a proton bound to the C-2, whereas Asp⁸¹ donates a proton. As a result, cis-1,2-enediol intermediate is produced.
Finally, the protonated Glu¹⁰³ acts as a general acid and donates an H⁺ at C-1 of the cis-1,2-enediol intermediate, while Asp⁸¹ acts as a general base accepting a proton from the hydroxyl group bound to C-2. The result is the formation of ribulose 5-phosphate.
During the synthesis of ribose 5-phosphate from ribulose 5-phosphate phosphopentose isomerase works in reverse.

Epimerization of ribulose 5-phosphate to xylulose 5-phosphate

The other metabolic fate of ribulose 5-phosphate in the pentose phosphate pathway is to be epimerized to xylulose 5-phosphate, a ketose like ribulose 5-phosphate, in the reaction catalyzed by phosphopentose epimerase (EC 5.1.3.1).

Ribulose 5-Phosphate ⇄ Xylulose 5-Phosphate

Note: Xylulose 5-phosphate is a regulatory molecule that inhibits gluconeogenesis and stimulates glycolysis by controlling the levels of fructose 2,6-bisphosphate in the liver.^[10]

Catalytic mechanism of phosphopentose epimerase

Also this reaction, like those catalyzed by 6-phosphogluconate dehydrogenase and ribose 5-phosphate isomerase, proceeds through the formation of an enediol intermediate, but with the double bond between C-2 and C-3 and not between C-1 and C-2.
During the reaction an amino acid residue present in the active site of the enzyme acts as a general base, as a nucleophile, and abstracts a proton bound to the C-3, leading to the formation of the cis-2,3-enediol intermediate. Then, an acidic amino acid residue donates a proton to C-3, but from the opposite side, hence, with an inversion at C-3 to form xylulose 5-phosphate.^[5][9]

To this point, the hexose monophosphate shunt has generated for each molecule of glucose 6-phosphate metabolized:

• a pool of three pentose 5-phosphates, namely, ribulose 5-phosphate, ribose 5-phosphate and xylulose 5-phosphate, that coexist at equilibrium;
• 2 molecules of NADPH.

In the following three steps, from the sixth to the eighth, transketolase (EC 2.2.1.1) and transaldolase (EC 2.2.1.2), two enzymes unique to the pentose phosphate pathway, catalyze a series of rearrangements of the carbon skeletons leading to the formation of three-, four-, six-, and seven carbon units, that can be used for various metabolic purposes, depending on the needs of the cell.^[2]
Analyzing the flow of metabolites through the different metabolic pathways, the concerted action of transketolase and transaldolase allows the interaction of the pentose phosphate pathway, in particular of its non-oxidative phase, with glycolysis, and gluconeogenesis, as well as with the pathways leading to the formation of numerous vitamins, coenzymes and nucleic acid precursors.

Transketolase

Transketolase is the rate-limiting enzyme of the nonoxidative phase of the pentose phosphate pathway, and the first enzyme that acts downstream of ribose 5-phosphate isomerase and phosphopentose epimerase.^[11]
Discovered independently in 1953 by Horecker and Racker, and named by Racker, it catalyzes in the sixth and eighth steps, the transfer of a two carbon unit from a ketose, the donor substrate, namely, xylulose 5-phosphate, sedoheptulose 7-phosphate or fructose 6-phosphate, to an aldose, the acceptor substrate, ribose 5-phosphate, glyceraldehyde 3-phosphate or erythrose 4-phosphate.^[8]

Taking as an example the forward reactions, in the sixth step, the ketose donor is xylulose 5-phosphate, whereas the aldose acceptor is ribose 5-phosphate, to form glyceraldehyde 3-phosphate, the remaining three-carbon fragment from xylulose 5-phosphate, and sedoheptulose 7-phosphate, a seven-carbon sugar that will be used in the next step, the seventh.
In the eighth step, the ketose donor is xylulose 5-phosphate, whereas the aldose acceptor is erythrose 4-phosphate, to form another glyceraldehyde 3-phosphate and a fructose 6-phosphate.
It should be noted that three of the four products of the reactions catalyzed by this enzyme, two molecules of glyceraldehyde 3-phosphate and one of fructose 6-phosphate, are also intermediates of glycolysis.^[11]
In addition to xylulose 5-phosphate, sedoheptulose 7-phosphate and fructose 6-phosphatetransketolase can use as substrates other 2-keto sugars, as well as a variety of different aldose phosphates.

Transketolase and thiamine pyrophosphate

Transketolase is an enzyme that requires thiamine pyrophosphate (TPP) as a cofactor.
Thiamine pyrophosphate is the biologically active form of thiamin or vitamin B₁, and is tightly bound to the enzyme.

Other enzymes that require TPP as a cofactor are:

• pyruvate decarboxylase (EC 4.1.1.1), that is involved in alcoholic fermentation;
• pyruvate dehydrogenase or E₁ (EC 1.2.4.1) of the pyruvate dehydrogenase complex;
• alpha-keto acid dehydrogenase or E₁ component (EC 1.2.4.4) of the branched-chain alpha-ketoacid dehydrogenase complex, which catalyzes the oxidative decarboxylation of branched-chain alpha-keto acids;
• alpha-ketoglutarate dehydrogenase or E₁ component (EC 1.2.4.2) of the alpha-ketoglutarate dehydrogenase complex, an enzyme of the citric acid cycle.

TPP is involved in the transfer of activated aldehyde intermediates by stabilizing the two-carbon carbanions formed during the reaction.

Catalytic mechanism of transketolase

The carbon atom between the sulfur and nitrogen atoms of the thiazolium ring of thiamine pyrophosphate, namely, the C-2 atom, is much more acidic than most =CH groups found in other molecules because of adjacent positively charged nitrogen atom that electrostatically stabilizes the carbanion resulting from dissociation of the proton. This causes the C-2 proton to be easily dissociable to form a carbanion, i.e. a carbon atom with a negative charge. Such proton abstraction is catalyzed by transketolase.^[5][14][22]
The carbanions attack the carbonyl carbon of the substrate, in the step 6, xylulose 5-phosphate or, in the reverse reaction, sedoheptulose 7-phosphate, whereas in the step 8, xylulose 5-phosphate or, in the reverse reaction, fructose 6-phosphate.
Taking as an example the forward reaction of step 6, the covalent adduct between thiamine pyrophosphate and xylulose 5-phosphate undergoes fragmentation, via the cleavage of the C2-C3 bond of xylulose 5-phosphate, to form glyceraldehyde 3-phosphate, that is released, and a two carbon unit, a negatively charged hydroxyethyl group, that remains bound to C-2 of the thiazolium ring.

The negative charge on the hydroxyethyl intermediate is stabilized by the thiazolium ring of thiamine pyrophosphate because of the positively charged nitrogen atom that acts as an electron trap. Therefore, thiazolium ring provides an electron deficient or electrophilic structure that can delocalize by resonance the carbanion electrons.
Then, the condensation occurs between the hydroxyethyl group and the ribose 5-phosphate, the acceptor aldehyde substrate, via carbanion attack on the aldehyde carbon of ribose 5-phosphate, to form a covalent adduct bound to thiamine pyrophosphate.
Finally, the cleavage of the adduct leads to the release of sedoheptulose 7-phosphate, and regenerates the TPP carbanion.

Transaldolase

Discovered in 1953 by Horecker and Smyrniotis in the brewer’s yeast, assigned to the species Saccharomyces cerevisiae, it catalyzes, in the seventh step of the pentose phosphate pathway, the transfer of a three carbon unit from a donor substrate, sedoheptulose 7-phosphate, to an acceptor substrate, glyceraldehyde 3-phosphate, to form fructose 6-phosphate and erythrose 4-phosphate.^[18]

Sedoheptulose 7-phosphate + Glyceraldehyde 3-phosphate ⇄ Fructose 6-phosphate + Erythrose 4-phosphate

Note: Like in transketolase catalyzed reactions, the carbon unit donor is a ketose while the acceptor is an aldose.
In the reverse reaction, the donor substrate is fructose 6-phosphate, while the acceptor substrate is erythrose 4-phosphate.

Catalytic mechanism of transaldolase

Unlike transketolase, transaldolase does not require a cofactor for activity.
The reaction occurs in two step, an aldol cleavage and an aldol condensation. Below, the catalytic mechanism of E. coli transaldolase B is analyzed, taking as an example the forward reaction leading to erythrose 4-phosphate and fructose 6-phosphate synthesis.
In the first step an ε-amino group of a lysine residue (Lys¹³²) in the active site, after a proton transfer to a glutamic acid residue (Glu⁹⁶) mediated by a water molecule, performs a nucleophilic attack on the carbonyl carbon of sedoheptulose 7-phosphate, that is, on the C-2 atom. The result is the formation of a carbinolamine with sedoheptulose 7-phosphate.^{[5][14][18][22]}

In the second step, the removal of a water molecule from carbinolamine leads to the formation of an enzyme-bound imine or Schiff base intermediate; this step, too, involves the transfer of a proton from Glu⁹⁶ to the “catalytic” water molecule.
Note: This enzyme-substrate covalent intermediate is quite similar to that formed in the reaction catalyzed by aldolase (EC 4.1.2.13) in the fourth step of glycolysis.
In the next step, the carboxylic group of an aspartic acid residue (Asp¹⁷) extracts a proton from the hydroxyl group bound to C-4, leading to the cleavage of the C–C bond between C-3 and C-4. This reaction is an aldol cleavage and releases the first product, erythrose 4-phosphate, an aldose, whereas a three-carbon carbanion remains bound to the enzyme and is stabilized by resonance, like in transketolase catalyzed reactions. In fact, like the nitrogen atom in the thiazolium ring of thiamine pyrophosphate, the nitrogen atom with a positive charge of the Schiff base acts as an electron trap stabilizing the negative charge carried by the carbanion.
Once the acceptor substrate glyceraldehyde 3-phosphate is in the active site, the carbanion performs a nucleophilic attack on the carbonyl carbon of glyceraldehyde 3-phosphate to form, by aldol condensation, a new C–C bond and an enzyme-bound ketose.
Then, the hydrolysis of the Schiff base releases fructose 6-phosphate, a ketose and the second product of the reaction. At this point, a new reaction cycle can start.
Finally, as seen previously, in the eighth step of the pentose phosphate pathway, transketolase catalyzes the synthesis of fructose 6-phosphate and glyceraldehyde 3-phosphate from erythrose 4-phosphate and xylulose 5-phosphate.

Cell’s need for NADPH, ribose 5-phosphate and ATP

From the molecular point of view, the fate of glucose 6-phosphate depends, to a large extent, on the relative activities of the enzymes that metabolize it in glycolysis and in the pentose phosphate pathway, in particular phosphofructokinase 1 (PFK-1) (EC 2.7.1.11) and glucose 6-phosphate dehydrogenase, whose activities that are highly regulated.
PFK-1 is inhibited when ATP and/or citrate concentrations increase, namely, when the energy charge of the cell is high, whereas it is activated when AMP and/or fructose 2,6-bisphosphate concentrations increase, namely, when the energy charge of the cell is low. Thus, when the energy charge of the cell is high, the carbon flow, and therefore the flow of glucose 6-phosphate, through the glycolytic pathway decreases.^[20][21]
Glucose 6-phosphate dehydrogenase is inhibited by NADPH and acyl-CoAs, intermediates in fatty acid biosynthesis.^[15] Thus, when the cytosolic levels of NADPH increases, the flow of glucose 6-phosphate through the pentose phosphate pathway is inhibited, whereas if NADPH levels drop, the inhibition disappears, the pathway switches on again, and NADPH and ribose 5-phosphate are synthesized.
Therefore, depending on the cell’s need for ATP, NADPH and ribose 5-phosphate, some reactions of glycolysis and the pentose phosphate pathway can be combined in novel ways to emphasize the synthesis of needed metabolites, also exploiting the fact that the non-oxidative phase of the hexose monophosphate shunt is essentially controlled by the availability of the substrates.
The four principal possibilities are described below.^[2]

The need for NADPH is much greater than that for ribose 5-phosphate and ATP

When much more NADPH than ribose 5-phosphate is needed, and there is no need for additional ATP to be produced, namely, the energy charge of the cell is high, glucose 6-phosphate enters the pentose phosphate pathway and is completely oxidized to CO₂. Such metabolic conditions are found, for example, in the adipose tissue during fatty acid synthesis.
Through a combination of the reactions of the non-oxidative phase and of some reactions of gluconeogenesis, namely, those catalyzed by triose phosphate isomerase (EC 5.3.1.1), aldolase (EC 4.1.2.13), phosphohexose isomerase (EC 5.3.1.9), and fructose 1,6-bisphosphatase (EC 3.1.3.11), six molecules of ribulose 5-phosphate are converted into five molecules of glucose 6-phosphate. Thus, it is possible to state that the reactions of the non-oxidative phase allow the reactions of the oxidative phase to proceed.
Three groups of reactions can be identify.

• In the first group there are the reactions catalyzed by the enzymes of the oxidative phase, leading to the formation of two molecules of NADPH and one molecule of ribulose 5-phosphate.

6 Glucose 6-phosphate + 12 NADP⁺ + 6 H₂0 → 6 Ribulose 5-phosphate + 6 CO₂ + 12 NADPH + 12 H⁺

• In the second group there are the reactions catalyzed by the enzymes phosphopentose epimerase, ribose 5-phosphate isomerase, transketolase and transaldolase, namely, those of the non-oxidative phase of the pathway, that lead to the conversion of ribulose 5-phosphate to fructose 6-phosphate and glyceraldehyde 3-phosphate.

6 Ribulose 5-phosphate → 4 Fructose 6-phosphate + 2 Glyceraldehyde 3-phosphate

• Finally, fructose 6-phosphate and glyceraldehyde 3-phosphate can be recycled to glucose 6-phosphate via some reactions of gluconeogenesis, so that the cycle can begin again.

4 Fructose 6-phosphate + 2 Glyceraldehyde 3-phosphate + H₂O → 5 Glucose 6-phosphate + P_i

The sum of the last two reactions shows that six molecules of ribulose 5-phosphate are converted to five molecules of glucose 6-phosphate.

6 Ribulose 5-phosphate+ H₂O → 5 Glucose 6-phosphate + P_i

The sum of the reactions of the first, second and third group gives the overall reaction:

Glucose 6-phosphate + 12 NADP⁺ + 7 H₂0 → 6 CO₂ + 12 NADPH + 12 H⁺ + P_i

Therefore, one molecule of glucose 6-phosphate, via six cycles of the pentose phosphate pathway coupled with some reactions of gluconeogenesis, is converted to six molecules of CO₂, with the concomitant production of 12 molecules of NADPH, and without net production of ribose-5-phosphate.

The need for NADPH and ATP is much greater than that for ribose 5-phosphate

When much more NADPH than ribose 5-phosphate is needed, and the energy charge of the cell is low, that is, there is a need for ATP, ribulose 5-phosphate formed in the oxidative phase of the pentose phosphate pathway is converted to fructose 6-phosphate and glyceraldehyde 3-phosphate through the reactions of the non-oxidative phase. These two intermediates, through the reactions of glycolysis, are oxidized to pyruvate, the conjugate base of pyruvic acid, with concomitant ATP production.
The net reaction is:

3 Glucose 6-phosphate + 6 NADP⁺ + 5 NAD⁺ + 5 P_i + 8 ADP → 5 Pyruvate + 3 CO₂ + 6 NADPH + 5 NADH + 8 ATP + 2 H₂O + 8 H⁺

If the cell requires more ATP, the pyruvate produced can be oxidized through the citric acid cycle.
Conversely, if there is no need for additional ATP to be produced, the carbon skeleton of pyruvate can be used as a building block in several biosynthetic pathways.

Note: As in the previous case, there is no net production of ribose 5-phosphate.

The need for ribose 5-phosphate is much greater than that for NADPH

When much more ribose 5-phosphate than NADPH is needed, as in rapidly dividing cells in which there is a high rate of synthesis of nucleotides, precursors of DNA, the reactions of the oxidative phase of the pentose phosphate pathway are bypassed, and there is no synthesis of NADPH. Conversely, because the reactions of the non-oxidative phase are easily reversible, the drop in ribose 5-phosphate levels, due to its rapid use, stimulates its synthesis.
What happens is that, through the glycolytic pathway, most of the glucose 6-phosphate is converted to fructose 6-phosphate and glyceraldehyde 3-phosphate. Then, transaldolase and transketolase lead to the synthesis of ribose 5-phosphate and xylulose 5-phosphate. Xylulose 5-phosphate, through the reactions catalyzed by phosphopentose epimerase and ribose 5-phosphate isomerase, is converted to ribose 5-phosphate.
The net reaction is:

6 Glucose 6-phosphate + ATP → 6 Ribose 5-phosphate + ADP + H⁺

Under this metabolic conditions therefore, what happens is an interplay between reactions of glycolysis and of the non-oxidative phase of the phosphogluconate pathway, with the latter in the direction of ribose 5-phosphate synthesis.
It should be noted no metabolites return to glycolysis.

The needs for ribose 5-phosphate and NADPH are balanced

If one molecule of ribose 5-phosphate and two molecules of NADPH per molecule of glucose 6-phosphate metabolized satisfy the metabolic needs of the cell, the reactions that predominate are those of the oxidative phase and that catalyzed by ribose 5-phosphate isomerase.
The net reaction is:

Glucose 6-phosphate + 2 NADP⁺ + H₂O → Ribose 5-phosphate + 2 NADPH + 2 H⁺ + CO₂

Under this metabolic conditions, too, no metabolites return to glycolysis.

References

1. ^ ^{a b c} Au S.W.N., Gover S., Lam V.M.S., Adams M.J. Human glucose-6-phosphate dehydrogenase: the crystal structure reveals a structural NADP⁺ molecule and provides insights into enzyme deficiency. Structure 2000;8(3):293-303. doi:10.1016/S0969-2126(00)00104-0
2. ^ ^{a b c d} Berg J.M., Tymoczko J.L., and Stryer L. Biochemistry. 5th Edition. W. H. Freeman and Company, 2002
3. ^{^} Brodie A.F., Lipmann F. Identification of a gluconolactonase. J Biol Chem 1955;212(2):677-85. doi:10.1016/S0021-9258(18)71006-5
4. ^{^} Cosgrove M.S., Naylor C., Paludan S., Adams M.J., Richard Levy H. On the mechanism of the reaction catalyzed by glucose 6-phosphate dehydrogenase. Biochemistry 1998;37(9);2759-2767. doi:10.1021/bi972069y
5. ^ ^{a b c d e f g} Garrett R.H., Grisham C.M. Biochemistry. 4th Edition. Brooks/Cole, Cengage Learning, 2010
6. ^{^} Hanau S., Montin K., Cervellati C., Magnani M., Dallocchio F. 6-Phosphogluconate dehydrogenase mechanism: evidence for allosteric modulation by substrate. J Biol Chem 2010;285(28):21366-21371. doi:10.1074/jbc.M110.105601
7. ^ ^{a b c d} Harvey R.A., Ferrier D.R. Lippincott’s illustrated reviews: biochemistry. 5th Edition. Lippincott Williams & Wilkins, 2011
8. ^ ^{a b} Horecker B.L. The pentose phosphate pathway. J Biol Chem 2002;277(50):47965-47971. doi:10.1074/jbc.X200007200
9. ^{^} Jelakovic S., Kopriva S., Süss K-H, Schulz G.E. Structure and catalytic mechanism of the cytosolic D-ribulose-5-phosphate 3-epimerase from rice. J Mol Biol 2003;326:127-135. doi:10.1016/S0022-2836(02)01374-8
10. ^{^} Kabashima T., Kawaguchi T., Wadzinski B.E., Uyeda K. Xylulose 5-phosphate mediates glucose-induced lipogenesis by xylulose 5-phosphate-activated protein phosphatase in rat liver. Proc Natl Acad Sci USA 2003;100:5107-5112. doi:10.1073/pnas.0730817100
11. ^ ^{a b} Kochetov G.A., Solovjeva O.N. Structure and functioning mechanism of transketolase. Biochim Biophys Acta 2014;1844(9):1608-1618. doi:10.1016/j.bbapap.2014.06.003
12. ^{^} Levy H.R. Glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides. Biochem Soc Trans. 1989 Apr;17(2):313-5. doi:10.1042/bst0170313
13. ^ ^{a b c d} Michal G., Schomburg D. Biochemical pathways. An atlas of biochemistry and molecular biology. 2nd Edition. John Wiley J. & Sons, Inc. 2012
14. ^ ^{a b c d e f} Nelson D.L., Cox M.M. Lehninger. Principles of biochemistry. 6th Edition. W.H. Freeman and Company, 2012
15. ^ ^{a b c d} Patra K.C., Hay N. The pentose phosphate pathway and cancer. Trends Biochem Sci 2014;39(8):347-354. doi:10.1016/j.tibs.2014.06.005
16. ^ ^{a b c} Rosenthal M.D., Glew R.H. Medical Biochemistry – Human Metabolism in Health and Disease. John Wiley J. & Sons, Inc., 2009
17. ^{^} Ruda G.F., Campbell G., Alibu V.P., Barrett M.P., Brenk R., Gilbert I.H. Virtual fragment screening for novel inhibitors of 6-phosphogluconate dehydrogenase. Bioorg Med Chem 2010;18(14):5056-62. doi:10.1016/j.bmc.2010.05.077
18. ^ ^{a b c d} Samland A.K., Sprenger G.A. Transaldolase: from biochemistry to human disease. Int J Biochem Cell Biol 2009;41(7):1482-1494. doi:10.1016/j.biocel.2009.02.001
19. ^{^} Sprenger G.A. Genetics of pentose-phosphate pathway enzymes of Escherichia coli K-12. Arch Microbiol 1995;164(5):324-30. doi:10.1007/BF02529978
20. ^{^} Van Schaftingen E., Hers H-G. Inhibition of fructose-1,6-bisphosphatase by fructose-2,6-bisphosphate. Proc Natl Acad Sci USA 1981;78(5):2861-2863. doi:10.1073/pnas.78.5.2861
21. ^{^} Van Schaftingen E., Jett M-F., Hue L., Hers H-G. Control of liver 6-phosphofructokinase by fructose 2,6-bisphosphate and other effectors. Proc Natl Acad Sci USA 1981;78(6):3483-3486. doi:10.1073/pnas.78.6.3483
22. ^ ^{a b c d e f g} Voet D. and Voet J.D. Biochemistry. 4th Edition. John Wiley J. & Sons, Inc. 2011
23. ^{^} Zhang R., Andersson C.E., Savchenko A., Skarina T., Evdokimova E., Beasley S., Arrowsmith C.H., Edwards A.M., Joachimiak A., Mowbray S.L. Structure of Escherichia coli ribose-5-phosphate isomerase: a ubiquitous enzyme of the pentose phosphate pathway and the Calvin cycle. Structure 2003;11(1):31-42. doi:10.1016/S0969-2126(02)00933-4