Modes of Nutrition - Definitions

An organism needs a source of carbon plus energy plus reducing power. These may all come from the same source (e.g. glucose provides all three) or they may come from different sources:

Where does the carbon come from?

a) Organic molecules - heterotrophs

b) Inorganic - in practice usually CO2 - autotrophs

Where does the energy come from?

a) Chemical reactions (redox reactions) - chemotrophs

b) Light - phototrophs

What molecule is the electron donor?

a) Organic molecules - organotrophs

b) Inorganic (e.g., H2O, H2, Sulfur) - lithotrophs

Do not confuse heterotrophy with organotrophy (or autotrophy with lithotrophy). In one case the organic molecule is a source of carbon, in the other, a source of electrons (reducing power). In practice, most heterotrophs are also organotrophs (and most lithotrophs are autotrophs) but NOT always. e.g., E. coli can live anaerobically as a heterotrophic chemolithotroph by using H2 and fumarate. Here fumarate is the C-source, and H2 the electron donor. Energy is derived from the oxidation of H2 and the concomitant reduction of fumarate to succinate.

Overview of Heterotrophic Metabolism

Bacteria such as E. coli can degrade a wide range of organic molecules. These are generally used as sources of carbon energy and reducing power. In air they are oxidised to CO2 via the Krebs Cycle (= Citric Acid Cycle) and yield energy. Anaerobically, sugars are fermented but other compounds are only used if alternative electron acceptors such as nitrate are available.

E. COLI GROWING ON SUGARS

We start by considering E. coli using sugars as carbon and energy source.

A) Sugar enters the periplasmic space

A) Entry Of Sugars Into Periplasmic Space

The outer membrane is penetrated by water-filled pores. Each pore consists of a trimer of porin and has a diameter of 1.5-2.0 nm. Substances of MW less than 650 may diffuse rapidly through the pore (if hydrophilic). Substances of MW greater than 900-1000 are excluded. The porins are arranged on the outer surface of the mucopeptide in a regular array and are mostly beta-sheet. They are tightly, but non-covalently, bound to the mucopeptide. Vesicles, reconstituted from lipid, LPS and porins show the same diffusion properties as an intact cell. E. coli K-12 has two porins, coded by genes ompC and ompF. (omp = Outer Membrane Protein) A regulatory gene ompB controls the expression of the ompC and F proteins, for example, when osmotic pressure increases, ompF decreases and ompC increases. Somehow, ompB detects changes in osmotic pressure. Both ompC and ompF act as receptors for certain bacteriophages. By selecting phage-resistant mutants it is possible to eliminate either or both of the porins. Porinless mutants are resistant to Cu2+ ions and to beta-lactams such as cephaloridine. They cannot take up sugars and amino acids at low concentrations. Porin-less mutants can grow on glucose if the concentration is greater than about 0.2%, but on lactose (a disaccharide) only if offered 1% sugar.

B) Transport Across The Cytoplasmic Membrane.

Several types of transport systems:

a) Facilitated Diffusion

b) Proton Motive Force Linked Active Transport

c) ATP Linked Active Transport

d) Group Translocation

a) Facilitated Diffusion

A facilitator protein is located in the cytoplasmic membrane and carries the substrate either into or out of the cell with equal ease. No energy is coupled to transport. This mechanism is rare. In E. coli glycerol is transported by the GlpF facilitator proein. After entry, glycerol (plus ATP) is converted to glycerol-3-phosphate (G3P) by glycerol kinase. The G3P cannot leave the cell, since only neutral molecules (such as glycerol, erythritol, glyceraldehyde) are carried by the facilitator.

b) Proton Motive Force Linked Transport.

Transport of some sugars (lactose, melibiose) and many amino acids is coupled directly to the PMF. Lactose transport is by proton symport. One proton and one lactose are transported by the lactose permease (product of lacY gene). Only lactose permease which has bound both lactose plus a proton will transport both substrates across the membrane. Neither lactose nor protons are moved alone. Purified lactose permease in lipid vesicles has been shown to transport lactose when an artificial PMF is imposed. Melibiose transport is by sodium symport. The principle is the same, except that a sodium ion is taken up with melibiose instead of a proton. The sodium ion is then expelled from the cell in exchange for a proton by the proton/sodium antiport system. The net result is that one proton enters for each melibiose (but two transport reactions are required). It is possible to isolate mutants of melibiose permease which use lithium ions or protons instead of sodium ions. Thus in principle, any ion which is in excess outside the cell can act as the driving force for transporting nutrients. In practice cells expel sodium but concentrate potassium inside, thus protons and sodium are coupled to nutrient transport of both sugars and amino acids (e.g., proline - protons, glutamic acid - sodium).

c) ATP Linked Active Transport.

Transport of sugars such as maltose. It has been suggested that acetyl phosphate rather than ATP is the donor of high energy phosphate - still unsettled. The maltose G and F proteins lie in the cytoplasmic membrane and transport maltose. The maltose K protein energises them by using ATP.

d) Group Translocation.

In this case the nutrient (outside) is chemically altered during transport and a modified nutrient appears inside. Two important examples:

1) Transport of fatty acids: The transport system uses ATP and converts the fatty acid to its CoA ester during uptake. No free acid is found inside the cell.

fatty acid (outside) = fatty acyl CoA (inside).

2) PEP:Sugar Phosphotransferase System: The transport system uses phosphoenolpyruvate as a source of high energy phosphate. Again, no free sugar enter the cell. Known also as vectorial phosphorylation.

sugar (outside) = sugar-6-phosphate (inside).

The Phosphotransferase System (PTS).

The high energy phosphate of phosphoenolpyruvate (PEP) is transferred via three or four proteins to the sugar. The phosphate group is attached to a histidine when on the PTS proteins. Enzyme I, III and HPr are soluble whereas enzyme II is a membrane protein. There are several slightly different types of PTS system, with or without an enzyme III i.e.:

A) Simple form: PEP, EnzI, HPr, EnzII, Sugar:

B) Complex forms: PEP, EnzI, HPr, EnzIII, EnzII, Sugar:

Complex systems are found for glucose, mannose, fructose and sorbitol whereas the simple form is found for mannitol, galactitol, N-acetyl glucosamine and beta-glucosides. Enzyme I and HPr are common to all sugars (except fructose which has its own version of HPr known as FPr), whereas enzymes II and III (if found) are specific for particular sugars. Specificity is often rather low eg the mannose PTS can take up glucose and fructose slowly. The sugar-6-P is always produced except that the fructose PTS produces Fru-1-P. In E. coli only monosaccharides (and derivatives) are transported by the PTS but in other organisms disaccharides may be taken up by the PTS eg lactose in Staphylococcus aureus.

C) Catabolite Repression

Presence of a favored carbon source (e.g., glucose) prevents use of less favored substrates (e.g., lactose). A hierarchy exists among carbon sources:

good: glucose, glucose-6P, gluconate, mannitol

medium: lactose, glycerol, mannose

poor: succinate, xylose, sorbitol, lactate, amino acids

very poor: acetate, fatty acids

Catabolite repression depends largely on the intracellular level of cyclic AMP (range approximately 1 to 10 micromolar). Cyclic AMP is bound by Catabolite Activator Protein (CAP) also known as cAMP Receptor Protein (CRP). The level of CRP is constant. Transcription of catabolite sensitive operons (e.g, the lac operon) requires binding of CRP-cAMP complex to the promoter region. This allows RNA polymerase to bind to and transcribe the operon. Many operons other than lac are subject to control by cyclic AMP (e.g., genes for glycerol, acetate, sorbitol, mannose catabolism, for the formation of flagella and components of the respiratory chain).

The regulation of cyclic AMP levels is due mostly to changes in activity of adenylate cyclase which catalyses:

ATP = cyclic AMP + PPi

How the information that a favored C-source is present is transmitted to cylase is not known for every sugar. The best understood case is for glucose, where the PTS is involved. Mutants lacking Enz-I or HPr are permanently catabolite repressed. Lack of Enz-III or II (for glucose) abolishes catabolite repression. The most likely explanation is that cyclase activity depends on the state of phosphorylation of Enz-III.

III not phosphorylated - cyclase inactive

III highly phosphorylated - cyclase activated

Presence of glucose results in transfer of phosphate from III-P to glucose, consesquently level of phosphorylation of III falls. If glucose is absent III-P accumulates. III-P is an activator of adenylate cyclase. In contrast, III(non-phos) causes inducer exclusion (see below). Note that glucose must be transported and phosphorylated to cause repression. Further breakdown is NOT essential. Thus a-methylglucose can be converted to its phosphate by the glucose PTS and causes catabolite repression. However it cannot be broken down any further by E. coli.

Permanent repression - the "usual" form of repression. Glucose keeps the level of cyclic AMP at a low but constant value.

Transient repression. Addition of glucose to a culture growing on e.g., lactose results in a sudden massive drop in cyclic AMP levels, to below the permanent repression level. After a while, the cyclic AMP levels recover somewhat and level out at the concentration characteristic of permanent repression. The transient effect is mostly due to excretion of cyclic AMP out of the cell into the culture medium. The excretion system is not understood.

Inducer exclusion. When the PTS is actively taking up PTS-sugars other types of sugar transport system are directly inhibited, hence uptake of inducer molecules is reduced. Unphosphorylated Enz-III of the PTS binds to and inhibits the activity of other transport proteins (e.g. lactose permease). Constitutive mutants (lacI ) do not require inducer, so the inducer exclusion effect is not relevant. Consequently, their lac operon expression depends only on cyclic AMP-CRP and they are repressed much less severely by glucose than wild type cells.

Non PTS sugars (e.g., gluconate, glucose 6-P) also cause catabolite repression. The link with adenylate cyclase is not understood. For catabolite repression to occur effectively, the electron transport chain must be functioning. Consequently catabolite repression does not work strongly in fermenting cells under anaerobic conditions. The link between cyclase and electron transport is mysterious.

Breakdown of cyclic AMP to AMP is catalysed by cyclic phosphodiesterase. The level of this enzyme does not vary significantly. Control of cyclic AMP levels is by synthesis or, if a sudden decrease is required, by excretion. Cyclic phosphodiesterase provides a steady state degradation system for cyclic AMP. (If there were no way to get rid of cyclic AMP, regulation of its level by changing the rate of production would be impossible. To some extent, cyclic AMP is diluted out by cell growth, however, this alone is not sufficient for accurate, rapid regulation).

Sugars and Sugar Derivatives

Many sugars, sugar alcohols and sugar amines can be transported and phosphorylated by the PTS system. Galactose and mannose are isomers of glucose. Sorbose and fructose are 2-keto sugars with two terminal alcohol groups - which is why the PTS can make fructose-1-P.

Sorbitol is the sugar alcohol corresponding to both sorbose and glucose. It is also called glucitol. Note that galactitol and mannitol are both symmetrical - i.e. carbon #6 is equivalent to carbon #1. Hence mannitol-1-P is identical to mannitol-6-P. By convention the first name is used as it has the lower number.

However, sugar acids are not substrates for the PTS. One reason is simply that they do not have a 6-hydroxyl group for the PTS to attach the phosphate group to. They are metabolised by separate pathways.

D) Production Of Fructose Diphosphate

PGI is specifically and competitively inhibited by 2-deoxyglucose-6P which is made from 2-deoxyglucose by the phosphotransferase system. PFK is an important control point. A large variety of sugars and derivatives feed into glycolysis via fructose diphosphate (FDP). PFK is activated by AMP and ADP and inhibited by PEP. (In eukaryotes inhibition is generally by ATP). PFK is a tetramer and its Km is altered by the allosteric effectors but the Vm is unchanged. Thus when energy is short (AMP and ADP are high) glycolysis is stimulated. When PEP, a product of glycolysis, is plentiful, glycolysis is slowed down. PFK essentially controls entry of substrate into the glycolytic pathway. Many sugars and their derivatives are converted to fructose diphosphate e.g.:

Disaccharides

Two basic approaches exist:

Lactose

The lactose operon encodes three proteins, LacZ - b-galactosidase, LacY - the lactose permease, and LacA - lactose transacetylase. The operon is repressed by the LacI repressor whose gene is close to, but not part of the lac operon. Presence of inducer inactivates the repressor. Lactose itself is not an inducer and only indirectly induces the lac operon after a small amount has been isomerized to allolactose. This is an isomer of lactose, generated in a side reaction by the low basal levels of b-galactosidase which are found before induction. In the laboratory, IPTG (isopropyl-thio-b-D-galactoside) is often used as inducer. IPTG is not metabolised and is of no use to the cell - it is a gratuitous inducer.

Lactose enters the cell via the inducible lactose permease. Lactose permease is NOT a phosphotransferase. Lactose entry is coupled to the proton motive force. The respiratory chain generates the PMF by pumping H+ ions out from the cell. The lactose permease operates by proton symport. An H+ ion, re-entering the cell provides the energy to move the lactose across the membrane.

Next, beta-galactosidase hydrolyses lactose, giving glucose plus galactose. Glucose formed INSIDE E. coli is phosphorylated by hexokinase (can phosphorylate many sugars) or glucokinase (specific for glucose):

Glucose + ATP = Glucose-6-P + ADP

The galactose induces the gal operon and the three enzymes produced convert galactose to glucose-1-P. Glucose-1-P and glucose-6-P are interconverted by phosphoglucomutase. Hence the overall product is glucose-6-P and from this, fructose-6-P is made, as previously discussed.

External galactose is metabolized in the same way after entering the cell unchanged via galactose permease (ATP linked). The UDP-galactose cycle is primed by the synthesis of UDP-glucose:

Glucose-1-P + UTP = UDP-Glc + PPi

The pyrophosphate is cleaved into 2Pi by pyrophosphatase. Most UDP-sugars are used for biosynthesis. The gal operon has two functions:

a) growth on galactose (or lactose)

b) synthesis of galactose (for cell wall lipopolysaccharide) in cells growing on other carbon sources. During synthesis glucose is converted to UDP-glucose which is then converted to UDP-galactose.

Because of this dual function the gal operon has two promoters. The catabolism promoter is controlled by cyclic AMP-CRP. The biosynthetic promoter is independent of cyclic AMP and allows a cell growing on glucose to still make galactose.

The system for using lactose is rather clumsy and lactose is not a favored carbon source. Comparison of E. coli with Salmonella and other close relatives indicates that E. coli has an extra segment of chromosome not present in other enterobacteria. Sometime in the recent evolutionary past E. coli probably picked up this extra DNA which carries the lac operon and some other genes. Although the lac operon is often used as the classic example of a catabolic system, it is possibly rather an afterthought from the physiological point of view.

Maltose

Maltose is a dimer of glucose (glucose a1Æ4 glucose). There are two maltose operons - malA and malB regions: These are widely separated on the E. coli chromosome:

malA region
malT - activator protein for all mal genes
malP - maltodextrin phosphorylase
malQ - amylomaltase

malB region
malG & malF - inner membrane transport proteins
malK - energy for MalG/F from ATP
malE - periplasmic binding protein
lamB - maltose specific porin in outer membrane (lambda receptor)

The receptor for bacteriophage lambda is an outer membrane porin intended for maltose transport. If the lambda receptor is missing, maltose (a disaccharide) can enter via the general porins, OmpF & OmpC. However, longer oligosaccharides related to maltose (maltotriose with 3 glucose residues, maltotetraose with 4 etc.) need the lambda receptor. Maltose enters the cell unchanged and there is a periplasmic maltose binding protein which conveys maltose from the lambda receptor pore in the outer membrane across the periplasmic space to the transport system in the cytoplasmic membrane.

Once inside the cell, maltose (and longer-relatives) are degraded by maltodextrin phosphorylase. This is more efficient than b-galactosidase since the Glc-Glc bond energy is saved as a Glc-P bond.

Maltose + Pi = Glc + Glc-1-P

Maltotriose + Pi = Maltose + Glc-1-P

Maltose is a breakdown product of starch and glycogen. These polymers are broken down by amylases, e.g. amylomaltase, which chop up the polymers into manageable pieces. The phosphorylase then deals with these pieces. The maltose system differs from the lactose system in four major respects:

When E. coli runs out of nitrogen, or phosphorus for example, but has plenty of sugar it converts the sugar to glucose-1-P and then to ADP-glucose. The ADP-glucose is polymerized to glycogen which is stored. When better times come the glycogen is broken down by amylase and the maltose system is involved in the metabolism of the sugars released from storage. The maltose system thus has two functions a) growth on external maltose etc., b) re-using stored glycogen. This is probably why the maltose system is divided up - to allow separate control of these two functions.

GLYCOLYSIS

Glycolysis is often defined as the breakdown of glucose to pyruvate. It is better to consider glycolysis as starting with fructose diphosphate since a wide range of sugars can feed into glycolysis and fructose diphosphate is the first intermediate common to all.

Overall one monosaccharide yields 2ATP plus 2NADH2. If oxygen is present the NADH2 may be respired, producing 3ATP per NADH2, and in addition pyruvate may be oxidized further by the Krebs cycle yielding the equivalent of 15ATP per pyruvate. In the absence of oxygen the NADH2 from glycolysis cannot be respired. It must be reoxidized somehow in order to recycle the NAD+. This is done by the process of fermentation. Pyruvate is converted to reduced fermentation products which are excreted into the medium.

The Embden-Meyerhof Pathway

The enzymes of glycolysis are soluble in the cytoplasm. All the intermediates are phosphorylated:

a) This conserves energy as high energy phosphate

b) The enzymes only recognize the phosphate derivatives

c) Massive accumulation of any one phosphate derivative tends to interfere with many other reactions in a not very specific way

d) Phosphorylated compounds cross membranes with very great difficulty so that phosphates are trapped inside the cell. This is particularly applicable to phosphates of glycerol, glyceraldehyde, etc., where the parent molecule can diffuse relatively easily across membranes.

e) Most glycolytic enzymes require divalent cations (Mg2+ or Mn2+) to help bind substrates which are negatively charged due to their phosphate groups.

The overall reaction is irreversible but most of the individual steps are reversible. There are two irreversible steps involved in sugar activation: glucose to glucose-6-P by the PTS and fructose-6-P to fructose diphosphate by PFK. There are two other irreversible steps, both involve substrate level phosphorylation - the phosphoglycerate kinase reaction and the pyruvate kinase reaction (see below).

Glycolytic Enzymes	DG (Kcal/mole)	Gene	Map position
phosphotransferase	-11.5	pts	52,24
phosphoglucose isomerase	0.4	pgi	91
phosphofructokinase	-3.4	pfkA	88
FDP aldolase	5.7	fda	63
triosephosphate isomerase	1.83	tpi	88
glyceraldehyde-P dehydrogenase	1.5	gap	39
phosphoglycerate kinase	-4.5	pgk	63
phosphoglyceromutase	0	gpm	--
enolase	0.44	eno	59
pyruvate kinase (2 isoenzymes)	-7.5	pyk	--
lactate dehydrogenase	-6.0	ldh	30

Net Energy Balance:

Glucose ----> 2 Lactate DGo' = -47.0 kcal/mole

2 ADP ----> 2 ATP DGo' = +14.6 kcal/mole

Overall energy released DGo = -32.4 kcal/mole

These figures include recycling the NADH produced in glycolysis by conversion of pyruvate to lactate, a fermentation product. The efficency is about 30% (14.6 kcal saved as ATP out of 47.0 = 31.1%). But remember that this value includes the uptake of glucose from the medium which may be regarded as work done by the cell - i.e., the actual breakdown of glucose is somewhat more efficient than 30%. For glucose giving two pyruvates DGo' is - 35 kcal/mole. However, in this case we still have 2NADH which must be removed somehow. In fermentation, conversion of pyruvate to lactate is coupled to regeneration of NAD. For this DGo' = -6 kcal/mole of pyruvate or -12 kcal/mole of original glucose. Thus, in fermentation a sizeable amount of energy is wasted in order to reoxidize NADH. In respiring cells NADH reoxidation is coupled to oxygen and as a result glycolysis itself becomes more efficient (slightly over 40%).

The fructose diphosphate aldolase reaction has a DGo' of +5.7. The only reason this goes in the forward direction is that the levels of glyceraldehyde-P and DHAP are extremely low and the reaction is thus pulled over. Most aldolases, of which FDA is an example, form Schiff base intermediates. However, bacteria and yeasts have class II aldolases which do not use Schiff bases. Instead they have a Zn2+ ion at the active site and also require K+ for activation.

Glyceraldehyde phosphate dehydrogenase oxidises the aldehyde of glyceraldehyde-P to a carboxyl group. Oxidation of aldehydes releases energy and this allows the formation of an acyl phosphate on the new carboxyl group. If we split this reaction into two parts:

a) RCHO + NAD+ Æ RCOOH + NADH DGo = -10.3

b) RCOOH + Pi Æ RCO-OP DGo = +11.8

Part (a) yields energy and part (b) consumes energy. Overall the reaction has DGo' = 1.5 and is reversible. The reaction involves an acyl enzyme intermediate attached via the -SH group of cysteine. The enzyme can use arsenate instead of phosphate so producing 3-phosphoglyceroyl-1-arsenate. This is unstable and hydrolyses spontaneously to give 3-phosphoglyceric acid. Thus arsenate uncouples oxidation from phosphorylation, and the energy is wasted.

If you eat sodium arsenate you will die because of the destruction of your high energy phosphates by arsenate mediated hydrolysis. If you eat the more popular white arsenic (As2O3, arsenic trioxide, rat poision) or other trivalent arsenic compounds, you will die too, but by a different mechanism, involving inhibition of enzymes with active SH groups.

We are now ready for substrate level phosphorylation. There are two reactions which form ATP, phosphoglycerate kinase and pyruvate kinase. Both 1,3-diphosphoglyceric acid and PEP are substantially more energy rich than ATP and both reactions have equilibria which greatly favor ATP production. Between these two steps 3-phosphoglyceric acid is converted to PEP by two reversible steps - phosphoglyceromutase and enolase. Enolase is killed by fluoride ions, in the presence of phosphate. A fluorophosphate, covalently attached to the active site is formed.

Pyruvate kinase terminates glycolysis. It is allosterically regulated. It is inhibited by excess ATP and activated by fructose diphosphate - an example of feed forward activation (or positive feedback).

OXYGEN TOXICITY

Anaerobes grow in the absence of oxygen. Organisms which can grow in the presence or absence of oxygen are facultative anaerobes. Obligate anaerobes grow only in the absence of oxygen. Many enzymes involved in anaerobic metabolism are inhibited by oxygen. In some cases the inhibition is reversible and removal of oxygen allows recovery. In other cases the enzymes are permanently killed by oxidation. Oxygen is, in fact, toxic to all organisms. In practice aerobes protect themselves against oxygen toxicity by catalase and superoxide dismutase which destroy toxic peroxides and superoxides. If the concentration of oxygen is greatly increased over the normal levels found in air the oxygen protection system of aerobes become overloaded - hyperbaric oxygen toxicity.

Oxygen toxicity is largely due to the production of superoxide ions (•O-O-) and hydroxyl radicals (HO•). Superoxide ions are generated as a side product of the operation of flavoprotein oxidases. The normal reaction of these enzymes is: FADH2 + O2 + XH Æ FAD + H2O + X-OH

However, the reaction is a two electron transfer but occurs in two single electron stages. The intermediate in the reduction of oxygen is superoxide, which may be released by mistake: O2 + e- Æ •O-O-

Superoxides are extremely reactive and toxic. Superoxide dismutase (SOD) is the fastest enzyme known and protects cells by breaking down superoxide to give hydrogen peroxide: 2•O-O- + 2H+ Æ H2O2 + O2

Most bacterial and mitochondrial SODs are iron or manganese containing enzymes and are all homologous in amino acid sequence. The SOD of eukaryotic cytoplasm is a Cu2+/Zn2+ enzyme and has a totally different sequence.

Hydrogen peroxide is also toxic but not nearly so rapid or dangerous in its reactions as superoxide. The main danger from hydrogen peroxide is its conversion to hydroxyl radicals by reaction with transition metal ions such as Fe2+. H2O2 + Fe2+ = HO• + HO- + Fe3+

Thus we must dispose of the H2O2 before it gives rise to hydroxyl radicals. There is no enzymatic protection mechanism against hydroxyl radicals, once made. Catalase is important in destroying H2O2. It is a heme enzyme which catalyses the reaction: 2H2O2 Æ 2H2O + O2

Peroxidases are also heme enzymes, but of little importance in prokaryotes and of no significance in preventing oxygen toxicity. They catalyse the reaction: XH2 + H2O2 = X + 2H2O

Note that there are three types of heme enzymes:

In addition to the above enzymes, most cells contain an excess of sulfhydryl compounds such as the tripeptide glutathione. This consists of glutamate, cysteine and glycine linked by peptide bonds. The glutamate is linked to the cysteine by its side chain carboxyl group (not its alpha-COOH). Glutathione and other -SH compounds act as scavengers for superoxides, peroxides and free radicals being oxidized themselves to the disulfides. The disulfides are then reduced back to the sulfhydryl derivatives by glutathione reductase, which uses NADPH.

FERMENTATION

In the absence of oxygen or any alternative electron acceptors, NADH2 is reoxidized by reactions which do not involve any electron transport chain. During fermentation ATP is formed only by substrate level phosphorylation. Most of the carbon source is wasted as fermentation products.

Fermentation Pathway in E. coli

The mixed acid fermentation of E. coli illustrates the key points involved in fermentations generally.

Enzymes of Fermentation: Encoded by Gene:

LDH = lactate dehydrogenase ldhA

ACDH = acetaldehyde dehydrogenase adhE

ADH = alcohol dehydrogenase adhE

FDH = formate dehydrogenase fdhF

HYD = hydrogenase hyd

PFL = pyruvate formate lyase pfl

ACK = acetate kinase ack

PTA = phosphotransacetylase pta

In E. coli there are two alternatives. In acid conditions the major fermentation product is lactate. LDH is induced when conditions are both anaerobic and acidic. Lactate dehydrogenase is also activated by pyruvate in a hyperbolic manner. Eukaryotic muscle (when in temporary oxygen deficit) and homofermentative lactic acid bacteria possess only the fermentative pathway leading to lactate.

When the medium is not acidic E. coli produces a mixture of ethanol and acetate, together with formic acid. If E. coli grows under intermediate conditions then both the lactate and alcohol fermentations occcur simultaneously, hence the term "mixed acid fermentation".

Pyruvate formate lyase is responsible for decarboxylation of pyruvate to acetyl-CoA during fermentation in E. coli NOT pyruvate dehydrogenase. The product of PFL is not CO2 but HCOOH. This may be converted to CO2 plus 2(H) by formate dehydrogenase. The 2(H), attached to an unknown carrier may be released as molecular H2 by hydrogenase. Conversion of formic acid to gas (CO2 + H2) is favored by acid conditions.

The acetyl-CoA formed by PFL is converted to a mixture of ethanol and acetate. Ethanol production consumes 4(H) per pyruvate. Acetate production consumes 0(H) per pyruvate. A 50:50 mixture of ethanol plus acetate production therefore disposes of 2(H) per pyruvate. The advantage of this system over lactate production is that an extra ATP is produced by acetate kinase. The enzymes for interconverting acetate and acetyl-CoA are expressed constitutively. In anaerobic cells acetate is excreted, in aerobic cells acetate is taken in and oxidised via the Krebs cycle.

Alcohol dehydrogenase and acetaldehyde dehydrogenase are induced anaerobically in response to build up of reduced NADH. Both enzyme activities are found as part of the same, very large protein, encoded by the adhE gene. Some PFL is present in aerobic cells but it is inactive. Under anaerobic conditions the level of PFL is induced but only by about 5-fold. The major change is that the PFL protein is activated by covalent modification. The enzyme PFL activase modifies PFL. Reduced flavodoxin (an FMN-protein) is used and S-adenosylmethionine is split. The details of the activation reaction are obscure, but a free radical is generated at the active site of PFL. The presence of oxygen kills PFL by destroying the free radical. However, normally the AdhE protein also acts as a PFL deactivase and shuts down PFL when aerobic conditions return.

In addition E. coli produces small amounts of succinate - about 5-10% of fermentation products irrespective of conditions. Fumarate is converted to succinate by fumarate reductase.

Other Fermentations