Tools and Applications of Biochemical Engineering Science

56	C. Wittmann

≈ ≈ ≈

Fig. 9. Schematic view of a linear MALDI-TOF MS instrument

3.2.3

Time-of-Flight

In a time-of-flight (TOF) analyzer the time of flight of ions between the ion source and the detector is measured [61]. This requires that the time at which the ions leave the ion source is well-defined. Therefore, ions are either formed by a pulsed ionization method or various kinds of rapid electric field switching. The single discontinuous laser pulses at distinct time points used in MALDI can be ideally combined with time-of-flight mass separation. TOF analyzers thus received increasing interest with the development of MALDI MS. The schematic draw of a linear MALDI-TOF MS is shown in Fig. (9).

Ions formed in the ion source are accelerated in a short electric field of 2 to 20 mm length to a high kinetic energy of about 20–30 keV, whereby ions of the same charge receive the same kinetic energy. During the acceleration phase, ions with lower masses therefore reach higher velocities, compared to ions with higher masses and need a shorter time to reach the detector. The mass to charge ratio is determined from the time elapsed from ion formation to ion arrival at the detector. It is proportional to the square of the time of flight. Flight tubes of TOF analysators are about 0.5 to 3 m in length.

4

Analysis of Metabolite Labeling by MS

The analytes of interest in 13C tracer studies are mainly low molecular metabolites from the cell interior or low molecular weight products excreted into the cell surrounding. The samples from biological systems for labeling measurements frequently exhibit different characteristics that impose specific constraints on the MS techniques used for their analysis. Generally, they are complex mixtures of diverse substances, which may require a separation step prior to MS analysis. The compounds of interest are generally extracted from the biological system, single cells, blood, urine or other biological fluids or obtained from culture supernatants by lyophilization. Often, amounts and concentrations

Metabolic Flux Analysis Using Mass Spectrometry

57

of these analytes in the samples are rather low. In addition to small molecules, also biopolymers such as proteins, glycogen or DNA, are compounds of interest, because they potentially carry valuable labeling information in their monomeric subunits. Their analysis is mainly carried out on the level of the subunits, obtained after hydrolysis of the polymer. The analysis of the labeling of a compound depends on the desired content of information on the labeling degree. The information content increases in the order molar enrichment, mass isotopomer distributions and positional isotopomer distributions. This chapter describes their direct measurement. The molar enrichment can be calculated from the mass or positional isotopomer distribution, and the mass isotopomer distribution itself can be calculated from the positional isotopomer distribution.

4.1

Measurement of Molar Enrichment

The molar enrichment of a compound can be directly measured by IR MS. The 13C molar enrichment can be obtained from the labeling of CO2 , formed after complete combustion of the analyte.

4.2

Measurement of Mass Isotopomer Distributions

MS can measure the ratio between molar fractions of mass isotopomers. The ratio between two mass isotopomer pools of masses m1 and m2 is defined in the present work as intensity ratio Im1/m2 . It is identical with a mass spectral intensity ratio. If more than two mass isotopomer pools are assessed, their relative ratios, normalized to the sum, are named mass isotopomer distribution. The mass distribution of a compound can be thus obtained from the analysis of ions, which contain the intact carbon skeleton of the analyte. In the area of metabolic flux analysis, mass distributions of various metabolites have been assessed by MS. The major method used is GC/MS, whereby the analytes are derivatized into forms with desired physico-chemical properties such as increased volatility, thermal stability and suitable MS properties [62]. The mass of the formed derivate must be sufficiently high (usually above 175 apparent mass units) to avoid background interference [48]. To obtain the 13C mass distribution of a compound, ions with the entire carbon skeleton of the analyte have to be present. For accurate quantification of the mass distribution of such ions, they should occur in high abundance and preferably be unique species, thus being formed by only one fragmentation pathway.

The compounds analyzed by GC/MS comprise e.g. amino acids [34, 39, 64–69], organic acids [33, 63, 65, 66, 69], sugars [39, 70, 71], lipids and fatty acids [72, 73]. Moreover, mass distributions of polymers and their building blocks, obtained via hydrolysis of the polymer, were assessed. Examples are glycogen [39, 70], cell protein [8, 10, 17], or DNA [74]. Most of the analytical methods have been developed for tissue samples. Since most of the compounds studied are polar or even charged molecules, derivatization is necessary in most of the cases of GC/MS analysis. The derivatization method of choice clearly depends

58	C. Wittmann

on the chemical nature of the analyte of interest. Often used derivatizations are methylation, acetylation, silylation and acylation. Instable metabolites such as oxaloacetate can be transformed into stable derivates by suitable derivatization such as oximation, whereby the reagent can be directly supplied in the extraction solvent [75].

In addition to GC/MS, also MALDI MS can be applied for labeling measurements of low molecular weight compounds related to the metabolism. Successful analysis of amino acids, organic acids, and sugars has been demonstrated [18, 25, 27]. Usually, derivatization is not required, which is a major advantage compared to GC/MS. To our knowledge, ESI MS has so far not been directly used for labeling measurements in metabolic flux analysis. However, its application for the analysis of low molecular weight metabolites has been demonstrated by several authors [e.g. 76, 77]. Membrane introduction MS has been used for the quantification of intensity ratios of labeled to non-labeled gases in bioprocesses such as CO2 or CH4 [32, 78].

4.3

Measurement of Positional Isotopomer Distributions

More detailed information on the labeling of a compound can be obtained with MS via additional analysis of fragment ions, which contain only specific parts of the carbon skeleton of the analyte. It should be noticed that the resolution of single positional isotopomer pools not necessarily leads to an increase in information for flux quantification [16]. Mathematically, isotopomer distributions can be obtained from mass distributions of the molecular ion, and the different assessed fragment ions via matrix calculus [8, 9]. By this approach, positional isotopomer distributions can be partially, in some cases even completely resolved. As example, the entire positional isotopomer distribution of pyruvate can be achieved via analysis of the molecular ion and three additional fragments formed in GC/MS analysis of pyruvate as methylester [9] or via analysis of the molecular ion and one fragment ion of derivatized alanine and one fragment ion of derivatized valine, both formed in the metabolism from pyruvate [8]. More complex protocols may be required for analytes containing more carbon atoms, thus exhibiting a substantially increased number of possible positional isotopomers. Di Donato et al. [64] determined 24 out of 32 possible positional isotopomers of glutamate by GC/MS, which was linked to an enormous experimental effort, requiring the synthesis and MS analysis of five different derivates in addition to glutamate. Also other protocols for the partial resolution of positional isotopomer pools are laborious, include various steps for purification and chemical or enzymatic conversion of the analytes [65, 66]. This impedes the broad and routine use of such techniques.

As underlined, the success of positional isotopomer analysis by MS is based on the availability of suitable fragments. Using hard ionization techniques such as EI ionization usually yields a number of fragments automatically. If the information content from the available fragments is too low, other derivates of the analyte, with a different fragmentation pattern, can be analyzed in addition. A major tool to obtain additional fragments is targeted fragmentation. A pro-

Metabolic Flux Analysis Using Mass Spectrometry

59

mising approach for positional isotopomer measurements is MS/MS, also called tandem MS, where selected ions can be isolated, subsequently fragmented and the fragments can be analyzed for their mass distribution (see chapter 2).

5

Application of MS to Metabolic Flux Analysis

The following chapter shows the application of MS to metabolic flux analysis with different examples. Whereas some of them focus on flux quantification of only a single or a few selected reactions, others aim at the analysis of larger parts of the metabolism. The overview given should illustrate the broad application potential of MS for metabolic flux analysis by examples from different fields of research. The majority of studies belongs to the medical field, whereas so far only few examples can be found in the area of biochemical engineering.

5.1

Microorganisms and Fungi

Park et al. showed the activity of pyruvate carboxylase in a pyruvate kinase deficient strain of lysine producing Corynebacterium glutamicum by 13C tracer experiments with NMR and GC/MS [68]. Wittmann and Heinzle [25, 27] applied MALDI-TOF MS to batch cultivations of lysine producing C. glutamicum either on 1-13C glucose or on a mixture of U-13C and naturally labeled glucose, respectively, in combination with metabolite balancing. From stoichiometric data and from selected intensity ratios of secreted lysine, alanine and trehalose, the flux distribution of the central metabolism was determined. During maximum lysine production, the strain exhibited high relative fluxes of 71% into the pentose phosphate pathway and significant anaplerotic formation of oxaloacetate 37%. The two alternative lysine producing branches, the succinylase and the dehydrogenase pathway, respectively were found active, whereby 63% of lysine were formed via the dehydrogenase branch. Glucose 6-phosphate isomerase was found highly reversible. Moreover significant backfluxes from the TCA cycle to the glycolysis causing a withdrawal of the lysine precursor oxaloacetate, were found. The benefits of 13C experiments and MS in addition to metabolite balancing are underlined by the comparison with a study applying only metabolite balancing for flux analysis of the same strain [79]. As described below, metabolite balancing alone does not provide sufficient information to resolve key fluxes in the central metabolism of C. glutamicum. As example metabolite balancing cannot discriminate between the two alternative routes in the lysine biosynthesis. In contrast, sufficient information on the flux partitioning in the lysine biosynthesis is provided by MS measurement of only one lysine intensity ratio [27]. The flux partitioning ratio between glycolysis and pentose phosphate pathway is accessible via metabolite balancing including a balance of NADPH, which is often found uncertain as described above. The flux distribution between glycolysis and PPP can be sensitively quantified by MS from only one measured intensity ratio of lysine [27]. In contrast to tracer experiments with MS, no information could be obtained by metabolite balancing on bidirectional

60	C. Wittmann

fluxes catalyzed by reversible enzymes such as glucose 6-phosphate isomerase or on cycling fluxes between C4 units of the TCA cycle and C3 units of the glycolysis.

Christensen et al. [17] determined flux distributions in the central metabolism of low and high-yielding strains of Penicillium chrysogenum applying chemostat cultures with 1-13C glucose. The authors analyzed the labeling pattern of amino acids in hydrolyzed cell protein by GC/MS, which provides a high amount of labeling information. Recently Gombert et al. [80] applied GC/MS for labeling analysis and quantified metabolic fluxes in the central metabolism of S. cerevisiae with 13C glucose as substrate. The authors found significant differences between cells grown in batch and in continuous culture. The relative flux through the pentose phosphate pathway was markedly higher in continuous culture compared to batch culture. A knockout of the MIG1 gene caused no changes in the intracellular flux distribution. A combined approach of 13C labeled fructose, glutamate or aspartate and NMR and GC/MS for labeling analysis was used to study the metabolism of actinomycin producing Streptomyces parvulus [81]. Labeling analysis of the actinomycin D peptide ring allowed the specification of the origin of its five amino acids. Kinetic studies in anaerobic mixed cultures were performed by Dornseiffer et al. [82] in a miniaturized bioreactor equipped with a membrane mass spectrometer with 13C labeled bicarbonate and subsequent measurement of 12CH4 and 13CH4 by membrane MS. Flux analysis of Saccharomyces cerevisiae with 13C and 18O labeled glucose as tracer substrate and measurement of different mass isotopes of CO2 was carried out in the same measurement cell [32].

5.2

Tissue Cultures

A number of authors have applied 13C labeling experiments with MS to metabolic flux analysis in tissues such as liver, kidney, brain, and heart [83]. The inherent complexity of the networks including phenomena such as compartmentation, metabolite channeling and complex dilution effects of the labeling usually allow the determination of only selected flux parameters by a tracer experiment. One application of stable isotopes and MS is the in vivo quantification of the biosynthesis of polymers. For this purpose, a mathematical approach, recently reviewed by Hellerstein and Neese [84], was developed, that allows the determination of the labeling of the actual precursor molecules of a particular polymer. With this approach, the biosynthesis of proteins [85], lipids [86], and steroids [87] was studied. It was also applied to quantify gluconeogenesis [88]. Extensive work has been carried out on quantifying fluxes in the central metabolism of tissue cultures as exemplified by studies on the pentose phosphate pathway in human hepatoma cells [89], or on anaplerosis and the TCA cycle in perfused rat liver [64, 66, 69], and rat heart [90, 91]. Fluxes in the central metabolism of different mammalian cell lines were quantified by Lin et al. [71].

Metabolic Flux Analysis Using Mass Spectrometry

61

5.3

In vivo Studies in Animals and Humans

Tracer studies with stable isotopes have a long tradition in medical biochemistry. 13C, 15N or 18O do not cause adverse physiological effects even at higher enrichments and are therefore especially suitable for in vivo studies in animals and humans. The quantification of metabolite turnover rates in vivo is a typical application field as reviewed by Wolfe [92]. In vivo studies on the glucose metabolism including gluconeogenesis, hepatic glucose cycling, glycogen synthesis and breakdown were conducted by several authors [e.g. 38, 39, 65, 70, 93, 94]. Additional studies were performed on the fatty acid and lipoprotein metabolism [46].

6

Conclusion and Future Perspectives

As shown in this review tracer experiments and MS are valuable tools for the analysis of metabolic fluxes in various different biological systems. Metabolic flux analysis by MS plays a central role in biochemical network characterization. Future perspectives of this rapidly developing field are seen in different areas. The first attributes to the fact that the real value of metabolic flux analysis is its comparative application, e.g. to different physiological situations or to different strains. However, metabolic flux analysis as a routine tool is currently not available. Thus, alterations of fluxes during industrially relevant batch or fed-batch cultivations or differences of flux distributions between wild type and genetically engineered over-producing strains are still widely unknown. Concerning the experimental methods, a major challenge will be therefore the simplification, miniaturization and automatization of the MS measurements for flux analysis, whereby especially novel techniques such as ESI MS or MALDITOF MS exhibit a high potential. An important requirement for broad and efficient flux analysis is the availability of general modeling frameworks with tools for model generation, experimental design, parameter estimation and statistical analysis. An interesting future direction is the elucidation of metabolic fluxes under dynamic conditions aiming at the determination of in vivo kinetics of enzymes and of metabolic regulation. MS exhibits a high sensitivity and an excellent applicability to both concentration and labeling measurements of intracellular metabolites, providing information on intracellular pool sizes and flux distributions, which makes it a promising tool to study flux dynamics, including the resolution of time-dependent flux alterations linked to the cell cycle or cell differentiation.

62	C. Wittmann

References

1.Sauer U, Hatzimanikatis V, Hohmann HP, Manneberg M, Van Loon AP, Bailey JE (1996) Appl Env Microbiol 62:3687

2.Hult K, Veide A, Gatenbeck S (1980) Arch Microbiol 128:253

3.Sazanov LA, Jackson JB (1994) FEBS Lett 344:109

4.Szyperski T (1998) Q Rev Biophys 31:41

5.Schmidt K, Marx A, de Graaf AA, Wiechert W, Sahm H, Nielsen J, Villadsen J (1998) Biotechnol Bioeng 58:254

6.Bonarius HP, Timmerarends B, de Gooijer CD, Tramper J (1998) Biotechnol Bioeng 58: 258

7.Christensen B, Nielsen J (2000) Adv Biochem Eng Biotechnol 66:209

8.Christensen B, Nielsen J (1999) Metab Eng 1:282

9.Wittmann C, Heinzle E (1999) Biotechnol Bioeng 62:739

10.Dauner M, Sauer U (2000) Biotechnol Prog 16:642

11.Spengler B (2001) The basics of matrix-assisted laser desorption ionization time-of flight mass spectrometry and post source decay. In: James P (ed) (2001) Proteome research: Mass spectrometry. Springer, New York, p 33

12.Sherry AD, Sumegi B, Miller B, Cottam GL, Gavva S, Jones JG, Malloy CR (1994) Biochemistry 33:6268

13.Debnam PM, Shearer G, Blackwood L, Kohl DH (1997) Eur J Biochem 246:283

14.Schmidt K, Carlsen M, Nielsen J, Villadsen J (1997) Biotechnol Bioeng 55:831

15.Kelleher JK (1999) Am J Physiol 277:E395

16.Wiechert W, Moellney M, Isermann N, Wurzel M, de Graaf AA (1999) Biotechnol Bioeng 66:69

17.Christensen B, Thykaer J, Nielsen J (2000) Appl Microbiol Biotechnol 54:212

18.Wittmann C, Heinzle E (2001) Biotechnol Bioeng 72:642

19.Lee WN, Byerley LO, Bergner EA, Edmond J (1991) Biol Mass Spectrom 20:451

20.Fernandez CA, Des Rosiers C, Previs SF, David F, Brunengraber H (1996) J Mass Spectrom 1996 31:255

21.Jones PJ, Leatherdale ST (1991) Clin Sci (Colch) 80:277

22.Wiechert W (1996) Habilitation Thesis, University of Bonn, Jülicher Forschungsbericht 3301, ISSN 0944-2952 (in German)

23.Moellney M, Wiechert W, Kownatzki D, de Graaf AA (1999) Biotechnol Bioeng 66:86

24.Zupke C, Stephanopoulos G (1994) Biotechnol Prog 10:489.

25.Wittmann C, Heinzle E (2000) Metabolic flux analysis of lysine producing Corynebacterium glutamicum by MALDI-TOF MS, 3rd European Symposium on Biochemical Engineering Sciences, Copenhagen, Denmark

26.Wittmann C, Heinzle E (2001) Metab Eng 3:171–190

27.Wittmann C, Heinzle E (2001) Eur J Biochem 268:2441–2455

28.Malloy CR, Sherry AD, Jeffrey FMH (1988) J Biol Chem 263:6964

29.Lee WN (1993) J Biol Chem 268:25522

30.Szyperski T (1995) Eur J Biochem 232:433

31.Klapa MI, Park SM, Sinskey AJ, Stephanopoulos G (1999) Biotechnol Bioeng 62:375

32.Luxenburger H, Wittmann C, Heinzle E (1998) Metabolic flux studies in Saccharomyces cerevisiae using dynamic membrane mass spectrometry. 2nd Conference on Metabolic Engineering, Elmau, Germany.

33.Erecinska M, Zaleska MM, Nissim I, Nelson D, Dagani F, Yudkoff M (1988) J Neurochem 51:892

34.Schadereit R, Krawielitzki K (1998) Isotopes Environ Health Stud 34:127

35.Doumit J, Le J, Frey J, Chamson A, Perier C (1999) Amino Acids 16:107

36.Yudkoff M, Daikhin Y, Nissim I, Nissim I (2000) Neurochem Int 36:329

37.Ben-Yoseph O, Kingsley PB, Ross BD (1994) Magn Reson Med 32:405

38.Weber JM, Klein S, Wolfe RR (1990) J Appl Physiol 68:1815

39.Hellerstein MK, Neese RA, Linfoot P, Christiansen M, Turner S, Letscher A (1997) J Clin Invest 1997 100:1305

Metabolic Flux Analysis Using Mass Spectrometry

63

40.Lee WN, Bassilian S,Ajie HO, Schoeller DA, Edmond J, Bergner EA, Byerley LO (1994) Am J Physiol 266:E699

41.Landau BR,Wahren J, Chandramouli V, Schumann WC, Ekberg K, Kalhan SC (1995) J Clin Invest 95:172

42.Costello CE (1997) Biophys Chem 68:173

43.Matsuo T, Seyama Y (2000) J Mass Spectrom 35:114

44.James P (ed) (2001) Proteome research: Mass spectrometry, Springer New York

45.Johnson RC, Cooks RG, Allen TM, Cisper ME, Hemberger PH (2000) Mass Spectrom Rev 19:1

46.Brenna JT (1994) Acc Chem Res 27:340

47.Corso TN, Brenna JT (1997) Proc Natl Acad Sci 94:1049

48.Thompson GN, Pacy PJ, Ford GC, Halliday D (1989) Biomed Environ Mass Spectrom 18:321

49.Caimi RJ, Brenna JT (1993) Anal Chem 65:3497

50.Karas M, Hillenkamp F (1988) Anal Chem 60:2299

51.Tanaka K, Waki H, Ido Y, Akita S, Yashida Y, Yoshida Y (1988) Rap Comm Mass Spectrom 8:151

52.Moore WT (1997) Methods Enzymol 289:520

53.Duncan MW, Matanovic G, Cerpa-Poljak A (1993) Rapid Commun Mass Spectrom 7:1090

54.Kang MJ, Tholey A, Heinzle E (2000) Rapid Commun Mass Spectrom 14:1972

55.Zhan Q, Gusev A, Hercules D M (1999) Rapid Commun Mass Spectrom 13:2278.

56.Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM (1989) Science 246:64

57.Emmet MR, Caprioli RM (1994) J Am Soc Mass Spectrom 5:605

58.Edmonds CG, Loo JA, Loo RR, Udseth HR, Barinaga CJ, Smith RD (1991) Biochem Soc Trans 19:943

59.Parker CE, Perkins JR, Tomer KB, Shida Y, O’Hara K (1993) J Chromatogr 616:45

60.Cooks RG, Kaiser RE (1990) Acc Chem Res 23:213

61.Cotter RJ (1989) Biomed Environ Mass Spectrom 18:513

62.Tsikas D (1998) J Chromatogr B Biomed Sci Appl 717:201

63.Nissim I, Nissim I, Yudkoff M (1990) Biochim Biophys Acta 1033:194

64.Di Donato L, Des Rosiers C, Montgomery JA, David F, Garneau M, Brunengraber H (1993) J Biol Chem 268:4170

65.Katz J, Wals P, Lee WN (1993) J Biol Chem. 268:25509

66.Des Rosiers C, Di Donato L, Comte B, Laplante A, Marcoux C, David F, Fernandez CA, Brunengraber H (1995) J Biol Chem 270:100 27

67.Neese RA, Schwarz JM, Faix D, Turner S, Letscher A, Vu D, Hellerstein M (1995) J Biol Chem 270:14452

68.Park SM, Shaw-Reid C, Sinskey AJ, Stephanopoulos G (1997) Appl Microbiol Biotechnol 47:430

69.Des Rosiers C, Fernandez CA, David F, Brunengraber H (1994) J Biol Chem 269:27179

70.Katz J, Lee WN, Wals PA, Bergner EA (1989) J Biol Chem 264:12994

71.Lin YY, Cheng WB, Wright CE (1993) Anal Biochem 209:267

72.Lligona Trulla L, Magistrelli A, Salmona M, Tacconi MT (1992) Melanoma Res 2:235

73.Hellerstein MK, Schwarz JM, Neese RA (1996) Annu Rev Nutr 16:523

74.Macallan DC, Fullerton CA, Neese RA, Haddock K, Park SS, Hellerstein MK (1998) Proc Natl Acad Sci USA 95:708

75.Laplante A, Comte B, Des Rosiers C (1995) Anal Biochem 224:580

76.Gerber SA, Scott CR, Turecek F, Gelb MH (1999) J Am Chem Soc 121:1102

77.Reetz MT, Becker MH, Klein HW, Stöckigt D (1999) Angew Chem 111:1872

78.Meyer B, Heinzle E (1998) Biotechnol Bioeng 57:127

79.Vallino JJ, Stephanopoulos G (1993) Biotechnol Bioeng 41:633

80.Gombert AK, Moreira Dos Santos M, Christensen B, Nielsen J (2001) J Bacteriol 183:1441

81.Inbar L, Lapidot A (1991) J Bacteriol 173:7790

82.Dornseiffer P, Meyer B, Heinzle E (1995) Biotechnol Bioeng 45:219

83.Lee K, Berthiaume F, Stephanopoulos G, Yarmush ML (1999) Tissue Eng 5:347

64	C. Wittmann

84.Hellerstein MK, Neese RA (1999) Am J Physiol 276:E1146

85.Papageorgopoulos C, Caldwell K, Shackleton C, Schweingrubber H, Hellerstein MK (1999) Anal Biochem 267:1

86.Hellerstein MK, Christiansen M, Kaempfer S, Kletke C, Wu K, Reid JS, Mulligan K, Hellerstein NS, Shackleton CH (1991) J Clin Invest 87:1841

87.Kelleher JK, Kharroubi AT, Aldaghlas TA, Shambat IB, Kennedy KA, Holleran AL, Masterson TM (1994) Am J Physiol 266:E384

88.Neese RA, Schwarz JM, Faix D, Turner S, Letscher A, Vu D, Hellerstein MK (1995) J Biol Chem 270:14452

89.Lee WN, Boros LG, Puigjaner J, Bassilian S, Lim S, Cascante M (1998) Am J Physiol 274:E843

90.Comte B, Vincent G, Bouchard B, Des Rosiers C (1997) J Biol Chem 272:26117

91.Laplante A, Vincent G, Poirier M, Des Rosiers C (1997) Am J Physiol 272:E74

92.Wolfe RR (1992) Radioactive and stable isotope tracers in biomedicine: Principle and practice of kinetic analysis. Whiley-Liss, New York

93.Kalderon B, Korman SH, Gutman A, Lapidot A (1989) Am J Physiol 257:E346

94.Katz J, Lee WN (1991) Am J Physiol 261:E332

Received: April 2001

The Molecular Mechanism of ATP Synthesis

by F1F0-ATP Synthase: A Scrutiny of the Major Possibilities

Sunil Nath

Department of Biochemical Engineering and Biotechnology, Indian Institute of Technology, Hauz Khas, New Delhi 110 016, India

E-mail: [e-mail]

Dedicated to Prof. Dr. Wolf-Dieter Deckwer on the occasion of his 60th birthday

There is a complicated hypothesis which usually entails an element of mystery and several unnecessary assumptions. This is opposed by a more simple explanation which contains no unnecessary assumptions. The complicated one is always the popular one at first, but the simpler one, as a rule, eventually is found to be correct. This process frequently requires 10 to 20 years. The reason for this long time lag was explained by Max Planck. He remarked that “Scientists never change their minds, but eventually die.”

J.H. Northrop

A critical goal of metabolism in living cells is the synthesis of adenosine triphosphate (ATP). ATP is synthesized by the enzyme F1F0-ATP synthase. This enzyme, the smallest-known molecular machine, couples proton translocation through its membrane-embedded, hydrophobic domain, F0 , to the synthesis of ATP from adenosine diphosphate (ADP) and inorganic phosphate (Pi) in its soluble, hydrophilic headpiece, F1. Animals, plants and microorganisms all capture and utilize energy by this important chemical reaction. How does it occur? The binding change mechanism and the torsional mechanism of energy transduction and ATP synthesis are two mechanisms that have been proposed in the literature.According to the binding change mechanism (which considers reversible catalysis and site-site cooperativity), energy is required primarily for release of synthesized ATP, but not for its synthesis. On the other hand, according to the torsional mechanism (which considers an irreversible mode of catalysis and absence of cooperativity), all the elementary steps require energy, and the ionprotein interaction energy obtained from the ion gradients is used to synthesize ATP, for Pi binding, and for straining the b-e bond in order to enable ADP to bind. The energy to release preformed ATP from the tight catalytic site (bDP) is provided by the formation of the b-e ester linkage. First, the central features of these mechanisms are clearly delineated. Then, a critical scrutiny of these mechanisms is undertaken. The predictions of the torsional mechanism are listed. In particular, how the torsional mechanism deals with the specific difficulties associated with other mechanisms, and how it seeks to explain a wealth of structural, spectroscopic, and biochemical data is discussed in detail. Recent experimental data in support of the mechanism are presented. Finally, in view of the molecular machine nature of energy transduction, the indispensability of applying engineering tools at the molecular level is highlighted. This paves the way for the development of a new field: Molecular Physiological Engineering.

Keywords. ATP synthase, Oxidative phosphorylation, Binding change mechanism, Torsional mechanism, Molecular physiological engineering

Advances in Biochemical Engineering/

Biotechnology, Vol. 74

Managing Editor: Th. Scheper

© Springer-Verlag Berlin Heidelberg 2002