- Email: [email protected]
Transport catalysis
Biochimica et Biophysica Acta 1757 (2006) 1229 – 1236 www.elsevier.com/locate/bbabio
Transport catalysis Martin Klingenberg Institute Physiological Chemistry, University of Munich, Schillerstr.44, 80336 München, Germany Received 13 February 2006; received in revised form 6 April 2006; accepted 7 April 2006 Available online 19 April 2006
Abstract Carrier linked solute transport through biomembranes is analysed with the viewpoint of catalysis. Different from enzymes, in carriers the unchanged substrate induces optimum fit in the transition state. The enhanced intrinsic binding energy pays for the energy required of the global conformation changes, thus decreasing the activation energy barrier. This “induced transition fit” (ITF) explains several phenomena of carrier transport, e.g., high or low affinity substrate requirements for unidirectional versus exchange, external energy requirement for “low affinity” transport, the existence of side specific inhibitors to ground states of the carrier, the requirement of external energy in active transport to supplement catalytic energy in addition to generate electrochemical gradients. © 2006 Elsevier B.V. All rights reserved. Keywords: Transport; Catalytic energy; Induced fit; Active transport; Electrochemical gradient
Whereas the concept of catalysis is well conceived and widely exploited in enzymes [1,2], it is infrequently applied with its ramifications to transport [1,3–7]. For illustrating the crucial elements of carrier mediated catalysis a comparison to familiar enzymatic catalysis is useful [4,8,9]. Reaction theory requires that the catalyst, e.g., the enzyme, assist in transforming the substrate into a transition state. The binding site is constructed not to bind optimally the substrate but to fit the distorted substrate in the transition state. Thus by using energy generated from the optimum binding in the transition state, the substrate is readied for the subsequent and spontaneous chemical alteration. In contrast, carriers catalyse transport of substrates through membranes without chemical changes and therefore the concept of catalysis has received little attention in the transport field. In particular, it was not clear what a transition state as an essential ingredient of catalysis would mean and look like. Pertinent models of carriers were concerned with the global changes comprising the binding site reorientation between the “out” and “in”-side configurations. But the question was barely addressed, what happens in between these two “ground states”. The single binding center gated pore model (SBGP) has been the underlying mechanism for the present analysis of transport catalysis [10–13]. It provides a good example for simple single substrate enzyme catalysis. The SBGP was first established on E-mail address: [email protected]. 0005-2728/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbabio.2006.04.010
the molecular level with the AAC using specific inhibitor ligands. With two different, side specific inhibitors the carrier population could be entirely transferred from the out- to the in-side ground state and vice-versa. The presence of ADP or ATP was obligatory as catalysts of this transition coupled to the translocation of one substrate molecule. The SBGP is manifest in a wide range of transporters, encompassing unidirectional facilitators, exchange type transporters, and primary or secondary active transporters. The emergent structures of carriers of widely different provenance provide a striking support to the SBGP model and discard tandem and multiple binding site models [14–20]. Based on extensive observations in the mitochondrial membrane, in the isolated state, and in the reconstituted system [21,22], the intricacies of the transition between the “c” (out) and “m”(in) states of the mitochondrial ADP/ATP carrier (AAC), with often seemingly contradictory results, challenged us to comprehend the underlying mechanism of transport catalysis. In line with reaction theory and in analogy to enzymatic catalysis, we introduced the transition state, where the carrier–substrate complex assumes an intermediate configuration between the two ground states [4,8]. Here the substrate induces an optimum fit and thus generates a maximum intrinsic binding energy which drives the carrier substrate complex into and through the transition state between the ground states. With the advent of crystal structures of some carriers, molecular modelling of the transport mechanism and the forces
1230
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
involved become more precise and challenging. But in recent models based on 3D structures, intermediate steps between the in and out states were not considered or only marginally described [15,23,24]. Here we want to demonstrate that the concept of the induced transition fit (ITF) may provide an important guide to interpret the substrate–protein interactions involved in the translocation process. 1. Enzyme and carrier catalysis cycles A comparison of the simple catalytic cycles of enzyme and carriers illustrate similarities and differences between these two protein mediated types of catalysis (Fig. 1). Whereas after product release the enzyme is recuperated in a state ready for the next catalytic cycle, the carrier transformed after release of the substrate and the catalytic cycle is complete only after the second “return” branch. In balance, the carrier catalyses a vectorial reaction of the substrate, by facilitating the translocation through a diffusion barrier. In some way, the roles of the protein and the substrate are reversed: the substrate remains chemically unchanged but the protein has assumed a different configuration with opposite access paths to the binding center. One may regard the substrate rather than the carrier as the catalyst.
Fig. 1. Catalytic cycles for a simple enzyme reaction and for unidirectional (uniport) and exchanging (antiport) carrier transport. During the cycle, the carrier assumes two different ground states, with the binding site directed outside or inside. As a result, the full carrier cycle has two transformation branches. At an intermediate stage, the carrier enters the transition state.
Whereas in the analysis of enzyme mechanism, the transition state is well conceived, it is not obvious to introduce this pivotal step to transport. Indeed, the need for introducing the transition state becomes more rational only by considering the often important changes in the carrier structure during the translocation. Evidently, these global structural shifts require intermediate steps, some of them with the highest energy level qualifying as classical transition states. Most importantly, this catalysis is highly selective for the substrate structure, often surpassing that of enzymes dealing with the same substrate. In enzymes in the transition state, the interaction between a distorted substrate and an adapted binding site is optimised [25]. However, the conformation changes are limited and focussed on the catalytic centre. In carriers, the substrate is unchanged but the protein undergoes a global conformation change in the transition state, midway in the transformation between the inand out-states. How is the energy for these important changes generated? Similar as in enzymes, the substrate is not optimally bound in the ground states but, different from enzymes, the unchanged substrate induces a major reorganisation in the binding centre going into the transition state to attain an optimum fit (Fig. 2). In the concomitant global structure change, the gates are partially opened and closed respectively. This substrate induced optimum fit of the binding site leading to the transition state (induced transition fit, ITF) is the essence of the carrier catalysis. An energy reaction profile illustrates the implications of the ITF (Fig. 3). At first, substrate binding releases little energy because of poor fit to the ground state. On proceeding into the transition state, the increased substrate protein interaction forces produce a larger “intrinsic” binding energy, The energy costs of the concomitant reconfigurations of the carrier, shown by the high energy barrier in the substrate less transition state (CT), are paid for by the intrinsic binding energy such that in balance the energy barrier for the carrier substrate complex in the transition state is low. The energy difference of the carrier transition states without and with substrate corresponds to “catalytic energy” which is responsible for the substrate linked rate acceleration. In other words the catalytic energy necessary for transport is recruited from the substrate–protein interaction forces reaching a maximum in the transition state. Also in enzymes substrates can induce changes in the active center, which may even propagate to some other “background” residues along hydrogen and ionic bond relays [25]. But those changes are generally small and centred, whereas in substrate induced transition fit (ITF) of carriers the changes are large and global, focused not only on the binding center but also on the gating region. Relays of interacting residues emanate from the binding center towards the gating regions along the translocation path as illustrated in Fig. 2. Hereby, intermediate states may be partially stabilised as shown in the energy reaction profile of Fig. 3. Although extensive regions of the protein are modified, these changes follow a precise trajectory in a multidimensional energy space, coordinating all residues involved. Changes may involve translational and rotational movements, bending and tilting of transmembrane helices,
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
1231
Fig. 2. The induced transition fit (ITF) for the single binding center gated pore. Only first half of steps leading to the transition state are shown and not the analogous steps to the internal ground state. In the ground states, the substrate only loosely fits to the binding site. In the transition state, the substrate reaches optimum fit by inducing more binding contacts and thus global changes in the flexible structure which lead to partial opening and closing of the opposite gates.
triggered by the substrate induced residues shifts at the binding center. The intrinsic binding forces produced in the transition state compensate the resulting intra-carrier strains. The size of these changes may depend among other factors on the size of the substrate molecule that defines the width of the binding niche and the access paths (Fig. 4). Thus, with larger substrates the conformation changes on closing and opening the gates require more extensive and complex rearrangements, resulting in higher energy barriers, more intermediate steps, and lower transport rates. 2. Implications of the ITF By defining the energetic constraints for the working of carriers, the ITF determines a variety of transport parameters. According to the ITF, exchange type of transport is the most efficient transport mode by supplying binding energy to both branches of the transport cycle. In unidirectional transport the substrate less branch should encounter a markedly higher barrier and be rate limiting. In the classical case of the well studied glucose facilitator the turnover is strongly accelerated when substrate is provided to both sides of the membrane [26]. Severe constraints are imposed on unidirectional transport: the activation barrier of the free carrier must be comparatively low, in order to allow a sufficient transport rate which requires some flexibility of the carrier structure even without substrate. The conformation changes must be small and as a consequence only small and weakly binding substrates should be able to accommodate the unidirectional facilitated transport. Because of lower binding force the transport should be less selective and tolerate broader specificity.
3. Active transport As a result of these restrictions most unidirectional transport systems are found among transporters driven by external energy, either in primary or secondary active transport. According to common wisdom in active transport external energy is to drive substrates “uphill”, against an electrochemical gradient. However, there are cases where the transport is not obviously uphill and yet coupled to energy supply, apparently contradicting the common dogma. The best studied cases of “energy losing” systems are among the ABC transporter family, where the MDR (multi drug resistance) transporters with its paradigm Pglycoprotein (Pgp) (see review [27]) may exemplify the role of external energy supply in transport catalysis according to the roles of the ITF mechanism. Crystal structures of a bacterial MDR member, the lipid A flippase (MsbA), show an open crevice formed by two monomers in the dimer MsbA, apparently ready to accept large molecules from the cytoplasmic side of the membrane [28,29]. MDR transporters exhibit a broad specificity for their hydrophobic substrates, indicating that binding is comparatively weak. On the other hand, the large size of the substrates handled by the MDR transporters requires comparatively large conformation changes. As a result, there exists a significant discrepancy between the energy generated from binding and the catalytic energy required for the large structural change in opening and closing the necessarily wide channels. The coupling of transport to the hydrolysis of ATP [30] is suggested to compensate this energy deficit (Fig. 5A). It seems now well established that ATP hydrolysis at the nucleotide binding domains causes a drastic reconfiguration of this cytosolic orientated domain, (see reviews [19,20,31]) which is
1232
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
Fig. 3. Energy profile of carrier catalysis. Initial substrate binding releases low binding energy. In the transition state (intrinsic) binding energy is high due to the optimum fit. This intrinsic binding energy attenuates the high energy barrier of the free carrier (CT) to the low level of the transition state complex CTST. Thus, catalytic energy is recruited from the high intrinsic binding energy in the transitions state. The energy profile also illustrates intermediate steps to the transition state, caused by the mobile network of interacting residues that propagate the reconfiguration into the transition state (see Fig. 2).
transferred by interdomain contact to the transmembrane domain, thus facilitating the transformation of the transporting domain between the two ground states. In other words, catalytic energy is obtained from ATP hydrolysis via inter-domain forcing. Reciprocally ATP hydrolysis is largely dependent on the presence of substrate, independently, whether there is an uphill substrate gradient. In analogy to substrate induced transition fit in passive transport, also in external energy dependent transport, substrate induces a transition fit with the additional assistance of ATP hydrolysis. The total process of coupled ATP hydrolysis is divided up into several steps including ATP binding in situ, cleavage, “transition state”, dissociation of Pi and ADP, which accompany the steps of the transport cycle [32]. The frequently quoted “transition state” of ATP hydrolysis in ABC transporters taking place in the NBD domain should not be confounded with the transition state of transport in the transmembrane domain, as discussed here. Concomitant with the transfer of catalytic energy, some steps of the hydrolysis cycle stabilise the “in- or outside” state of the carrier depending whether the function is import or export, thus enabling the release of the substrate against a concentration gradient. It may be speculated that out of the 2 ATP required for transport of one substrate, one ATP is more focused on the facilitation of the transport, i.e., providing catalytic energy, and the second ATP is for skewing the transport site such as to stabilise the “outside” state for releasing the substrate. Thus the second ATP would be more involved in
providing energy for the “uphill” component of the substrate transport. It is not clear, however, to what extent MDR type carriers concentrate substrates, as some of their substrates are localised in the lipid bilayers and only flipped between the inner and outer membrane leaflets whereas others substrates are expelled from the membrane. Active import of substrates may also employ electrochemical gradients as an “external” energy supply for substrate uptake. In many of these “secondary active” transport systems the electrochemical gradients of H+ or Na+ drive uptake, mostly of sugars, amino acids, and neurotransmitters (Fig. 5B). Prime examples are the lactose H+ cotransporter of E. coli [33] and the renal Na+ glucose cotransporter [34]. For identifying of how the external energy is conveyed upon the carrier cycle, it is important to note that the cations serve to neutralise key negatively charged residues resident in the carrier, resulting in an electro-neutral substrate uptake branch whereas in the substrate less “return” branch “free” negative charges are exposed. Since in H+ or Na+ electrochemical gradients (ΔμH+ and ΔμNa+), in most cases the electrical component (nFΔψ) carries more weight than the osmotic gradient (nRT ln (H+ or Na+)e/(H+ or Na+)i, the major energy input is conveyed upon the rate limiting [35] return branch. By driving the negative charges of the cation binding site, electrical gradients accelerate the substrate less rate limiting step, receiving catalytic energy concomitant with the energy investment into the substrate gradient. This catalytic energy input distribution between the two translocation branches conforms to the postulates of the ITF, i.e.,
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
1233
large conformation changes, however, with low selectivity and consequently insufficient intrinsic binding energy necessary to overcome the large activation energy barrier. Energy from ATP has to make up for this major catalytic energy deficit. In general, utilisation of external energy for accumulating electrochemical gradient energy and for catalytic energy is intertwined and depends in a particular transport system on the external conditions of substrate and ion concentrations, as well as on the Δψ. In addition, in some systems accessory components for delivering substrate to the carrier and for extracting substrates from the carrier have to be taken into account for the overall energy balance. 4. Inhibitors
Fig. 4. Substrate size and catalytic energy in uni-directional transport. (A) Comparison of energy profiles of transport with carriers designed for large or for small substrates. The substrate induced energy barrier decrease (ΔGT) is stronger with larger substrate as a result of higher intrinsic binding energy. (B) Illustration of the different demands of small and large substrates on the binding site, on the width of the access channel and the contact surface.
the step lacking catalytic energy from substrate binding requires the largest share of the external energy supply. In conclusion, the standpoint that in active transport external energy only serves to sustain the electrochemical gradients cannot explain a variety of observations. In extreme case, external energy can be used primarily for catalytic energy, for instance, in MDR type transport systems. These have to deal with the dilemma of transporting large substrates, requiring
According to the ITF in the ground states of the carrier, substrate poorly fits to the bind center. Substrate analogues may exist exhibiting a better fit to the ground states. They would feature additional contacts and may fill the binding cavity more completely (Fig. 6). With the enhanced interactions, the binding energy increases pushing the inhibitor into an energy trap. Being unable to move from this low level into the transition state these compounds are not transported. With a much higher binding strength they block transported substrates and thus act as inhibitors of transport. Whereas for enzymes substrate analogues have to mimic the transition state of the substrate to qualify as inhibitors, in carriers substrate analogues are “ground state inhibitors” [4,8,9]. They could mirror the configuration of binding site in the ground state. Since in the two ground states the binding site assumes different configurations, whether being in- or outside oriented, there should exist two types of mutually excluding transport inhibitors docking from the in- or outside [21,36]. These inhibitors, exemplified by the highly specific, naturally occurring inhibitors of the AAC, carboxyatractylate for the outside and bongkrekate for the inside [21], may be larger than the substrates engaging a greater share of the residues lining the binding cavity thus producing higher binding energy. Some
Fig. 5. In active transport external energy transfer to the transition states for enhancing catalytic energy. (A) In ATP driven transport (ABC-transporter) energy is transferred to the two transition states in the two translocation branches, but primarily to the substrate less branch (CT). (B) In cation (H+ transport of Na+) co-transport the H+ or Na− electrochemical potential influences the two branches differently: the transition state of the substrate cation carrier complex is influenced by the cation concentration gradient, the “backward” branch with the “free” carrier now containing two uncompensated negative charges (C2− T ) is influenced by the membrane potential component which often is larger than the concentration gradient energy thus providing a larger supply to the substrate less return branch.
1234
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
Fig. 6. Illustration of inhibitor binding to the two ground states of the carrier. Inhibitors bind only to ground states. The two different interfaces in the external (Ce) and internal (Ci) ground states require different inhibitor structures. Enhanced interaction by inhibitor ligands may change ground state configuration to “abortive” ground states.
observations indicate that because of the high flexibility the binding center can be induced to assume within the “e” and “i” ground state a more open configuration to adapt to larger inhibitor molecules. We have called these “abortive ground states” [21], for emphasizing the structural and energetic positions opposite from the transition state (Fig. 7). 5. The AAC as an example of the ITF In many respects the AAC of mitochondria (reviews, [11, 21,37,38]) represents a tutorial case for illustrating the ITF mecha-
Fig. 7. Energy profile of inhibitor binding. The good fit of inhibitors to the binding site in the ground state fully releases binding energy leading to tight binding (low energy level). Larger inhibitors may induce “abortive” ground states (C′I′).
nism. Physiologically the ADP/ATP exchange is a secondary active transport, which under the Δψ of the inner membrane imports ADP and exports ATP each against a concentration gradient exploiting the charge difference of ADP3− against ATP4− [39]. It also efficiently catalyses electro-neutral “homo” exchange between ADP/ADP and ATP/ATP. These unproductive modes are important for the auto-regulation of the AAC by the ATP/ADP ratios on both sides of the membrane, by occupying to a varying degree AAC transport capacity. Exclusive homo-exchange can be studied in reconstituted vesicles [40]. Interestingly, the rates of homo and Δψ driven hetero-exchanges do not strongly differ, indicating a high efficiency of transport catalysis by the AAC. Although the large substrates ADP and ATP require large structural reconfiguration in the AAC during transport, the high activation energy is compensated by the strong interaction energy of the highly charged and polar molecules in line with the ITF. The nucleotide–protein interaction is especially strong as compared to most other ATP binding proteins since the fully charged ATP4− and ADP3− are bound instead of the less charged ATPMg2− and ATPMg1− complexes. Further, an unusually high selectivity for the adenine base indicates a precise and strong fit. Taken together, these binding elements create a high intrinsic binding energy necessary to compensate the high conformational energy associated with the translocation of large hydrophilic substrates. As a result, the seemingly paradoxical observation of low affinity of nucleotide binding and unusually high specificity are well explained by the ITF, as they show the balance between the two large energy amounts of the extensive conformation change and strong binding. A striking example for the influence of substrate size affords a comparison of the AAC with the phosphate carrier having similar structures. Both cooperate in oxidative phosphorylation such that
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
for each ATP delivered from the mitochondria one Pi has to be taken up. The molecular weight of the substrates differ by a factor of 5 and the turnover rates are about 5 to 10 times higher in the phosphate carrier than in the AAC [41] illustrating the correlation between substrate size and transport rate discussed above. As a result, to cope with their parallel metabolic role in mitochondria correspondingly different level of these carriers are present, the AAC attaining about 5 to 10 time higher levels than the Pi carrier. 6. Conclusion and outlook With the advent of crystal structures of biomembrane carriers, an advanced analysis of the intramolecular events during transport catalysis becomes realistic. These are guided by the thermodynamic and kinetic laws of carrier catalysis. Access channels and gates on both sides of the structural binding center have to be opened and closed requiring often large global conformation changes depending among other factors on the size of the substrate molecule. These reconfigurations involve tilting, turning, shifting of transmembrane helices and other structural sub-domains. To cope with the changes, often large energy inputs are required which on the first line are covered by the intrinsic binding energies in the pivotal transition state. More often, these will be insufficient and external energy must be called upon to supplement the insufficiency of catalytic energy. Both components of external energy supply, uphill transport and sustaining sufficient catalytic energy, are intertwined. Under these aspects carriers can be viewed to have evolved a mechanism of channelling both internal binding energy and external energy into catalytic rate acceleration to sustain cellular demands. For the future structure based analysis of residue — substrates interactions in carriers and their changes during the translocation steps, the induced transition fit with its implications should become a heuristic and interpretative guide. References [1] W.P. Jencks, in: A. Meister (Ed.), Binding Energy, Specificity, and Enzymatic Catalysis: The Circe Effect, in Advances in Enzymology, vol. 43, John Wiley & Sons, New York, 1975, pp. 219–410. [2] A.R. Fersht, Enzyme Structure and Mechanism, vol. 1, Freeman, San francisco, 1977. [3] W.P. Jencks, The utilization of binding energy in coupled vectorial processes, Adv. Enzymol. 51 (1980) 57–106. [4] M. Klingenberg, Catalytic Energy and Carrier-Catalyzed Solute Transport in Biomembranes, in: E. Quaglieriello, E.C. Slater, F. Palmieri, C. Saccone, A. Kroon (Eds.), Achievements and Perspectives of Mitochondrial Research. Vol. I: Bioenergetics, Elsevier, Amsterdam, 1985, pp. 303–315. [5] R.M. Krupka, Role of substrate binding forces in exchange-only transport systems: I. Transition-state theory, J. Membr. Biol. 109 (1989) 151–158. [6] R.M. Krupka, Role of substrate binding forces in exchange-only transport systems: II. Implications for the mechanism of the anion exchanger of red cells, J. Membr. Biol. 109 (1989) 159–171. [7] S.C. King, The “Transport Specificity Ratio”: a structure–function tool to search the protein fold for loci that control transition state stability in membrane transport catalysis, BMC Biochem. 5 (2004) 16. [8] M. Klingenberg, Mechanistic and energetic aspects of carrier catalysis — exemplified with mitochondrial translocators, in: S.A. Kuby (Ed.), A Study of Enzymes, Vol.II: Mechanism of Enzyme Action, CRC Press, Boca Raton, 1991, pp. 367–388. [9] M. Klingenberg, Ligand–protein interaction in biomembrane carriers. The induced transition fit of transport catalysis, Biochemistry 44 (2005) 8563–8570.
1235
[10] M. Klingenberg, The adenine nucleotide transport of mitochondria, in: L. Packer, A. Gomez-Puyou (Eds.), Mitochondria: Bioenergetics, Biogenesis and Membrane Structure, Academic Press, New York, 1976, pp. 127–149. [11] M. Klingenberg, The ADP/ATP shuttle of the mitochondrion, TIBS 4 (1979) 249–252. [12] M. Klingenberg, Membrane protein oligomeric structure and transport function, Nature 290 (1981) 449–454. [13] M. Klingenberg, The ADP/ATP carrier in mitochondrial membranes, in: A.N. Martonosi (Ed.), The Enzymes of Biological Membranes, Plenum Publishing Corp., New York, 1985, pp. 511–553. [14] J. Abramson, I. Smirnova, V. Kasho, G. Verner, S. Iwata, H.R. Kaback, The lactose permease of Escherichia coli: overall structure, the sugarbinding site and the alternating access model for transport, FEBS Lett. 555 (2003) 96–101. [15] Y. Huang, M.J. Lemieux, J. Song, M. Auer, D.N. Wang, Structure and mechanism of the glycerol-3-phosphate transporter from Escherichia coli, Science 301 (2003) 616–620. [16] E. Pebay-Peyroula, C. Dahout-Gonzalez, R. Kahn, V. Trezeguet, G.J. Lauquin, G. Brandolin, Structure of mitochondrial ADP/ATP carrier in complex with carboxyatractyloside, Nature 426 (2003) 39–44. [17] E.R. Kunji, M. Harding, Projection structure of the atractyloside-inhibited mitochondrial ADP/ATP carrier of Saccharomyces cerevisiae, J. Biol. Chem. 278 (2003) 36985–36988. [18] K.P. Locher, Structure and mechanism of ABC transporters, Curr. Opin. Struct. Biol. 14 (2004) 426–431. [19] S.G. Dahl, I. Sylte, A.W. Ravna, Structures and models of transporter proteins, J. Pharmacol. Exp. Ther. 309 (2004) 853–860. [20] A.L. Davidson, J. Chen, ATP-binding cassette transporters in bacteria, Annu. Rev. Biochem. 73 (2004) 241–268. [21] M. Klingenberg, The ADP/ATP carrier in mitochondrial membranes, in: A.N. Martonosi (Ed.), The Enzymes of Biological Membranes: Membrane Transport, vol. 3, Plenum Publishing Corp, New York, 1976, pp. 383–438. [22] R. Krämer, M. Klingenberg, Reconstitution of adenine nucleotide transport with purified ADP/ATP-carrier protein, FEBS Lett. 82 (1977) 363–367. [23] J. Abramson, I. Smirnova, V. Kasho, G. Verner, H.R. Kaback, S. Iwata, Structure and mechanism of the lactose permease of Escherichia coli, Science 301 (2003) 610–615. [24] M.J. Lemieux, Y. Huang, D.N. Wang, The structural basis of substrate translocation by the Escherichia coli glycerol-3-phosphate transporter: a member of the major facilitator superfamily, Curr. Opin. Struct. Biol. 14 (2004) 405–412. [25] S.J. Benkovic, S. Hammes-Schiffer, A perspective on enzyme catalysis, Science 301 (2003) 1196–1202. [26] P.G. LeFevre, G.F. McGinniss, Tracer exchange vs. net uptake of glucose through human red cell surface. New evidence for carrier-mediated diffusion, J. Gen. Physiol. 44 (1960) 87–103. [27] S.V. Ambudkar, C. Kimchi-Sarfaty, Z.E. Sauna, M.M. Gottesman, P-glycoprotein: from genomics to mechanism, Oncogene 22 (2003) 7468–7485. [28] G. Chang, C.B. Roth, Structure of MsbA from E. coli: a homolog of the multidrug resistance ATP binding cassette (ABC) transporters, Science 293 (2001) 1793–1800. [29] G. Chang, Structure of MsbA from Vibrio cholera: a multidrug resistance ABC transporter homolog in a closed conformation, J. Mol. Biol. 330 (2003) 419–430. [30] A.E. Senior, M.K. al-Shawi, I.L. Urbatsch, The catalytic cycle of P-glycoprotein, FEBS Lett. 377 (1995) 285–289. [31] A.L. Davidson, J. Chen, Structural biology. Flipping lipids: is the third time the charm? Science 308 (2005) 963–965. [32] M.K. Al-Shawi, M.K. Polar, H. Omote, R.A. Figler, Transition state analysis of the coupling of drug transport to ATP hydrolysis by P-glycoprotein, J. Biol. Chem. 278 (2003) 52629–52640. [33] H.R. Kaback, Structure and mechanism of the lactose permease, C. R. Biol. 328 (2005) 557–567. [34] E.M. Wright, E. Turk, The sodium/glucose cotransport family SLC5, Pflugers. Arch. 447 (2004) 510–518. [35] G.J. Kaczorowski, D.E. Robertson, H.R. Kaback, Mechanism of lactose translocation in membrane vesicles from Escherichia coli. 2. Effect of
1236
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
imposed delata psi, delta pH, and Delta mu H+, Biochemistry 18 (1979) 3697–3704. [36] J.E. Barnett, G.D. Holman, R.A. Chalkley, K.A. Munday, Evidence for two asymmetric conformational states in the human erythrocyte sugar-transport system, Biochem. J. 145 (1975) 417–429. [37] M. Klingenberg, H. Rottenberg, Relation between the gradient of the ATP/ ADP ratio and the membrane potential across the mitochondrial membrane, Eur. J. Biochem. 73 (1977) 125–130. [38] C. Fiore, V. Trezeguet, A. LeSaux, P. Roux, C. Schwimmer, A.C. Dianoux, F. Noel, G.J. Lauquin, G. Brandolin, P.V. Vignais, The mitochondrial
ADP/ATP carrier: structural, physiological and pathological aspects, Biochimie 80 (1998) 137–150. [39] M. Klingenberg, The ADP-ATP translocation in mitochondria, a membrane potential controlled transport, J. Membr. Biol. 56 (1980) 97–105. [40] R. Krämer, M. Klingenberg, Electrophorectic control of reconstituted adenine nucleotide translocation, Biochemistry 21 (1982) 1082–1089. [41] F. Palmieri, C. Indiveri, F. Bisaccia, R. Kramer, Functional properties of purified and reconstituted mitochondrial metabolite carriers, J. Bioenerg. Biomembr. 25 (1993) 525–535.
Transport catalysis Martin Klingenberg Institute Physiological Chemistry, University of Munich, Schillerstr.44, 80336 München, Germany Received 13 February 2006; received in revised form 6 April 2006; accepted 7 April 2006 Available online 19 April 2006
Abstract Carrier linked solute transport through biomembranes is analysed with the viewpoint of catalysis. Different from enzymes, in carriers the unchanged substrate induces optimum fit in the transition state. The enhanced intrinsic binding energy pays for the energy required of the global conformation changes, thus decreasing the activation energy barrier. This “induced transition fit” (ITF) explains several phenomena of carrier transport, e.g., high or low affinity substrate requirements for unidirectional versus exchange, external energy requirement for “low affinity” transport, the existence of side specific inhibitors to ground states of the carrier, the requirement of external energy in active transport to supplement catalytic energy in addition to generate electrochemical gradients. © 2006 Elsevier B.V. All rights reserved. Keywords: Transport; Catalytic energy; Induced fit; Active transport; Electrochemical gradient
Whereas the concept of catalysis is well conceived and widely exploited in enzymes [1,2], it is infrequently applied with its ramifications to transport [1,3–7]. For illustrating the crucial elements of carrier mediated catalysis a comparison to familiar enzymatic catalysis is useful [4,8,9]. Reaction theory requires that the catalyst, e.g., the enzyme, assist in transforming the substrate into a transition state. The binding site is constructed not to bind optimally the substrate but to fit the distorted substrate in the transition state. Thus by using energy generated from the optimum binding in the transition state, the substrate is readied for the subsequent and spontaneous chemical alteration. In contrast, carriers catalyse transport of substrates through membranes without chemical changes and therefore the concept of catalysis has received little attention in the transport field. In particular, it was not clear what a transition state as an essential ingredient of catalysis would mean and look like. Pertinent models of carriers were concerned with the global changes comprising the binding site reorientation between the “out” and “in”-side configurations. But the question was barely addressed, what happens in between these two “ground states”. The single binding center gated pore model (SBGP) has been the underlying mechanism for the present analysis of transport catalysis [10–13]. It provides a good example for simple single substrate enzyme catalysis. The SBGP was first established on E-mail address: [email protected]. 0005-2728/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbabio.2006.04.010
the molecular level with the AAC using specific inhibitor ligands. With two different, side specific inhibitors the carrier population could be entirely transferred from the out- to the in-side ground state and vice-versa. The presence of ADP or ATP was obligatory as catalysts of this transition coupled to the translocation of one substrate molecule. The SBGP is manifest in a wide range of transporters, encompassing unidirectional facilitators, exchange type transporters, and primary or secondary active transporters. The emergent structures of carriers of widely different provenance provide a striking support to the SBGP model and discard tandem and multiple binding site models [14–20]. Based on extensive observations in the mitochondrial membrane, in the isolated state, and in the reconstituted system [21,22], the intricacies of the transition between the “c” (out) and “m”(in) states of the mitochondrial ADP/ATP carrier (AAC), with often seemingly contradictory results, challenged us to comprehend the underlying mechanism of transport catalysis. In line with reaction theory and in analogy to enzymatic catalysis, we introduced the transition state, where the carrier–substrate complex assumes an intermediate configuration between the two ground states [4,8]. Here the substrate induces an optimum fit and thus generates a maximum intrinsic binding energy which drives the carrier substrate complex into and through the transition state between the ground states. With the advent of crystal structures of some carriers, molecular modelling of the transport mechanism and the forces
1230
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
involved become more precise and challenging. But in recent models based on 3D structures, intermediate steps between the in and out states were not considered or only marginally described [15,23,24]. Here we want to demonstrate that the concept of the induced transition fit (ITF) may provide an important guide to interpret the substrate–protein interactions involved in the translocation process. 1. Enzyme and carrier catalysis cycles A comparison of the simple catalytic cycles of enzyme and carriers illustrate similarities and differences between these two protein mediated types of catalysis (Fig. 1). Whereas after product release the enzyme is recuperated in a state ready for the next catalytic cycle, the carrier transformed after release of the substrate and the catalytic cycle is complete only after the second “return” branch. In balance, the carrier catalyses a vectorial reaction of the substrate, by facilitating the translocation through a diffusion barrier. In some way, the roles of the protein and the substrate are reversed: the substrate remains chemically unchanged but the protein has assumed a different configuration with opposite access paths to the binding center. One may regard the substrate rather than the carrier as the catalyst.
Fig. 1. Catalytic cycles for a simple enzyme reaction and for unidirectional (uniport) and exchanging (antiport) carrier transport. During the cycle, the carrier assumes two different ground states, with the binding site directed outside or inside. As a result, the full carrier cycle has two transformation branches. At an intermediate stage, the carrier enters the transition state.
Whereas in the analysis of enzyme mechanism, the transition state is well conceived, it is not obvious to introduce this pivotal step to transport. Indeed, the need for introducing the transition state becomes more rational only by considering the often important changes in the carrier structure during the translocation. Evidently, these global structural shifts require intermediate steps, some of them with the highest energy level qualifying as classical transition states. Most importantly, this catalysis is highly selective for the substrate structure, often surpassing that of enzymes dealing with the same substrate. In enzymes in the transition state, the interaction between a distorted substrate and an adapted binding site is optimised [25]. However, the conformation changes are limited and focussed on the catalytic centre. In carriers, the substrate is unchanged but the protein undergoes a global conformation change in the transition state, midway in the transformation between the inand out-states. How is the energy for these important changes generated? Similar as in enzymes, the substrate is not optimally bound in the ground states but, different from enzymes, the unchanged substrate induces a major reorganisation in the binding centre going into the transition state to attain an optimum fit (Fig. 2). In the concomitant global structure change, the gates are partially opened and closed respectively. This substrate induced optimum fit of the binding site leading to the transition state (induced transition fit, ITF) is the essence of the carrier catalysis. An energy reaction profile illustrates the implications of the ITF (Fig. 3). At first, substrate binding releases little energy because of poor fit to the ground state. On proceeding into the transition state, the increased substrate protein interaction forces produce a larger “intrinsic” binding energy, The energy costs of the concomitant reconfigurations of the carrier, shown by the high energy barrier in the substrate less transition state (CT), are paid for by the intrinsic binding energy such that in balance the energy barrier for the carrier substrate complex in the transition state is low. The energy difference of the carrier transition states without and with substrate corresponds to “catalytic energy” which is responsible for the substrate linked rate acceleration. In other words the catalytic energy necessary for transport is recruited from the substrate–protein interaction forces reaching a maximum in the transition state. Also in enzymes substrates can induce changes in the active center, which may even propagate to some other “background” residues along hydrogen and ionic bond relays [25]. But those changes are generally small and centred, whereas in substrate induced transition fit (ITF) of carriers the changes are large and global, focused not only on the binding center but also on the gating region. Relays of interacting residues emanate from the binding center towards the gating regions along the translocation path as illustrated in Fig. 2. Hereby, intermediate states may be partially stabilised as shown in the energy reaction profile of Fig. 3. Although extensive regions of the protein are modified, these changes follow a precise trajectory in a multidimensional energy space, coordinating all residues involved. Changes may involve translational and rotational movements, bending and tilting of transmembrane helices,
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
1231
Fig. 2. The induced transition fit (ITF) for the single binding center gated pore. Only first half of steps leading to the transition state are shown and not the analogous steps to the internal ground state. In the ground states, the substrate only loosely fits to the binding site. In the transition state, the substrate reaches optimum fit by inducing more binding contacts and thus global changes in the flexible structure which lead to partial opening and closing of the opposite gates.
triggered by the substrate induced residues shifts at the binding center. The intrinsic binding forces produced in the transition state compensate the resulting intra-carrier strains. The size of these changes may depend among other factors on the size of the substrate molecule that defines the width of the binding niche and the access paths (Fig. 4). Thus, with larger substrates the conformation changes on closing and opening the gates require more extensive and complex rearrangements, resulting in higher energy barriers, more intermediate steps, and lower transport rates. 2. Implications of the ITF By defining the energetic constraints for the working of carriers, the ITF determines a variety of transport parameters. According to the ITF, exchange type of transport is the most efficient transport mode by supplying binding energy to both branches of the transport cycle. In unidirectional transport the substrate less branch should encounter a markedly higher barrier and be rate limiting. In the classical case of the well studied glucose facilitator the turnover is strongly accelerated when substrate is provided to both sides of the membrane [26]. Severe constraints are imposed on unidirectional transport: the activation barrier of the free carrier must be comparatively low, in order to allow a sufficient transport rate which requires some flexibility of the carrier structure even without substrate. The conformation changes must be small and as a consequence only small and weakly binding substrates should be able to accommodate the unidirectional facilitated transport. Because of lower binding force the transport should be less selective and tolerate broader specificity.
3. Active transport As a result of these restrictions most unidirectional transport systems are found among transporters driven by external energy, either in primary or secondary active transport. According to common wisdom in active transport external energy is to drive substrates “uphill”, against an electrochemical gradient. However, there are cases where the transport is not obviously uphill and yet coupled to energy supply, apparently contradicting the common dogma. The best studied cases of “energy losing” systems are among the ABC transporter family, where the MDR (multi drug resistance) transporters with its paradigm Pglycoprotein (Pgp) (see review [27]) may exemplify the role of external energy supply in transport catalysis according to the roles of the ITF mechanism. Crystal structures of a bacterial MDR member, the lipid A flippase (MsbA), show an open crevice formed by two monomers in the dimer MsbA, apparently ready to accept large molecules from the cytoplasmic side of the membrane [28,29]. MDR transporters exhibit a broad specificity for their hydrophobic substrates, indicating that binding is comparatively weak. On the other hand, the large size of the substrates handled by the MDR transporters requires comparatively large conformation changes. As a result, there exists a significant discrepancy between the energy generated from binding and the catalytic energy required for the large structural change in opening and closing the necessarily wide channels. The coupling of transport to the hydrolysis of ATP [30] is suggested to compensate this energy deficit (Fig. 5A). It seems now well established that ATP hydrolysis at the nucleotide binding domains causes a drastic reconfiguration of this cytosolic orientated domain, (see reviews [19,20,31]) which is
1232
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
Fig. 3. Energy profile of carrier catalysis. Initial substrate binding releases low binding energy. In the transition state (intrinsic) binding energy is high due to the optimum fit. This intrinsic binding energy attenuates the high energy barrier of the free carrier (CT) to the low level of the transition state complex CTST. Thus, catalytic energy is recruited from the high intrinsic binding energy in the transitions state. The energy profile also illustrates intermediate steps to the transition state, caused by the mobile network of interacting residues that propagate the reconfiguration into the transition state (see Fig. 2).
transferred by interdomain contact to the transmembrane domain, thus facilitating the transformation of the transporting domain between the two ground states. In other words, catalytic energy is obtained from ATP hydrolysis via inter-domain forcing. Reciprocally ATP hydrolysis is largely dependent on the presence of substrate, independently, whether there is an uphill substrate gradient. In analogy to substrate induced transition fit in passive transport, also in external energy dependent transport, substrate induces a transition fit with the additional assistance of ATP hydrolysis. The total process of coupled ATP hydrolysis is divided up into several steps including ATP binding in situ, cleavage, “transition state”, dissociation of Pi and ADP, which accompany the steps of the transport cycle [32]. The frequently quoted “transition state” of ATP hydrolysis in ABC transporters taking place in the NBD domain should not be confounded with the transition state of transport in the transmembrane domain, as discussed here. Concomitant with the transfer of catalytic energy, some steps of the hydrolysis cycle stabilise the “in- or outside” state of the carrier depending whether the function is import or export, thus enabling the release of the substrate against a concentration gradient. It may be speculated that out of the 2 ATP required for transport of one substrate, one ATP is more focused on the facilitation of the transport, i.e., providing catalytic energy, and the second ATP is for skewing the transport site such as to stabilise the “outside” state for releasing the substrate. Thus the second ATP would be more involved in
providing energy for the “uphill” component of the substrate transport. It is not clear, however, to what extent MDR type carriers concentrate substrates, as some of their substrates are localised in the lipid bilayers and only flipped between the inner and outer membrane leaflets whereas others substrates are expelled from the membrane. Active import of substrates may also employ electrochemical gradients as an “external” energy supply for substrate uptake. In many of these “secondary active” transport systems the electrochemical gradients of H+ or Na+ drive uptake, mostly of sugars, amino acids, and neurotransmitters (Fig. 5B). Prime examples are the lactose H+ cotransporter of E. coli [33] and the renal Na+ glucose cotransporter [34]. For identifying of how the external energy is conveyed upon the carrier cycle, it is important to note that the cations serve to neutralise key negatively charged residues resident in the carrier, resulting in an electro-neutral substrate uptake branch whereas in the substrate less “return” branch “free” negative charges are exposed. Since in H+ or Na+ electrochemical gradients (ΔμH+ and ΔμNa+), in most cases the electrical component (nFΔψ) carries more weight than the osmotic gradient (nRT ln (H+ or Na+)e/(H+ or Na+)i, the major energy input is conveyed upon the rate limiting [35] return branch. By driving the negative charges of the cation binding site, electrical gradients accelerate the substrate less rate limiting step, receiving catalytic energy concomitant with the energy investment into the substrate gradient. This catalytic energy input distribution between the two translocation branches conforms to the postulates of the ITF, i.e.,
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
1233
large conformation changes, however, with low selectivity and consequently insufficient intrinsic binding energy necessary to overcome the large activation energy barrier. Energy from ATP has to make up for this major catalytic energy deficit. In general, utilisation of external energy for accumulating electrochemical gradient energy and for catalytic energy is intertwined and depends in a particular transport system on the external conditions of substrate and ion concentrations, as well as on the Δψ. In addition, in some systems accessory components for delivering substrate to the carrier and for extracting substrates from the carrier have to be taken into account for the overall energy balance. 4. Inhibitors
Fig. 4. Substrate size and catalytic energy in uni-directional transport. (A) Comparison of energy profiles of transport with carriers designed for large or for small substrates. The substrate induced energy barrier decrease (ΔGT) is stronger with larger substrate as a result of higher intrinsic binding energy. (B) Illustration of the different demands of small and large substrates on the binding site, on the width of the access channel and the contact surface.
the step lacking catalytic energy from substrate binding requires the largest share of the external energy supply. In conclusion, the standpoint that in active transport external energy only serves to sustain the electrochemical gradients cannot explain a variety of observations. In extreme case, external energy can be used primarily for catalytic energy, for instance, in MDR type transport systems. These have to deal with the dilemma of transporting large substrates, requiring
According to the ITF in the ground states of the carrier, substrate poorly fits to the bind center. Substrate analogues may exist exhibiting a better fit to the ground states. They would feature additional contacts and may fill the binding cavity more completely (Fig. 6). With the enhanced interactions, the binding energy increases pushing the inhibitor into an energy trap. Being unable to move from this low level into the transition state these compounds are not transported. With a much higher binding strength they block transported substrates and thus act as inhibitors of transport. Whereas for enzymes substrate analogues have to mimic the transition state of the substrate to qualify as inhibitors, in carriers substrate analogues are “ground state inhibitors” [4,8,9]. They could mirror the configuration of binding site in the ground state. Since in the two ground states the binding site assumes different configurations, whether being in- or outside oriented, there should exist two types of mutually excluding transport inhibitors docking from the in- or outside [21,36]. These inhibitors, exemplified by the highly specific, naturally occurring inhibitors of the AAC, carboxyatractylate for the outside and bongkrekate for the inside [21], may be larger than the substrates engaging a greater share of the residues lining the binding cavity thus producing higher binding energy. Some
Fig. 5. In active transport external energy transfer to the transition states for enhancing catalytic energy. (A) In ATP driven transport (ABC-transporter) energy is transferred to the two transition states in the two translocation branches, but primarily to the substrate less branch (CT). (B) In cation (H+ transport of Na+) co-transport the H+ or Na− electrochemical potential influences the two branches differently: the transition state of the substrate cation carrier complex is influenced by the cation concentration gradient, the “backward” branch with the “free” carrier now containing two uncompensated negative charges (C2− T ) is influenced by the membrane potential component which often is larger than the concentration gradient energy thus providing a larger supply to the substrate less return branch.
1234
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
Fig. 6. Illustration of inhibitor binding to the two ground states of the carrier. Inhibitors bind only to ground states. The two different interfaces in the external (Ce) and internal (Ci) ground states require different inhibitor structures. Enhanced interaction by inhibitor ligands may change ground state configuration to “abortive” ground states.
observations indicate that because of the high flexibility the binding center can be induced to assume within the “e” and “i” ground state a more open configuration to adapt to larger inhibitor molecules. We have called these “abortive ground states” [21], for emphasizing the structural and energetic positions opposite from the transition state (Fig. 7). 5. The AAC as an example of the ITF In many respects the AAC of mitochondria (reviews, [11, 21,37,38]) represents a tutorial case for illustrating the ITF mecha-
Fig. 7. Energy profile of inhibitor binding. The good fit of inhibitors to the binding site in the ground state fully releases binding energy leading to tight binding (low energy level). Larger inhibitors may induce “abortive” ground states (C′I′).
nism. Physiologically the ADP/ATP exchange is a secondary active transport, which under the Δψ of the inner membrane imports ADP and exports ATP each against a concentration gradient exploiting the charge difference of ADP3− against ATP4− [39]. It also efficiently catalyses electro-neutral “homo” exchange between ADP/ADP and ATP/ATP. These unproductive modes are important for the auto-regulation of the AAC by the ATP/ADP ratios on both sides of the membrane, by occupying to a varying degree AAC transport capacity. Exclusive homo-exchange can be studied in reconstituted vesicles [40]. Interestingly, the rates of homo and Δψ driven hetero-exchanges do not strongly differ, indicating a high efficiency of transport catalysis by the AAC. Although the large substrates ADP and ATP require large structural reconfiguration in the AAC during transport, the high activation energy is compensated by the strong interaction energy of the highly charged and polar molecules in line with the ITF. The nucleotide–protein interaction is especially strong as compared to most other ATP binding proteins since the fully charged ATP4− and ADP3− are bound instead of the less charged ATPMg2− and ATPMg1− complexes. Further, an unusually high selectivity for the adenine base indicates a precise and strong fit. Taken together, these binding elements create a high intrinsic binding energy necessary to compensate the high conformational energy associated with the translocation of large hydrophilic substrates. As a result, the seemingly paradoxical observation of low affinity of nucleotide binding and unusually high specificity are well explained by the ITF, as they show the balance between the two large energy amounts of the extensive conformation change and strong binding. A striking example for the influence of substrate size affords a comparison of the AAC with the phosphate carrier having similar structures. Both cooperate in oxidative phosphorylation such that
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
for each ATP delivered from the mitochondria one Pi has to be taken up. The molecular weight of the substrates differ by a factor of 5 and the turnover rates are about 5 to 10 times higher in the phosphate carrier than in the AAC [41] illustrating the correlation between substrate size and transport rate discussed above. As a result, to cope with their parallel metabolic role in mitochondria correspondingly different level of these carriers are present, the AAC attaining about 5 to 10 time higher levels than the Pi carrier. 6. Conclusion and outlook With the advent of crystal structures of biomembrane carriers, an advanced analysis of the intramolecular events during transport catalysis becomes realistic. These are guided by the thermodynamic and kinetic laws of carrier catalysis. Access channels and gates on both sides of the structural binding center have to be opened and closed requiring often large global conformation changes depending among other factors on the size of the substrate molecule. These reconfigurations involve tilting, turning, shifting of transmembrane helices and other structural sub-domains. To cope with the changes, often large energy inputs are required which on the first line are covered by the intrinsic binding energies in the pivotal transition state. More often, these will be insufficient and external energy must be called upon to supplement the insufficiency of catalytic energy. Both components of external energy supply, uphill transport and sustaining sufficient catalytic energy, are intertwined. Under these aspects carriers can be viewed to have evolved a mechanism of channelling both internal binding energy and external energy into catalytic rate acceleration to sustain cellular demands. For the future structure based analysis of residue — substrates interactions in carriers and their changes during the translocation steps, the induced transition fit with its implications should become a heuristic and interpretative guide. References [1] W.P. Jencks, in: A. Meister (Ed.), Binding Energy, Specificity, and Enzymatic Catalysis: The Circe Effect, in Advances in Enzymology, vol. 43, John Wiley & Sons, New York, 1975, pp. 219–410. [2] A.R. Fersht, Enzyme Structure and Mechanism, vol. 1, Freeman, San francisco, 1977. [3] W.P. Jencks, The utilization of binding energy in coupled vectorial processes, Adv. Enzymol. 51 (1980) 57–106. [4] M. Klingenberg, Catalytic Energy and Carrier-Catalyzed Solute Transport in Biomembranes, in: E. Quaglieriello, E.C. Slater, F. Palmieri, C. Saccone, A. Kroon (Eds.), Achievements and Perspectives of Mitochondrial Research. Vol. I: Bioenergetics, Elsevier, Amsterdam, 1985, pp. 303–315. [5] R.M. Krupka, Role of substrate binding forces in exchange-only transport systems: I. Transition-state theory, J. Membr. Biol. 109 (1989) 151–158. [6] R.M. Krupka, Role of substrate binding forces in exchange-only transport systems: II. Implications for the mechanism of the anion exchanger of red cells, J. Membr. Biol. 109 (1989) 159–171. [7] S.C. King, The “Transport Specificity Ratio”: a structure–function tool to search the protein fold for loci that control transition state stability in membrane transport catalysis, BMC Biochem. 5 (2004) 16. [8] M. Klingenberg, Mechanistic and energetic aspects of carrier catalysis — exemplified with mitochondrial translocators, in: S.A. Kuby (Ed.), A Study of Enzymes, Vol.II: Mechanism of Enzyme Action, CRC Press, Boca Raton, 1991, pp. 367–388. [9] M. Klingenberg, Ligand–protein interaction in biomembrane carriers. The induced transition fit of transport catalysis, Biochemistry 44 (2005) 8563–8570.
1235
[10] M. Klingenberg, The adenine nucleotide transport of mitochondria, in: L. Packer, A. Gomez-Puyou (Eds.), Mitochondria: Bioenergetics, Biogenesis and Membrane Structure, Academic Press, New York, 1976, pp. 127–149. [11] M. Klingenberg, The ADP/ATP shuttle of the mitochondrion, TIBS 4 (1979) 249–252. [12] M. Klingenberg, Membrane protein oligomeric structure and transport function, Nature 290 (1981) 449–454. [13] M. Klingenberg, The ADP/ATP carrier in mitochondrial membranes, in: A.N. Martonosi (Ed.), The Enzymes of Biological Membranes, Plenum Publishing Corp., New York, 1985, pp. 511–553. [14] J. Abramson, I. Smirnova, V. Kasho, G. Verner, S. Iwata, H.R. Kaback, The lactose permease of Escherichia coli: overall structure, the sugarbinding site and the alternating access model for transport, FEBS Lett. 555 (2003) 96–101. [15] Y. Huang, M.J. Lemieux, J. Song, M. Auer, D.N. Wang, Structure and mechanism of the glycerol-3-phosphate transporter from Escherichia coli, Science 301 (2003) 616–620. [16] E. Pebay-Peyroula, C. Dahout-Gonzalez, R. Kahn, V. Trezeguet, G.J. Lauquin, G. Brandolin, Structure of mitochondrial ADP/ATP carrier in complex with carboxyatractyloside, Nature 426 (2003) 39–44. [17] E.R. Kunji, M. Harding, Projection structure of the atractyloside-inhibited mitochondrial ADP/ATP carrier of Saccharomyces cerevisiae, J. Biol. Chem. 278 (2003) 36985–36988. [18] K.P. Locher, Structure and mechanism of ABC transporters, Curr. Opin. Struct. Biol. 14 (2004) 426–431. [19] S.G. Dahl, I. Sylte, A.W. Ravna, Structures and models of transporter proteins, J. Pharmacol. Exp. Ther. 309 (2004) 853–860. [20] A.L. Davidson, J. Chen, ATP-binding cassette transporters in bacteria, Annu. Rev. Biochem. 73 (2004) 241–268. [21] M. Klingenberg, The ADP/ATP carrier in mitochondrial membranes, in: A.N. Martonosi (Ed.), The Enzymes of Biological Membranes: Membrane Transport, vol. 3, Plenum Publishing Corp, New York, 1976, pp. 383–438. [22] R. Krämer, M. Klingenberg, Reconstitution of adenine nucleotide transport with purified ADP/ATP-carrier protein, FEBS Lett. 82 (1977) 363–367. [23] J. Abramson, I. Smirnova, V. Kasho, G. Verner, H.R. Kaback, S. Iwata, Structure and mechanism of the lactose permease of Escherichia coli, Science 301 (2003) 610–615. [24] M.J. Lemieux, Y. Huang, D.N. Wang, The structural basis of substrate translocation by the Escherichia coli glycerol-3-phosphate transporter: a member of the major facilitator superfamily, Curr. Opin. Struct. Biol. 14 (2004) 405–412. [25] S.J. Benkovic, S. Hammes-Schiffer, A perspective on enzyme catalysis, Science 301 (2003) 1196–1202. [26] P.G. LeFevre, G.F. McGinniss, Tracer exchange vs. net uptake of glucose through human red cell surface. New evidence for carrier-mediated diffusion, J. Gen. Physiol. 44 (1960) 87–103. [27] S.V. Ambudkar, C. Kimchi-Sarfaty, Z.E. Sauna, M.M. Gottesman, P-glycoprotein: from genomics to mechanism, Oncogene 22 (2003) 7468–7485. [28] G. Chang, C.B. Roth, Structure of MsbA from E. coli: a homolog of the multidrug resistance ATP binding cassette (ABC) transporters, Science 293 (2001) 1793–1800. [29] G. Chang, Structure of MsbA from Vibrio cholera: a multidrug resistance ABC transporter homolog in a closed conformation, J. Mol. Biol. 330 (2003) 419–430. [30] A.E. Senior, M.K. al-Shawi, I.L. Urbatsch, The catalytic cycle of P-glycoprotein, FEBS Lett. 377 (1995) 285–289. [31] A.L. Davidson, J. Chen, Structural biology. Flipping lipids: is the third time the charm? Science 308 (2005) 963–965. [32] M.K. Al-Shawi, M.K. Polar, H. Omote, R.A. Figler, Transition state analysis of the coupling of drug transport to ATP hydrolysis by P-glycoprotein, J. Biol. Chem. 278 (2003) 52629–52640. [33] H.R. Kaback, Structure and mechanism of the lactose permease, C. R. Biol. 328 (2005) 557–567. [34] E.M. Wright, E. Turk, The sodium/glucose cotransport family SLC5, Pflugers. Arch. 447 (2004) 510–518. [35] G.J. Kaczorowski, D.E. Robertson, H.R. Kaback, Mechanism of lactose translocation in membrane vesicles from Escherichia coli. 2. Effect of
1236
M. Klingenberg / Biochimica et Biophysica Acta 1757 (2006) 1229–1236
imposed delata psi, delta pH, and Delta mu H+, Biochemistry 18 (1979) 3697–3704. [36] J.E. Barnett, G.D. Holman, R.A. Chalkley, K.A. Munday, Evidence for two asymmetric conformational states in the human erythrocyte sugar-transport system, Biochem. J. 145 (1975) 417–429. [37] M. Klingenberg, H. Rottenberg, Relation between the gradient of the ATP/ ADP ratio and the membrane potential across the mitochondrial membrane, Eur. J. Biochem. 73 (1977) 125–130. [38] C. Fiore, V. Trezeguet, A. LeSaux, P. Roux, C. Schwimmer, A.C. Dianoux, F. Noel, G.J. Lauquin, G. Brandolin, P.V. Vignais, The mitochondrial
ADP/ATP carrier: structural, physiological and pathological aspects, Biochimie 80 (1998) 137–150. [39] M. Klingenberg, The ADP-ATP translocation in mitochondria, a membrane potential controlled transport, J. Membr. Biol. 56 (1980) 97–105. [40] R. Krämer, M. Klingenberg, Electrophorectic control of reconstituted adenine nucleotide translocation, Biochemistry 21 (1982) 1082–1089. [41] F. Palmieri, C. Indiveri, F. Bisaccia, R. Kramer, Functional properties of purified and reconstituted mitochondrial metabolite carriers, J. Bioenerg. Biomembr. 25 (1993) 525–535.