Could tyrosine and tryptophan serve multiple roles in biological redox processes?

Abstract

Single-step electron tunnelling reactions can transport charges over distances of 15–20 Åin proteins. Longer-range transfer requires multi-step tunnelling processes along redox chains, often referred to as hopping. Long-range hopping via oxidized radicals of tryptophan and tyrosine, which has been identified in several natural enzymes, has been demonstrated in artificial constructs of the blue copper protein azurin. Tryptophan and tyrosine serve as hopping way stations in high-potential charge transport processes. It may be no coincidence that these two residues occur with greater-than-average frequency in O₂- and H₂O₂-reactive enzymes. We suggest that appropriately placed tyrosine and/or tryptophan residues prevent damage from high-potential reactive intermediates by reduction followed by transfer of the oxidizing equivalent to less harmful sites or out of the protein altogether.

1. Background

It is well established that multi-step tunnelling, called hopping, is required for functional charge transport in many redox enzymes (examples include ribonucleotide reductase [1–9], photosystem II [10–12], DNA photolyase [13–21], MauG [22–25] and cytochrome c peroxidase [26,27]). Here, we advance the hypothesis that many such enzymes, most especially those that generate high-potential intermediates during turnover, could be irreversibly damaged if the intermediates are not inactivated in some way. We suggest that appropriately placed tyrosine (Tyr) and/or tryptophan (Trp) residues can prevent such damage by rapid reduction of the intermediates followed by transfer of the oxidizing equivalent to less harmful sites or out of the protein altogether [28]. A protective role of this sort will not be apparent in catalytic rates and binding constants; the enzyme survival time is more likely to be an indicator of this function.

The blue copper protein Pseudomonas aeruginosa azurin has been a test-bed for mechanistic investigations of Trp and Tyr radical formation in biological electron transfer (ET) reactions [29–33]. Our initial investigations revealed that Cu^I oxidation by a photoexcited Re^I-diimine (Re^I(CO)₃(dmp), dmp=4,7-dimethyl-1,10-phenanthroline) covalently bound at His¹²⁴ on a His¹²⁴Gly¹²³Trp¹²²Met¹²¹ β-strand (ReHis¹²⁴Trp¹²²Cu^I-azurin) occurs in a few nanoseconds, fully two orders of magnitude faster than documented for single-step electron tunnelling at a 19 Ådonor–acceptor distance [29]. We attributed the accelerated ET to a two-step hopping mechanism involving a Trp^•+ radical intermediate. In recent work, we examined ET in ReHis¹²⁶Trp¹²²Cu^I-azurin, which has three redox sites at well-defined distances in the protein fold (Re–Trp¹²²(indole)=13.1 Å, dmp–Trp¹²²(indole)=10.0 Å, Re–Cu=25.6 Å) [32]. Near-UV excitation of the Re chromophore leads to prompt Cu^I oxidation (less than 50 ns), followed by slow back-ET (more than 0.2 μs) to regenerate Cu^I and ground-state Re^I. Spectroscopic measurements performed with varying protein concentrations suggest that the photoinduced ET reactions occur in protein dimers, (ReHis¹²⁶Trp¹²²Cu^I)₂, and that forward ET is accelerated by intermolecular electron hopping through the interfacial tryptophan: (∥ denotes a protein–protein interface). Solution mass spectrometry confirms a broad oligomer distribution with prevalent monomers and dimers, and the crystal structure of the Cu^II form reveals two ReHis¹²⁶Trp¹²²Cu^II molecules oriented such that redox cofactors Re(dmp) and Trp¹²²-indole on different protein molecules are located in the interface at much shorter intermolecular distances (Re–Trp¹²²(indole)=6.9 Å, dmp–Trp¹²²(indole)=3.5 Åand Re–Cu=14.0 Å) than within single protein folds. Whereas forward ET is accelerated by hopping through Trp¹²², Re^I(dmp^•−)∥→Cu^II ET is sluggish, probably due to poor electronic coupling across the protein–protein interface as well as the absence of an energetically accessible radical intermediate. These findings provided new insights into the factors that regulate Trp^•+ radical formation in protein ET reactions involving high-potential oxidants.

Work by theorists has shed much light on the main factors controlling biological ET reactions [34,35]: and, in our experimental programme, we have found semiclassical ET theory [36] to be particularly useful in analyses of results. Notably, given a particular spatial arrangement of redox cofactors, we can predict driving force dependences of the relative time constants for single-step (τ_ss=1/k_ss) and multi-step (τ_hop) electron transport [31]. Alternatively, given the redox and reorganization energetics, we can predict the hopping propensity for different cofactor arrangements [33]. We considered azurins labelled with Ru(bpy)₂(im)(His^X)²⁺ (bpy=2,2^′-bipyridine; im=imidazole; His^X=surface histidine) labelled at three surface sites (RuHis¹⁰⁷, RuHis¹²⁴ and RuHis¹²⁶) and examined the hopping advantage (τ_ss/τ_hop) for a protein with a generalized intermediate (Int) situated between a diimine-Ru^III oxidant and Cu^I [33]. In all cases, the greatest hopping advantage occurs in systems where the Int–Ru^III distance (r₁) is up to 5 Å shorter than the Int–Cu^I distance (r₂). The hopping advantage increases as systems orient nearer a linear donor–Int–acceptor configuration, owing to minimized intermediate tunnelling distances. The smallest predicted hopping advantage is in RuHis¹²⁴-azurin, which has the shortest Ru–Cu distance of the three proteins. The hopping advantage is nearly lost as ΔG^° for the first step () rises above +0.15 eV. Isoergic initial steps provide a wide distribution of arrangements, where advantages as great as 10⁴ are possible (for a fixed donor–acceptor distance of 23.7 or 25.4 Å). A slightly exergonic Int→Ru^III step provides an even larger distribution of arrangements for productive hopping, which will be the case as long as the driving force for the first step is not more favourable than that for overall transfer.

We tested these predictions experimentally in three Ru–His-labelled azurins using nitrotyrosinate (NO₂TyrO⁻) as a redox intermediate (RuHis¹⁰⁷(NO₂TyrOH)¹⁰⁹; RuHis¹²⁴(NO₂TyrOH)¹²²; and RuHis¹²⁶(NO₂TyrOH)¹²²; E^°′(NO₂TyrO^•/−)≈1.02 V versus NHE) [33]. The first two systems have cofactor placements that are close to the predicted optimum; the last system has a larger first-step distance, which is predicted to decrease the hopping advantage. The phenol pK_a of 3-nitrotyrosine (7.2) permitted us to work at near-neutral pH, rather than the high pH (more than 10) required for hopping with tyrosinate. ET via nitrotyrosinate avoids the complexities associated with the proton-coupled redox reactions of tyrosine [37–39]. We found specific rates of Cu^I oxidation more than 10 times greater than those of single-step ET in the corresponding azurins lacking NO₂TyrOH, confirming that NO₂TyrO⁻ accelerates long-range ET. In this case, the proposed reaction sequence is [, although the nitrotyrosyl radical intermediate was not detected by transient spectroscopy in any of the proteins investigated. The results are in excellent agreement with hopping maps developed using semiclassical ET theory and parameters derived from our body of protein ET measurements [31–42].

2. Protecting P450s?

The cytochromes P450 are members of a superfamily of haem oxygenases that perform two broad functional roles: xenobiotic metabolism and biosynthesis [43,44]. In mammals, these functions include drug metabolism, conversion of lipophilic molecules to more polar products for enhanced elimination, steroid biosynthesis and conversion of polyunsaturated fatty acids to biologically active products [43,45]. P450s are monooxygenases that incorporate one oxygen atom from O₂ into the substrate while the second is reduced to H₂O [43,44]. Substrate binding triggers reduction of the ferric enzyme to the ferrous state by a reductase with reducing equivalents originating in NAD(P)H. Oxygen binding, followed by delivery of a second electron, induces O−O bond cleavage, producing H₂O and a ferryl-porphyrin cation radical (Cmpd-1). Cmpd-1 transfers an O atom to substrate, regenerating the ferric haem [43]. The UniProtKB/Swiss-Prot database indicates that P450s are rich in aromatic amino acids: 90% of the sequences in the P450 family (956 sequences) have above-average occurrence of Phe residues; and 68% of the sequences have more Trp residues than average. Phe residues are unlikely to participate as real intermediates in ET reactions, but their role in enhanced superexchange coupling for long-range ET remains unresolved [46–48]. Cytochrome P450 BM3 (CYP102A1) is frequently used as a soluble surrogate for human microsomal P450 s [44]. A Trp⁹⁶ residue is found within 5 Åof the haem in the structure of CYP102A1 [49] (figure 1), and sequence alignment (UniProtKB) in the P450 family suggests that Trp is conserved at this position in more than 75% of the members of this group. The most common alternative residue at this position is His. Interestingly, of the 698 sequences with Trp at this position, all but five derive from eukaryotic sources, whereas about half of the proteins with His at this position derive from bacterial or archaeal sources. The strong conservation of the Trp⁹⁶ residue has been noted previously [50]. An ET mediator role for this residue had been suggested, but replacement of Trp⁹⁶ with Ala, Phe or Tyr has little impact on catalytic turnover kinetics [50]. To the best of our knowledge, no role other than structural has been reported for this highly conserved Trp residue in P450 [44].

Figure 1.

(a) Space-filling structural model of the haem domain of CYP102A1 (PDB #2IJ2) highlighting the surface locations of terminal residues in two potential radical transfer pathways (Tyr³³⁴ and Tyr³⁰⁵). (b) Space-filling model of the residues comprising CYP102A1 putative radical transfer pathways. Blue spheres represent structurally resolved water molecules. (Online version in colour.)

The oxygenation chemistry catalysed by some P450s is tightly coupled to substrate hydroxylation: one mole of product is produced for each mole of O₂ consumed. In many enzymes, particularly the eukaryotic proteins with broad substrate specificities, hydroxylation is much less efficiently coupled to O₂ reduction (frequently less than 10%) [51–53]. When the enzyme does not transfer an O atom to substrate, it can produce reactive oxygen species (ROS: , H₂O₂) or a second H₂O molecule [54]. The production of ROS can lead to rapid degradation of the enzyme. In the case of oxidase chemistry (formation of 2H₂O from O₂), two reducing equivalents must be delivered by sources other than the substrate. Uncoupled P450 catalysis leads to oxidation of bilirubin and uroporphyrinogen [55,56]. The physiological consequences of this uncoupling could involve uroporphyria and lowered plasma bilirubin concentrations [56]. Halogenated P450 substrates, including environmental toxins such as polyhalogenated biphenyls, are associated with inhibition of monooxygenase activity and uncoupled O₂ consumption [55,56]. We suggest that redox-active Tyr and/or Trp residues in P450, quite possibly analogues of Trp⁹⁶, protect the enzyme from oxidative degradation during uncoupled catalysis.

3. Parting shots

Across the six enzyme classes (oxidoreductases, transferases, hydrolases, lyases, isomerases and ligases) defined by the Enzyme Data Bank of the Swiss Institute of Bioinformatics, greater than half of the oxidoreductases have Tyr and Trp frequencies above the database average (figure 2). For the remaining five classes, most proteins have Trp frequencies below the database average, and only among the ligases does a majority of proteins have Tyr frequencies substantially above the database average. Refinement of this analysis reveals that, of the O₂-reactive oxidoreductases, 66% have Tyr frequencies above the database average and 81% have above-average Trp frequencies. Could the direct participation of Tyr and Trp residues in enzymatic redox chemistry be far more common than currently recognized? We think so!

Figure 2.

Occurrence frequencies relative to the UniProtKB/Swiss-Prot database average of aromatic residues (Phe, Trp and Tyr) in different enzyme classes. (Online version in colour.)

Funding statement

Our work on biological electron transfer processes is supported by the National Institutes of Health (DK-019038) and the Arnold and Mabel Beckman Foundation.