Kinetics of CO and NO ligation with the Cys(331)-->Ala mutant of neuronal nitric-oxide synthase.

Nitric-oxide synthases (NOS) catalyze the conversion of l-arginine to NO, which then stimulates many physiological processes. In the active form, each NOS is a dimer; each strand has both a heme-binding oxygenase domain and a reductase domain. In neuronal NOS (nNOS), there is a conserved cysteine motif (CX(4)C) that participates in a ZnS(4) center, which stabilizes the dimer interface and/or the flavoprotein-heme domain interface. Previously, the Cys(331) --> Ala mutant was produced, and it proved to be inactive in catalysis and to have structural defects that disrupt the binding of l-Arg and tetrahydrobiopterin (BH(4)). Because binding l-Arg and BH(4) to wild type nNOS profoundly affects CO binding with little effect on NO binding, ligand binding to the mutant was characterized as follows. 1) The mutant initially has behavior different from native protein but reminiscent of isolated heme domain subchains. 2) Adding l-Arg and BH(4) has little effect immediately but substantial effect after extended incubation. 3) Incubation for 12 h restores behavior similar but not quite identical to that of wild type nNOS. Such incubation was shown previously to restore most but not all catalytic activity. These kinetic studies substantiate the hypothesis that zinc content is related to a structural rather than a catalytic role in maintaining active nNOS.

Enzymatic synthesis of caged NADP cofactors: aqueous NADP photorelease and optical properties.

The synthesis of caged NADP analogues 18, 19, and 20 has been accomplished by utilizing the transglycosidase activity of solubilized NAD glycohydrolase (porcine brain) to incorporate caged nicotinamides 2, 3, and 4 into NADP. The synthesis of several nicotinamides modified at the carboxamide with o-nitrobenzyl photolabile groups is demonstrated as well as their potential for enzymatic transglycosidation. These results further demonstrate the feasibility of direct enzymatic transglycosidation of sterically hindered substrates into NAD(P), although high nicotinamide analogue water solubility was found to be a necessary trait for yield enhancement with certain analogues. Caged analogues were surveyed under aqueous conditions for net NADP photorelease, while the UV and fluorescent properties of both analogues and their photobyproducts were assessed for compatibility with systems that rely on optical monitoring of enzyme activity. A highly water-soluble alpha-methyl-o-nitrobenzyl group 8 was developed for the synthesis of 20 in order to enhance net NADP photorelease. Compound 20 demonstrated a high 75% net NADP photoreleased without substantial UV optical blackening or fluorescent byproducts. Analogues 18 and 19 were shown by ESI/MALDI-MS to photogenerate primarily adducts of NADP with deleterious UV and fluorescent properties. Our work stresses the superior release properties conferred by alpha-methyl substitution on aqueous carboxamide photorelease from o-nitrobenzyl compounds.

Activation of soluble guanylate cyclase by carbon monoxide and nitric oxide: a mechanistic model.

Soluble guanylate cyclase (GC) from bovine lung is activated 4-fold by carbon monoxide (CO) and 400-fold by nitric oxide (NO). Spectroscopic and kinetic data for ligation of CO and NO with GC are summarized and compared with similar data for myoglobin (Mb), hemoglobin (Hb), and heme model compounds. Kinetic, thermodynamic, and structural data form a basis on which to construct a model for the manner in which the two ligands affect protein structure near the heme for heme proteins in general and for GC in particular. The most significant datum is that although association rates of ligands with GC are similar to those with Mb and Hb, their dissociation rates are dramatically faster. This suggests a delicate balance between five- and six-coordinate heme iron in both NO and CO complexes. Based on these and other data, a model for GC activation is proposed: The first step is formation of a six-coordinate species concomitant with tertiary and quaternary structural changes in protein structure and about a 4-fold increase in enzyme activity. In the second step, applicable to NO, the bond from iron to the proximal histidine ruptures, leading to additional relaxation in the quaternary and tertiary structure and a further 100-fold increase in activity. This is the main event in activation, available to NO and possibly other activators or combinations of activators. It is proposed, finally, that the proximal base freed in step 2, or some other protein base suitably positioned as a result of structural changes following ligation, may provide a center for nucleophilic substitution catalyzing the reaction GTP --> cGMP. An example is provided for a similar reaction in a derivatized protoheme model compound. The reaction mechanism attempts to rationalize the relative enzymatic activities of GC, heme-deficient GC, GC-CO, and GC-NO on a common basis and makes predictions for new activators that may be discovered in the future.

Kinetics and equilibria of soluble guanylate cyclase ligation by CO: effect of YC-1.

Previous work has proved that the enzyme-soluble guanylate cyclase, GC, is activated several 100-fold by the combination of carbon monoxide plus a benzylindazole derivative called YC-1. That is about the same as activation by nitric oxide, which has a well-established role both in vivo and in vitro. This report addresses several spectroscopic, equilibrium, and kinetic effects wrought by YC-1 on carboxyl guanylate cyclase, including the following: a shift in the Soret absorption band by 4 nm to shorter wavelength; an increase in CO affinity by an order of magnitude; a dramatic change in the kinetics of CO association. After photolytic dissociation of CO, the majority, but not all, of bimolecular ligand recombination occurs with a time constant about 1000-fold faster than in the absence of YC-1, while a smaller fraction recombines almost, but not quite, the same as usual. This is reminiscent of the kinetics of NO association with GC, which also shows two prominent phases. The results just listed pertain in the presence of GTP/cGMP, which would be present during enzyme catalysis. Qualitatively similar, but smaller, effects occur in the absence of GTP/cGMP. Measurements are reported to characterize other changes in buffer conditions. The results are consistent with a mechanistic model that attributes a crucial role to the proximal bond that connects the heme iron to a histidine side chain in GC but also requires protein control of the distal environment.

Kinetics of NO ligation with nitric-oxide synthase by flash photolysis and stopped-flow spectrophotometry.

Nitric-oxide synthase (NOS) catalyzes conversion of L-arginine to nitric oxide, which subsequently stimulates a host of physiological processes. Prior work suggests that NOS is inhibited by NO, providing opportunities for autoregulation. This contribution reports that NO reacts rapidly (ka congruent with 2 x 10(7) M-1 s-1) with neuronal NOS in both its ferric and ferrous oxidation states. Association kinetics are almost unaffected by L-arginine or the cofactor tetrahydrobiopterin. There is no evidence for the distinct two phases previously reported for association kinetics of CO. Small amounts of geminate recombination of NO trapped in a protein pocket can be observed over nanoseconds, and a much larger amount is inferred to take place at picosecond time scales. Dissociation rates are also very fast from the ferric form, in the neighborhood of 50 s-1, when measured by extrapolating association rates to the zero NO concentration limit. Scavenging experiments give dissociation rate constants more than an order of magnitude slower: still quite fast. For the ferrous species, extrapolation is not distinguishable from zero, while scavenging experiments give a dissociation rate constant near 10(-4) s-1. Implications of these results for interactions near the heme binding site are discussed.

Soluble guanylate cyclase: effect of YC-1 on ligation kinetics with carbon monoxide.

Recently it has been reported that in the presence of YC-1, a benzyl indazole derivative, carbon monoxide activates soluble guanylate cyclase, GC, to about the same extent as its best known activator, nitric oxide. Kinetic studies utilizing flash photolysis of GC complexed with CO in the presence and absence of YC-1 show, in contrast to another recent report of a mixing experiment, that YC-1 has a profound effect on bimolecular association kinetics and a smaller, but significant, effect on ligand affinity. Most prominent is the appearance of a major, new phase in the bimolecular recombination kinetics in the presence of 200 microM YC-1: This major fraction rebinds CO approximately 1000-fold more rapidly than in the absence of YC-1. Another portion, considerably less than half, exhibits kinetics that are almost exactly the same as in the absence of YC-1. It is now clear that both YC-1 and CO have a strong synergistic effect on enzyme activity and also a dramatic effect on ligand binding behavior. It is, therefore, a reasonable inference that ligand binding at the heme iron atom is intimately connected with enzyme activation, a hypothesis that would have been difficult to maintain if the earlier report, that YC-1 has no effect on CO binding, were correct. Possible reasons for the discrepancy between the two measurements are suggested.

Arterial blood pressure responses to cell-free hemoglobin solutions and the reaction with nitric oxide.

Changes in mean arterial pressure were monitored in rats following 50% isovolemic exchange transfusion with solutions of chemically modified hemoglobins. Blood pressure responses fall into three categories: 1) an immediate and sustained increase, 2) an immediate yet transient increase, or 3) no significant change either during or subsequent to exchange transfusion. The reactivities of these hemoglobins with nitric monoxide (.NO) were measured to test the hypothesis that different blood pressure responses to these solutions result from differences in .NO scavenging reactions. All hemoglobins studied exhibited a value of 30 microM-1 s-1 for both .NO bimolecular association rate constants and the rate constants for .NO-induced oxidation in vitro. Only the .NO dissociation rate constants and, thus, the equilibrium dissociation constants varied. Values of equilibrium dissociation constants ranged from 2 to 14 pM and varied inversely with vasopressor response. Hemoglobin solutions that exhibited either transient or no significant increase in blood pressure showed tighter .NO binding affinities than hemoglobin solutions that exhibited sustained increases. These results suggest that blood pressure increases observed upon exchange transfusion with cell-free hemoglobin solutions can not be the result of .NO scavenging reactions at the heme, but rather must be due to alternative physiologic mechanisms.

Dissociation of nitric oxide from soluble guanylate cyclase.

Kinetic studies of soluble guanylate cyclase complexed with nitric oxide prove that NO dissociation in the presence of the substrate GTP and Mg2+ is as much as 50 times faster than in their absence. In the presence of those two reagents the dissociation rate constant is k(obs) = 0.04 +/- 0.01 s-1 at 20 degrees C, which is by far the fastest NO dissociation rate constant ever reported for a ferrous heme protein. Extrapolated to 37 degrees C, this corresponds to a half life of about 5 s for NO dissociation from soluble guanylate cyclase at physiological conditions, which is presumably fast enough to account for deactivation of the enzyme in biological systems. Dissociation rate constants are also reported for a variety of other reagent conditions.

Kinetics of nitric oxide dissociation from five- and six-coordinate nitrosyl hemes and heme proteins, including soluble guanylate cyclase.

Kinetics of NO dissociation were characterized for three five-coordinate systems, heme-NO, HSA-heme-NO (human serum albumin), GC-NO (soluble guanylate cyclase), and for the six-coordinate system, Im-heme-NO. Nitrosyl myoglobin was redetermined for comparison. Previously known, six-coordinate R and T state nitrosyl hemoglobins are also included in the comparison. The data indicate that NO dissociates more than 1000 times faster from five-coordinate model heme than it does from the six-coordinate analog. Such a negative trans-effect between NO and a proximal base is in sharp contrast to carboxy heme derivatives, in which ligand dissociation rates are greatly slowed in when a trans base is present. As a result of opposite trans-effects, six-coordinate carboxy and nitrosyl derivatives have comparable dissociation rates, even though the five-coordinate species are very different. In proteins, five- and six-coordinate forms do not show a large difference in dissociation rates. Part of the reason may be due to different probabilities for geminate recombination in the different proteins, but this cannot explain all the facts. There must also be influences of the protein structure on bond-breaking rate constants themselves. With the exception of hemoglobin in the T state, nitrosyl guanylate cyclase shows the highest NO dissociation rate constant, k(obs) = 6 x 10(-4) s(-1). This would yield a half-life of about 2 min at 37 degrees C for dissociation of NO from GC-NO, a number that has implications for the mechanism of regulation of the activity of this key heme enzyme.

Kinetics of CO ligation with nitric-oxide synthase by flash photolysis and stopped-flow spectrophotometry.

Interaction of CO with hemeproteins has physiological importance. This is especially true for nitric-oxide synthases (NOS), heme/flavoenzymes that produce .NO and citrulline from L-arginine (Arg) and are inhibited by CO in vitro. The kinetics of CO ligation with both neuronal NOS and its heme domain module were determined in the presence and absence of tetrahydrobiopterin and Arg to allow comparison with other hemeproteins. Geminate recombination in the nanosecond time domain is followed by bimolecular association in the millisecond time domain. Complex association kinetics imply considerable heterogeneity but can be approximated with two forms, one fast (2-3 x 10(6) M-1 s-1) and another slow (2-4 x 10(4) M-1 s-1). The relative proportions of the two forms vary with conditions. For the heme domain, fast forms dominate except in the presence of both tetrahydrobiopterin and Arg. In the holoenzyme, slow forms dominate except when both reagents are absent. Geminate recombination is substantial, approximately 50%, only when fast forms predominate. Stopped-flow mixing found dissociation constants near 0.3 s-1. These data imply an equilibrium constant such that very little CO should bind at physiological conditions unless large CO concentrations are present locally.

Evidence for a slow tertiary relaxation in the reaction of tert-butyl isocyanide with horseradish peroxidase.

The kinetics of tert-butyl isocyanide binding to the heme protein horseradish peroxidase (HRP) at 22 degrees C was examined on all time scales, from minutes to picoseconds, in aqueous borate buffer at pH 9.08. Unlike myoglobin (Mb) or hemoglobin, HRP shows two bimolecular ligand binding processes. For comparison, binding of the same ligand with Mb was measured under identical conditions. Ligand entry into the protein from the solvent in a mixing experiment is extremely slow in HRP: the bimolecular association constant is 0.04 M-1 s-1, while in Mb it is 4 x 10(3) M-1 s-1. Surprisingly, in view of that difference, picosecond and nanosecond photolyses reveal that once the ligand has reached the iron(II) site there is no difference in cage return or escape from the protein. The rate for the fastest cage return (from the contact pair) is close to 6 x 10(10) s-1 in both proteins. The rates of escape from the contact pair to form a secondary protein-caged pair are also similar: for Mb, 10 x 10(10) s-1, and for HRP, 8.5 x 10(10) s-1. The rate of rebinding from the protein-separated cage is near 4 x 10(6) s-1 in both proteins, and the rate of escape from protein to solvent is close to 3.7 x 10(6) s-1 in both. The difference between the two proteins lies in the low-millisecond time domain. After flash photolysis of HRP, there is a concentration-dependent recombination not seen in mixing experiments. This bimolecular rate constant varies slightly for different HRP preparations, being 2.6 x 10(4) or 4.0 x 10(4) M-1 s-1 in two cases, both of which are much faster than is observed in mixing experiments, namely, 0.04 M-1 s-1. In Mb, photolysis and mixing experiments consistently give the same combination rate, which is somewhat slower than the faster part of the HRP recombination. Similar measurements for the smaller ligand methyl isocyanide revealed no anomalous behavior. The interpretation proposed involves tertiary relaxation after ligand escape, which is significant in blocking the return of the large t-BuNC, but has no apparent effect on smaller ligands. Thus, HRP-t-BuNC reveals in dramatic fashion a phenomenon merely hinted at in earlier work involving the T-state binding kinetics of hemoglobin.

Basis of guanylate cyclase activation by carbon monoxide.

Kinetics of CO association with guanylate cyclase [GTP pyrophosphate-lyase (cyclizing), EC 4.6.1.2] and dissociation from carboxy guanylate cyclase have been studied at pH 7.5 by flash photolysis, yielding rate constants at 23 degrees C of 1.2 +/- 0.1 x 10(5) M-1.sec-1 and 28 +/- 2 sec-1, respectively. While the CO combination rate constant is the same as for the T state of hemoglobin, the CO dissociation rate constant is much higher than expected for a six-coordinate carboxyheme protein; yet the absorption spectrum is indicative of a six-coordinate heme. The two observations are reconciled by a reaction mechanism in which CO dissociation proceeds via a five-coordinate intermediate. This intermediate is structurally very similar to the five-coordinate nitrosyl heme derivative of guanylate cyclase and is presumably responsible for the observed 4-fold activation of guanylate cyclase by CO. Thus, we provide a model that explains enzyme activities of the nitrosyl and carboxy forms of the enzyme on the basis of a common mechanism.

Myoglobin-NO at low pH: free four-coordinated heme in the protein pocket.

In either sperm whale or horse heart myoglobin, binding of NO and lowering of solution pH work together to weaken, and ultimately break, the bond between iron and the proximal histidine. This is reminiscent of the reaction observed at neutral pH in the case of guanylate cyclase, the heme enzyme that catalyzes the conversion of GTP to cGMP. Bond breaking is characterized by a spectral change from a nine-line to a three-line ESR signal and accompanied by a shift from 420 to 387 nm in the UV-vis spectrum of the Soret band maximum. Analysis of the pH-dependent spectral changes shows that they are reversible, at least within a few hours, that the transition is cooperative, involving six protons during pH lowering but only two as it is raised, and that the pK is about 4.7. Different proteins exhibit different pK values, which are generally lower than that for "chelated" protoheme. The pK differences reflect the extra bond stability afforded by the protein structure. Investigations of thermal and photochemical NO displacement by CO suggest that the local pocket around the ligand, although significantly altered (according to circular dichroism investigations), nonetheless still imposes a barrier against the outward diffusion of ligand into the solvent. Nanosecond and picosecond flash photolysis shows that in proteins at low pH there is an extremely efficient geminate recombination of the ligand with the four-coordinated species through a single-exponential process. This occurs to a significantly larger extent than for the case of NO-"chelated" protoheme (where no distal barrier for ligand is present). At neutral pH, when the proximal histidine bond is intact, the geminate recombination for NO takes longer and displays multiexponential kinetics. Altogether, these results suggest that, even though distal effects probably also play a role, proximal effects make an important contribution in modulating ligand-iron bond formation.

Geminate recombination of diatomic ligands CO, O2, and NO with myoglobin.

The kinetics of geminate recombination of horse heart myoglobin with the diatomic ligands carbon monoxide, dioxygen, and nitric oxide have been reexamined. The new measurements are distinguished from previous studies by (1) consideration of the complete time range longer than 1 ps, (2) inclusion of the effect of temperature changes near ambient, (3) attention to the relation between recombination kinetics and the yield of dissociated partners on the millisecond time scale, and (4) use of singular value decomposition in the analysis. These picosecond results, together with earlier nanosecond data, for O2 prove that models incorporating one, two, or even three discrete intermediates are not sufficient to account for all features of geminate recombination kinetics. Instead, a continuous evolution of the geminate pair distribution is preferred.

Quaternary structure and geminate recombination in hemoglobin: flow-flash studies on alpha 2CO beta 2 and alpha 2 beta 2CO.

The kinetics of geminate recombination for the diliganded species alpha 2CO beta 2 and alpha 2 beta 2CO of human hemoglobin were studied using flash photolysis. The unstable diliganded species were generated just before photolysis by chemical reduction in a continuous flow reactor from the more stable valency hybrids alpha 2CO beta 2+ and alpha 2+ beta 2CO, which could be prepared by high pressure liquid chromatography. Before the flash photolysis studies, the hybrids had been characterized by double-mixing stopped-flow kinetics experiments. At pH 6.0 in the presence of inositol hexaphosphate (IHP) both of the diliganded species show second order kinetics for overall addition of a third CO that is clearly characteristic of the T state (l' = 1-2 x 10(5) M-1 s-1), whereas at higher pH and in the absence of IHP they show combination rates characteristic of an R state. The kinetics of geminate recombination following photolysis of a bound CO, however, showed little dependence on pH and IHP concentration. This surprising observation is explained on the basis that the kinetics of geminate recombination of CO primarily depends on the tertiary structure of the ligand binding site, which apparently does not differ much between the R state and the liganded T state formed on adding IHP in this system. Since this explanation requires distinguishing different tertiary structures within a particular quaternary structure, it amounts to a contradiction to the two-state allosteric model.

Analysis of hydride transfer and cofactor fluorescence decay in mutants of dihydrofolate reductase: possible evidence for participation of enzyme molecular motions in catalysis.

A remarkable correlation has been discovered between fluorescence lifetimes of bound NADPH and rates of hydride transfer among mutants of dihydrofolate reductase (DHFR) from Escherichia coli. Rates of hydride transfer from NADPH to dihydrofolate change by a factor of 1,000 for the series of mutant enzymes. Since binding constants for the initial complex between coenzyme and DHFR change by only a factor of 10, the major portion of the change in hydride transfer must be attributed to losses in transition-state stabilization. The time course of fluorescence decay for NADPH bound to DHFR is biphasic. Lifetimes ranging from 0.3 to 0.5 ns are attributed to a solvent-exposed dihydronicotinamide conformation of bound coenzyme which is presumably not active in catalysis, while decay times (tau 2) in the range of 1.3 to 2.3 ns are assigned to a more tightly bound species of NADPH in which dihydronicotinamide is sequestered from solvent. It is this slower component that is of interest. Ternary complexes with three different inhibitors, methotrexate, 5-deazafolate, and trimethoprim, were investigated, along with the holoenzyme complex; 3-acetylNADPH was also investigated. Fluorescence polarization decay, excitation polarization spectra, the temperature variation of fluorescence lifetimes, fluorescence amplitudes, and wavelength of absorbance maxima were measured. We suggest that dynamic quenching or internal conversion promotes decay of the excited state in NADPH-DHFR. When rates of hydride transfer are plotted against the fluorescence lifetime (tau 2) of tightly bound NADPH, an unusual correlation is observed. The fluorescence lifetime becomes longer as the rate of catalysis decreases for most mutants studied. However, the fluorescence lifetime is unchanged for those mutations that principally alter the binding of dihydrofolate while leaving most dihydronicotinamide interactions relatively undisturbed. The data are interpreted in terms of possible dynamic motions of a flexible loop region in DHFR which closes over both substrate and coenzyme binding sites. These motions could lead to faster rates of fluorescence decay in holoenzyme complexes and, when correlated over time, may be involved in other motions which give rise to enhanced rates of catalysis in DHFR.

Conformational dynamics of the formyl peptide receptor: a prototype for studies of receptor dynamics and binding pocket structure.

We have used spectrofluorometric and flow cytometric techniques to examine the interactions of formyl peptide ligands (L) with their cell surface receptors (R). Kinetic studies suggest that L binds to R at a diffusion limited rate and that R undergoes rapid transitions involving three states (LR, LRG, the ternary complex of L and R with the G protein, and a desensitized receptor "LRX" which forms within seconds) prior to internalization. A spectroscopic analysis of the interaction between L and R show that the binding pocket of R is large enough to contain no more than 6 amino acids and that a fluorescein-labelled pentapeptide is quenched upon binding to R. We hypothesize that histidine 90 (putatively located in the extracellular loop connecting the second and third transmembrane domains) protonates L and quenches the probe. New technology will extend the analysis of structure and dynamics to low affinity peptide receptors of living biological systems. Such technology will have implications in the design of peptidomimetic ligand and drug molecules.

Quaternary structure and the geminate recombination of carp hemoglobin with methylisocyanide.

The kinetics of geminate recombination were studied for the methylisocyanide derivative of carp hemoglobin. Carp hemoglobin is of interest because it has been established that the fully liganded form switches between a high affinity R state at pH 9 and a low affinity T state at pH 6 in the presence of IHP. Geminate recombination was observed on both the picosecond and the nanosecond time scales under all conditions; however, only a small variation is observed in the rates and the yields of geminate recombination as the protein switches from the R to the T state. Taken together with overall "on" and "off" rates, the data indicate that the change from the R to the T configuration affects bond breaking most, but also influences subsequent escape from the protein as well as both entry into the protein and bond formation. There is some reason to postulate tertiary conformational change in the T state on the microsecond time scale following ligand escape from the protein.

Picosecond kinetics of cytochromes b5 and c.

Ligand photolysis and subsequent recombination in cytochromes b5 and c have been studied with picosecond resolution. In both proteins, an iron-histidine bond is broken after excitation with 314-nm light, and recombination occurs with a rate constant of about 1.4 x 10(11) s-1. Photolysis and reformation of the iron-histidine bond may be surprising as these hemoproteins do not reversibly bind ligands in nature. The findings are explained using results both from experiments on model hemes and from computer investigations with atomic resolution on the three-dimensional structure of the protein. After photolysis, the formation and recombination of the geminate contact pair are attributed to simple low amplitude ligand bond rotations, a result that can be applied to geminate processes in other hemoproteins and model heme compounds as well.

Metal ion binding sites of bacteriorhodopsin. Laser-induced lanthanide luminescence study.

Laser-excited luminescence lifetimes of lanthanide ions bound to bacteriorhodopsin have been measured in deionized membranes. The luminescence titration curve, as well as the binding curve of apomembrane (retinal-free) with Eu3+, has shown that the removal of the retinal does not significantly affect the affinity of Eu3+ for the two high affinity sites of bacteriorhodopsin. The D2O effects on decay rate constants indicate that Eu3+ bound to the high affinity sites of native membrane or apomembrane is coordinated by about six ligands in the first coordination sphere. Tb3+ is shown to be coordinated by four ligands. The data indicate that metal ions bind to the protein with a specific geometry. From intermetal energy transfer experiments using Eu3+-Pr3+, Tb3+-Ho3+, and Tb3+-Er3+, the distance between the two high affinity sites is estimated to be 7-8 A.