Substrate Binding Primes Human Tryptophan 2,3-Dioxygenase for

Subscriber access provided by UNIV OF DURHAM

Article

Substrate Binding Primes Human Tryptophan 2,3-Dioxygenase for Ligand Binding Karin Nienhaus, Vincent Hahn, Manuel Hüpfel, and Gerd Ulrich Nienhaus J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b03463 • Publication Date (Web): 17 Jul 2017 Downloaded from http://pubs.acs.org on July 18, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Substrate Binding Primes Human Tryptophan 2,3-Dioxygenase for Ligand Binding Karin Nienhaus1, Vincent Hahn1, Manuel Hüpfel1, G. Ulrich Nienhaus1,2,3*

1

Institute of Applied Physics Karlsruhe Institute of Technology (KIT) Wolfgang-Gaede-Straße 1 76131 Karlsruhe, Germany 2

Institute of Nanotechnology (INT) and Institute of Toxicology and Genetics (ITG) Karlsruhe Institute of Technology (KIT) 76344 Eggenstein-Leopoldshafen, Germany

3

Department of Physics University of Illinois at Urbana-Champaign 1110 W. Green Street Urbana, Il 61801, USA

*Address correspondence to G. Ulrich Nienhaus, Institute of Applied Physics Karlsruhe Institute of Technology (KIT), 76133 Karlsruhe, Germany, Tel. +49 (0)721 608-43401 Fax. +49 (0)721 608-48480 Email: [email protected]

Running title: Catalytic complex formation in hTDO

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT The human heme enzyme tryptophan 2,3-dioxygenase (hTDO) catalyzes the insertion of dioxygen into its cognate substrate, L-tryptophan (L-Trp). Its active site structure is highly dynamic, and the mechanism of enzyme-substrate-ligand complex formation and the ensuing enzymatic reaction is not yet understood. Here we have studied complex formation in hTDO by using time-resolved optical and infrared spectroscopy with carbon monoxide (CO) as a ligand. We have observed that both substrate-free and substrate-bound hTDO coexist in two discrete conformations with greatly different ligand binding rates. In the fast rebinding hTDO conformation, there is facile ligand access to the heme iron, but it is greatly hindered in the slowly rebinding conformation. Spectroscopic evidence implicates active site solvation as playing a crucial role for the observed kinetic differences. Substrate binding shifts the conformational equilibrium markedly toward the fast species and thus primes the active site for subsequent ligand binding, ensuring that formation of the ternary complex occurs predominantly by first binding L-Trp and then the ligand. Consequently, the efficiency of catalysis is enhanced because O2 binding prior to substrate binding, resulting in nonproductive oxidation of the heme iron, is greatly suppressed.

2 ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION The heme enzymes tryptophan 2,3-dioxygenase (TDO) and indoleamine 2,3-dioxygenase (IDO) catalyze the first and rate-limiting step of the kynurenine pathway, i.e., the oxidative cleavage of the L-tryptophan (L-Trp) pyrrole ring to produce N-formylkynurenine.1 The monomeric IDO is ubiquitously expressed throughout the body, except for the liver. By contrast, the tetrameric TDO is predominantly found in the liver. Structurally, a description as a dimer of dimers is more appropriate because part of the substrate binding pocket of one subunit is formed by residues from another subunit. Although TDO and IDO were discovered decades ago, the molecular details of their catalytic activity are still controversially discussed.1-3 Recently, we presented a kinetic model that provides a quantitative description of ligand and substrate binding in human IDO1 (hIDO1).4 Our results showed that, for hIDO1 to be catalytically functional, ligand binding to the heme iron must precede substrate binding in the active site, as was also claimed by an earlier report.5 If the substrate binds first, which is more likely at high L-Trp concentrations, it suppresses ligand access to the heme iron and inhibits catalysis. Kolawole et al.6 subsequently reported that self-inhibition is incomplete at physiologically relevant substrate concentrations, possibly because ligands can still bypass the bulky substrate, although with low probability. For TDO, the reverse binding order has been proposed because O2-ligated ferrous TDO is unstable and decays swiftly to the ferric form.7-8 Based on the x-ray structure of ferrous TDO isolated from Xanthomonas campestris, xcTDO, it was suggested that the enzymatic action of TDO involves an induced fit mechanism.9 Upon recognition of L-Trp, an extensive network of interactions is formed that stabilizes the substrate in the active site (Figure 1a, b). Interestingly, the L-Trp ammonium ion is hydrogen-bonded to the side chain of Thr254, a residue in the αJ-αK loop (residues 248 ‒ 258) that is within a stretch of residues with small side chains (GTGGSSG) ensuring high flexibility. The αJ-αK loop is unstructured in the substrate-free enzyme and folds onto the substrate upon binding. Concomitantly, it sequesters the active site from the solvent. Recently, x-ray structures of substrate-free Drosophila melanogaster TDO (DmTDO10) and apo human TDO (hTDO11) have also been solved. These eukaryotic enzymes have an additional small domain adjacent to the active site of the neighboring subunit. It reduces structural heterogeneity of the αJ-αK loop and keeps it closer to the active site. DmTDO and 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

hTDO both contain the GTGGSSG stretch, and it was proposed that their αJ-αK loops also approach the active site upon substrate binding to shield it from the solvent.11 This conjecture was subsequently confirmed by the x-ray structure of the ternary hTDO-O2-L-Trp complex,12 which indeed features an active site structure with a closed αJ-αK loop (Figure 1C), similar to the one of the binary xcTDO complex (Figure 1A). The presently known x-ray structures suggest that the active site of TDO is highly dynamic and assembles into the proper environment competent of enzymatic action upon binding of the substrate L-Trp. Here we have used time-resolved optical absorption spectroscopy after photolysis to investigate the dynamic interplay between ligand and substrate binding and conformational dynamics in hTDO. Its heme group gives rise to an intense Soret absorption band at ~400 nm that is sensitive to the heme iron oxidation and ligation states. Here we use CO as a ligand instead of the physiological O2 because of its similar size and heme binding properties. Moreover, there is no enzyme activity with CO, so we can focus on assembly of the ternary complex. For CO-ligated ferrous hTDO and ligand-free (deoxy) hTDO, the Soret band is centered on 420 and 430 nm, respectively. These spectral changes allow us to measure the progress of ligand rebinding as a function of time. To reveal the effects of structural changes and protein dynamics on the kinetics of CO rebinding, we have varied solvent pH and composition, and the concentrations of L-Trp substrate and CO. In addition, we have used Fourier transform infrared (FTIR) spectroscopy to explore the active-site structure. The CO bond stretching vibration of the heme-bound CO gives rise to strong mid-infrared absorption bands that can be measured with exquisite sensitivity.13-14 The frequency of CO bound to the heme iron, νCO, is typically in the 1900 – 2000 cm‒1 spectral range and depends on the heme iron-ligand bond strength and the local electric field due to charges in the CO vicinity.15-16 The local electric field is created by charges in the CO vicinity. In heme proteins, we typically observe multiple stretching bands of the hemebound CO (so-called ‘A’ bands),17-19 indicating co-existence of multiple conformations with distinct active-site structures.20-23 We have also used FTIR in combination with temperature derivative spectroscopy (TDS),14,

24

a measurement protocol designed to study thermally

activated rate processes by taking spectra while the temperature is slowly raised over time.25-26 We have observed that hTDO exists as two discrete species with markedly different ligand binding properties. The active site environment of the slowly rebinding species is structurally 4 ACS Paragon Plus Environment

Page 4 of 30

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


very heterogeneous and ligand access to greatly hindered. Binding of the L-Trp substrate, however, stabilizes the fast-rebinding species, which features a well-structured active site and offers facile access of the ligand to the heme iron. This mechanism ensures that the ternary complex forms mainly by first binding the L-Trp substrate, which primes the active site for the subsequent binding of O2, which ensures efficient enzyme action.7-8

EXPERIMENTAL PROCEDURES Sample Preparation. Expression and purification of hTDO and hIDO1 followed published procedures.27,28 Site-directed mutagenesis was performed using the QuikChange II mutagenesis kit (Agilent Technologies, Waldbronn, Germany). Flash photolysis experiments were carried out at ambient temperature on dilute protein solutions held in airtight 1 × 1 × 3 cm3 glass cuvettes. For sample preparation, the solvent (100 mM sodium phosphate/citrate buffer (pH 4 ‒ 6), sodium phosphate buffer (pH 6.2 ‒ 8.6) and sodium carbonate buffer (pH > 8.6) without or with glycerol in specific volume proportions) was equilibrated with 1 bar CO or 0.05 bar CO/0.95 bar N2 in the airtight cuvette for 15 min to obtain a CO concentration of 1 mM or 50 µM, respectively. Subsequently, a few microliters of an anaerobically prepared sodium dithionite solution (1 M) at two-fold molar excess over the protein as well as a few microliters of an hTDO stock solution were added to obtain a solution containing ~10 μM of reduced protein. To determine the effects of bound substrate on the CO binding kinetics, a few microliters of a concentrated stock solution of L-Trp (equilibrated with CO, substrate concentration determined spectroscopically) were added. To achieve higher CO concentrations than 1 mM, the sample solution was transferred anaerobically into a home-built pressure cell for up to ~40 bar CO.4 For FTIR measurements, purified protein was dissolved at final a concentration of ~2 mM in cryosolvent (55% glycerol/45% 0.1 M potassium phosphate buffer, pH as indicated), stirred under a CO atmosphere for 1 h and reduced with a two-fold molar excess of an anaerobically prepared sodium dithionite solution. Subsequently, excess substrate was added (as a solid) and the mixture was stirred under CO for an additional 20 min. To remove undissolved material, the samples were centrifuged for 15 min at 5,000 rpm (EBA table top centrifuge, Hettich, Tuttlingen, Germany). 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

Flash photolysis. After photolysis by a 6-ns (full width at half maximum) pulse from a frequency doubled Nd:YAG laser (Surelite II-10, Continuum, Santa Clara, CA), ligand binding was monitored with light from a tungsten source that was passed through a monochromator set at 436 nm. The intensity of the light transmitted through the sample was measured by a photomultiplier tube (R5600U, Hamamatsu Corp., Middlesex, NJ). Its photocurrent output was amplified and digitized by two analog-to-digital converter cards NI-PCI 5114 and NI-PCI 6221 (National Instruments, Munich, Germany) in the time ranges 4 ns ‒ 3 ms and >4 µs, respectively. Up to 200 individual traces were averaged for the final rebinding trace. In the experiment, we use the absorption change at 436 nm due to ligand photodissociation, ΔA(t), as a measure of the fraction of proteins that are without a bound ligand at time t after the photolyzing pulse, =

∆ ∆0

,

(1)

with ΔA0 denoting the absorbance change upon complete photolysis. Because kinetic traces extend over many orders of magnitude in time, we plot them typically on a logarithmic time scale. In heme proteins, such traces typically display two distinct processes, internal, or geminate rebinding of ligands that did not escape from the protein after photodissociation, and bimolecular rebinding of ligands from the solvent. In the geminate process, the very same ligand that was photolyzed rebinds to the heme iron, and the corresponding rate coefficient does not depend on the ligand concentration in the solvent. Rebinding from the solvent, however, is a bimolecular process; its rate depends on both the concentrations of ligand, [L], and protein, [P] −

= .

(2)

If ligand binding studies are carried out with a large excess of ligand, [L] >> [P], [L] stays practically constant over the course of the reaction, so the kinetics simplify to pseudo-first order, with rate coefficient, = ,

(3)

which scales linearly with the ligand concentration.

FTIR Spectroscopy at Cryogenic Temperatures. Sample solutions were sandwiched between two CaF2 windows separated by a 75-µm Mylar spacer and enclosed by a sample holder made from oxygen-free high-conductivity copper. The holder was mounted onto a cold 6 ACS Paragon Plus Environment

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


finger of a closed-cycle helium refrigerator (model PT-SHI-4-5-331S, Sumitomo, Tokyo, Japan). A silicon temperature sensor diode and a digital temperature controller (model 330, Lake Shore Cryotronics, Westerville, OH) enabled precise temperature regulation of the sample. Samples were photolyzed with a continuous-wave, frequency-doubled Nd-YAG laser (model Cobolt Samba 0532-04-01-0300-300, Cobolt AB, Solna, Sweden) emitting 300 mW output power at 532 nm. The laser beam was split and focused onto the sample from both sides. FTIR transmission spectra were collected between 1,700 – 2,300 cm–1 with a resolution of 2 cm–1 with a Vertex 80v FTIR spectrometer (Bruker, Karlsruhe, Germany) using an InSb detector. Photolysis difference spectra were calculated from transmission spectra taken at 4 K before and after photolysis, ΔA = log(Ilight/Idark). Absorption spectra were calculated from transmission spectra of CO-ligated and CO-free samples. Samples were photolyzed either by 10-s illumination at 4 K or by slowly cooling the sample from 160 to 4 K under continuous illumination. The first protocol predominantly populates transient ligand docking sites close to the heme; the slow-cool protocol also probes alternative docking sites.29

Temperature-derivative Spectroscopy (TDS). At cryogenic temperatures, CO ligands that are photodissociated from the heme iron cannot escape from the protein but may dock at welldefined sites within the protein.23,

30

Thermally activated crossing of enthalpy barriers

against rebinding to the heme can be conveniently study with the TDS method.14, 24 From our TDS experiments on a wide variety of myoglobin mutants and hemoglobins, we anticipate that the CO ligands stay preferentially near the active site upon photolysis at 4 K for a few seconds, whereas they may become dispersed among other docking sites within the protein upon photolysis at higher temperatures and for longer times.31-34 FTIR/TDS data were acquired using the following protocol. After photodissociation as described above, the sample temperature, T, was raised from 4 K at a rate of 0.3 K/min, while FTIR transmittance spectra, I(v, T), were taken every 1 K. The temperature ramp protocol ensures that rebinding occurs sequentially with respect to the temperature at which processes become activated on the characteristic time scale of the experiment (~100 s). Ligand rebinding gives rise to absorbance changes from one spectrum to the next, which are usually displayed as two-dimensional contour plots versus temperature and 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

wavenumber, with contours spaced logarithmically to enhance weak features. Alternatively, we plot integrated intensities, i.e., the band areas missing from the A substates at a particular temperature due to photolysis at 4 K, to survey the overall recombination behavior of the different samples.

RESULTS AND DISCUSSION Flash Photolysis at Ambient Temperature. Substrate-free hTDO-CO dissolved in buffer solution (100 mM potassium phosphate buffer, pH 7.4, 1 mM CO) shows two sequential rebinding processes following photodissociation by a nanosecond laser flash (Figure 2A), with a fast step extending from micro- to milliseconds followed by a slower step on the millisecond scale. Increasing [L-Trp] up to ~1 mM systematically changed the kinetics in two characteristic ways: (i) The amplitude at the earliest time (1 µs) decreased; this variation can be quantified by an equilibrium binding curve with a dissociation coefficient, Kd(CO) = 10 ± 1 µM (Supporting Figure S1), indicating that L-Trp binds to hTDO-CO and interferes with CO escape from the active site after photodissociation. (ii) The slow kinetic step gradually disappears until only the fast step remains above ~100 µM. Larger structural rearrangements are strongly coupled to the viscosity of the surrounding solvent.35 Therefore, we performed flash photolysis experiments on hTDO-CO dissolved in a viscous solvent (75% glycerol (by volume) in potassium phosphate buffer, pH 7.5, 1 mM CO). The kinetics appear quite different but are still biphasic (Figure 2B). As in buffer, the amplitude at early times decreased with increasing [L-Trp], although to a lesser extent. This change can also be described by an equilibrium binding curve with a dissociation coefficient, Kd(CO) = 35 ± 8 µM (Supporting Figure S1). Apparently, L-Trp binds less tightly to hTDO-CO in glycerol buffer, as expected from the different protein-solvent partition coefficients of L-Trp in the two solvents.4 Again, as in buffer, the slow step gradually decreased with increasing [LTrp]; however, it did not completely disappear even at the highest L-Trp concentrations. In solutions containing between 0 and 75% glycerol (by volume) and 1 mM L-Trp, there is a continuous slowing of the second step with increasing glycerol content (Supporting Figure S2). Figure 2 shows a strong, systematic effect of L-Trp substrate binding on CO recombination; however, the nature of the two sequential processes remains obscure. Because geminate 8 ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


rebinding in heme proteins typically takes

Substrate Binding Primes Human Tryptophan 2,3-Dioxygenase for

Recommend Documents