Kinetic and Spectroscopic Characterization of the Catalytic Ternary Complex of Tryptophan 2,3-Dioxygenase

Abstract

The first step of the kynurenine pathway for l-tryptophan (l-Trp) degradation is catalyzed by heme-dependent dioxygenases, tryptophan 2,3-dioxygenase (TDO) and indoleamine 2,3-dioxygenase. In this work, we employed stopped-flow optical absorption spectroscopy to study the kinetic behavior of the Michaelis complex of Cupriavidus metallidurans TDO (cmTDO) to improve our understanding of oxygen activation and initial oxidation of l-Trp. On the basis of the stopped-flow results, rapid freeze-quench (RFQ) experiments were performed to capture and characterize this intermediate by Mössbauer spectroscopy. By incorporating the chlorite dismutase-chlorite system to produce high concentrations of solubilized O₂, we were able to capture the Michaelis complex of cmTDO in a nearly quantitative yield. The RFQ-Mössbauer results confirmed the identity of the Michaelis complex as an O₂-bound ferrous species. They revealed remarkable similarities between the electronic properties of the Michaelis complex and those of the O₂ adduct of myoglobin. We also found that the decay of this reactive intermediate is the rate-limiting step of the catalytic reaction. An inverse α-secondary substrate kinetic isotope effect was observed with a k_H/k_D of 0.87 ± 0.03 when (indole-d₅)-l-Trp was employed as the substrate. This work provides an important piece of spectroscopic evidence of the chemical identity of the Michaelis complex of bacterial TDO.

Figure 1.

Time-resolved (0−60 s) UV−vis spectra of ferrous cmTDO upon stopped-flow mixing with O₂. (A) For the single-mixing stopped-flow experiment, an O₂-saturated solution was mixed with an anaerobic ferrous cmTDO solution. The concentration of TDO after mixing was 15 μM. (B) For the sequential-mixing stopped-flow experiment, in the first mix, Cld (5 μM) was combined with chlorite (20 mM) to generate a burst of O₂ at 10 mM; in the second mix, the O₂-enriched solution was combined with an anaerobic ferrous TDO solution. There was a 50 ms delay between the first mix and the second mix to ensure a complete production of O₂ from chlorite. The data were collected immediately after the second mix. The concentration of TDO after mixing was 100 μM, and the concentration of O₂ in the final reaction mixture was 5 mM. Notably, the concentration of TDO was increased compared to that of the single-mixing experiment to minimize the spectral contribution from Cld, which also contains heme. In panel B, the light path of the stopped-flow UV−vis measurement was reduced from 10 to 2 mm to maintain linearity of detector response. In both sets of plots, the arrows indicate the trends of the changes in the spectra, and the asterisks indicate the isosbestic points (i.e., 415, 531, and 607 nm) identified on the spectra. The insets show the time-resolved absorption change at 432 nm and corresponding exponential fittings. The X-axis of the inset of panel B is plotted on a log 2 scale for a better comparison between the two fitting results.

Figure 2.

Time-resolved (0−60 s) UV−vis spectra of ferrous cmTDO upon stopped-flow mixing with an O₂-saturated solution containing L-Trp. The final concentration of TDO after mixing was 15 μM. The final concentration of L-Trp was 2.5 mM. The final concentration of O₂ in the reaction mixture was ∼1 mM. The time course of the reaction can be divided into three distinct stages: (A) 0−0.2 s, (B) 0.2−3 s, and (C) 3−60 s. The arrows indicate the trends of the changes in the spectra.

Figure 3.

Time-resolved (0−60 s) change in the optical absorbance at 321 nm. The blue line is a linear fit of the experimental data during the steady state (0.2−3 s) of the reaction. See the legend of Figure 2 for experimental details.

Figure 4.

Spectral comparison between two ferrous TDO species captured in the stopped-flow reaction of cmTDO. See the legend of Figure 2 for experimental details.

Figure 5.

Time-resolved (0−100 s) UV−vis spectra of cmTDO in a sequential-mixing stopped-flow experiment. In the first mix, Cld (5 μM) was combined with a buffered solution containing chlorite (20 mM) and L-Trp (10 mM) to generate a burst of O₂ at 10 mM; in the second mix, the O₂-enriched solution containing L-Trp was combined with an anaerobic ferrous TDO solution. There was a 50 ms delay between the first mix and the second mix to ensure a complete production of O₂ from chlorite. Data collection was initiated immediately after the second mix. The final concentration of cmTDO after mixing was 15 μM. The final concentration of L-Trp was 2.5 mM. The final concentration of O₂ was 5 mM. The time course of the reaction can be divided into three distinct stages: (A) 0−0.04 s, (B) 0.04−1.2 s, and (C) 1.2−100 s. The arrows indicate the trends of the changes in the spectra. Notably, the absorbance at 432 nm first increased but then decreased in panel C. A control experiment was performed to assess the spectral contribution of Cld. The control experiment used the same sequential-mixing setup as described above, but with the anaerobic ferrous cmTDO solution replaced by an anaerobic buffer. The time-resolved UV−vis spectra of the control experiment showed that the spectral contribution of Cld was minimal (Figure S1).

Figure 6.

Mössbauer spectra of ⁵⁷Fe-enriched cmTDO in a sequential-mixing RFQ experiment. In the first mix, Cld (100 μM) was combined with a buffered solution containing chlorite (20 mM) and L-Trp (25 mM) in a 1:1 ratio to generate a burst of O₂; in the second mix, the O₂-enriched solution containing L-Trp was combined with an anaerobic ⁵⁷Fe-enriched ferrous TDO solution (2.1 mM) in a 2:1 ratio. There was a 50 ms pause between the first mix and the second mix to ensure a complete production of O₂ from chlorite. The reaction was quenched (A) 40 ms and (B) 7 s following the second mix. The final concentration of TDO after mixing is 0.7 mM. The final concentration of L-Trp is 8.3 mM. The final concentration of O₂ is 6.7 mM. The Mössbauer data were collected at 100 K. The doublet parameters of the simulations are indicated as δ/ΔE_Q, both in millimeters per second. The spectra are displayed for equal areas.

Scheme 1.

Proposed Catalytic Mechanism for IDO and TDO

Scheme 2.

Proposed Catalytic Mechanisms of cmTDO