The Nitrogen Atom of Vitamin B6 Is Essential for the Catalysis of Radical Aminomutases

Abstract

Radical aminomutases are pyridoxal 5'-phosphate (PLP, a B₆ vitamer)-dependent enzymes that require the generation of a 5'-deoxyadenosyl radical to initiate the catalytic cycle, to perform a 1,2 amino group shift reaction. The role of the nitrogen atom of PLP in radical aminomutases has not been investigated extensively yet. We report an alternative synthetic procedure to provide easy access to 1-deazaPLP (dAPLP), an isosteric analog of PLP which acts as a probe for studying the role of the nitrogen atom. Our results revealed that lysine 5,6-aminomutase (5,6-LAM), a radical aminomutase, reconstituted with dAPLP cannot turn over a substrate, demonstrating that the nitrogen atom is essential for radical aminomutases. In contrast, biochemical and spectroscopic studies on the S238A variant reconstituted with PLP revealed a minuscule loss of activity. This apparent anomaly can be explained by a water-mediated rescue of activity in S238A, as if mimicking the active site of lysine 2,3-aminomutase. This study leads to a better comprehension of how enzymes harness the optimum capability of PLP to realize catalysis.

Keywords: DFT; EPR; PLP; aminomutase; coenzyme B12; dAdoCbl; mutagenesis.

Figure 1

Chemical structures of PLP and 1-deaza-PLP (dAPLP).

Scheme 1

Proposed reaction mechanism of radical aminomutases.

Scheme 2

Chemoenzymatic synthesis of dAPLP. (A), The reaction steps followed by Griswold et al; (B), The reaction steps followed in this work.

Figure 2

TLC chromatogram of reaction of D-lysine with WT, PLP/WT, dAPLP/WT and PLP/S238A in the presence of dAdoCbl. The time intervals were chosen at 3 min (left), 30 min (middle), and 60 min (right), respectively. Bottom spots are of D-lysine, whereas top spots represent the product 2,5-diaminohexanoic acid. Thus, WT and dAPLP/WT did not show any turnover, while PLP/S238A showed significant but less turnover than PLP/WT.

Figure 3

EPR spectrum in the reaction with 4-thialysine with dAPLP/5,6-LAM. Experimental parameters: microwave frequency, 9.487 GHz; power, 2 mW; modulation, 8 G at 100 kHz; T = 80 K. Simulation: g_Co = [2.278 2.227 2.000], g_radical = [2.014 2.01 2.004], Euler’s angle = [22° 28° −5°]; A_Co = [20 3 110] G; A_N = 19 G; J = 8178 G; D = −168 G; E = 0; Euler angle = [31° 48° 5°]; line width = 18 G.

Figure 4

Stopped-flow absorbance changes (black) following transaldimination and homolysis in the S238A variant (top panel) and WT 5,6-LAM (bottom panel). Transaldimination (left panel) was monitored at 423 nm. Homolysis (right panel) was monitored at 522 nm. The rate constant was extracted with a fitted single-exponential decay function (red).

Figure 5

EPR spectra of S238A variant reaction with 4-thial-L-lysine at reaction time 8 s. EPR spectra of WT (red) and S238A (black) showing decreased signal intensity. The features of both spectra are identical. Experimental parameters: microwave frequency, 9.467 GHz; power, 2 mW; modulation, 8 G at 100 kHz; T = 80 K.

Scheme 3

Model of radical intermediates in the reaction of 5,6-LAM with lysine.

Figure 6

Relative energies of models of important radical intermediates in the reaction of 5,6-LAM.

Figure 7

Stereo view of the active-site interactions of D-lysyl-PLP-H₂O docked into the open state of the S238A variant of 5,6-LAM. Selected residues of 5,6-LAM (cyan) are shown in stick representation. D-lysyl-PLP (yellow) and water are also shown in stick representation.