Biosynthesis of Firefly Luciferin in Adult Lantern: Decarboxylation of L-Cysteine Is a Key Step for Benzothiazole Ring Formation

The biochemical ground truth, finally pinned down. Earlier ¹⁴C work from the 1970s (Okada, McCapra, Colepicolo) had pointed at cysteine and benzoquinone/hydroquinone as precursors but none of it was conclusive, and CBT, the long-standing textbook intermediate, had never actually been detected in a firefly. Oba injected stable-isotope-labeled precursors into living Luciola lateralis lanterns and tracked exactly which atoms ended up where via LC/ESI-TOF-MS. This paper is the in vivo proof.

• Two L-cysteines + one benzoquinone/hydroquinone → luciferin. Double-labeling with deuterated hydroquinone plus ¹³C-cysteine showed both molecules incorporated into the same luciferin in living tissue. One cysteine ends up in the benzothiazole ring (left half), the other in the thiazoline ring (right half). This is the three-precursor model your pathway is built on.
• The carboxyl carbon of cysteine is lost as CO₂ during benzothiazole ring formation. Comparing L-Cys[1-¹³C] vs L-Cys[3-¹³C] incorporation: the C3 label was retained, the C1 (carboxyl) label was not. This is the title finding and it nails down the mechanism McCapra had only predicted in 1976.
• Benzoquinone is the preferred substrate, not hydroquinone. Even at 10× lower concentration (55 vs. 550 nmol, benzoquinone is too toxic to inject more), benzoquinone showed higher incorporation efficiency. This implies hydroquinone is oxidized to benzoquinone before condensing with cysteine. The oxidation step is part of the pathway, not optional.
• Both D- and L-luciferin are made from L-cysteine. When enantiomers were resolved on a chiral column, L-cysteine fed into both forms. The basal D:L ratio (~9:1) shifted toward L (~7:3) after injection. This means the firefly must have a racemization or inversion step, likely the CoA-mediated D/L cycling proposed by Niwa et al. (2006), and L-luciferin is on-pathway, not a dead-end byproduct.
• Arbutin (hydroquinone-β-glucoside) is present in lanterns; free hydroquinone is not. This points to arbutin as a storage form, with a β-glucosidase releasing hydroquinone on demand for biosynthesis. It also explains why fireflies don't accumulate toxic free hydroquinone, they keep it sugar-capped until needed.

Bottom line for the project: This is the paper that justifies every gene downstream of the precursor pool. BGL is justified because arbutin storage is real. The hydroquinone → benzoquinone oxidation step is justified because benzoquinone is the actual substrate that condenses with cysteine. The benzothiazole-ring chemistry (Kanie 2016's spontaneous condensation in a buffer) was shown here to be what's actually happening inside a living firefly. If you're writing a paper or grant, this is the citation that turns “we think the pathway works like this” into “this is the established pathway.”