The Chemical Basis of Fungal Bioluminescence

Article abstract of DOI:10.1002/anie.201501779

Many species of fungi naturally produce light, a phenomenon known as bioluminescence, however, the fungal substrates used in the chemical reactions that produce light have not been reported. We identified the fungal compound luciferin 3-hydroxyhispidin, which is biosynthesized by oxidation of the precursor hispidin, a known fungal and plant secondary metabolite. The fungal luciferin does not share structural similarity with the other eight known luciferins. Furthermore, it was shown that 3-hydroxyhispidin leads to bioluminescence in extracts from four diverse genera of luminous fungi, thus suggesting a common biochemical mechanism for fungal bioluminescence.

Full text of DOI:10.1002/anie.201501779

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201501779
German Edition: DOI: 10.1002/ange.201501779
Bioluminescence
The Chemical Basis of Fungal Bioluminescence**
Konstantin V. Purtov, Valentin N. Petushkov, Mikhail S. Baranov, Konstantin S. Mineev,
Natalja S. Rodionova, Zinaida M. Kaskova, Aleksandra S. Tsarkova, Alexei I. Petunin,
Vladimir S. Bondar, Emma K. Rodicheva, Svetlana E. Medvedeva, Yuichi Oba, Yumiko Oba,
Alexander S. Arseniev, Sergey Lukyanov, Josef I. Gitelson,* and Ilia V. Yampolsky*
Abstract: Many species of fungi naturally produce light,
a phenomenon known as bioluminescence, however, the fungal
substrates used in the chemical reactions that produce light
have not been reported. We identified the fungal compound
luciferin 3-hydroxyhispidin, which is biosynthesized by oxida-
tion of the precursor hispidin, a known fungal and plant
secondary metabolite. The fungal luciferin does not share
structural similarity with the other eight known luciferins.
Furthermore, it was shown that 3-hydroxyhispidin leads to
bioluminescence in extracts from four diverse genera of
luminous fungi, thus suggesting a common biochemical
mechanism for fungal bioluminescence.
extracts prepared from the mycelium of Collybia velutipes
and Armillaria mellea fungi. The terms “cold and hot
extracts” were introduced by Dubois in 1885.[3] He found
that extraction of luminous beetle biomass with cold water
preserved the enzymatic activity of luciferase, while the same
extraction using hot water denatured the enzymes to leave
intact the low-molecular-weight luminescence substrate—
luciferin. The two extracts produced light when mixed
together.
The bioluminescence of fungi is thought to be a two-step
process. In the first step, a luciferin precursor is reduced by an
NAD(P)H-dependent enzyme to a true luciferin. In the
second step, luciferin is oxidized by air under luciferase
catalysis to produce visible light[4,5] (Scheme 1).
B
ioluminescence of higher fungi has been known at least
since the time of the ancient Greeks.[1] Fruiting bodies of
several species produce a constant bright glowing light, which
can be seen easily with the naked eye. The luminescence of
cell-free extracts from luminous fungi was demonstrated by
Airth and McElroy in 1959,[2] with the emission of light
achieved by adding nicotinamide adenine dinucleotide phos-
phate (NADPH) to a mixture of cold and hot aqueous
Scheme 1. Two-stage fungal bioluminescence scheme, proposed by
Airth and Foerster.[5]
Despite multiple attempts to isolate and structurally
characterize fungal luciferins,[6–10] the structural and chemical
basis of bioluminescence has remained obscure. The main
obstacle impeding the study of fungal luciferins is the
difficulty to obtain pure substances suitable for structural
studies because of their likely instability and low content in
the biomass. As a result, no reliable structural data has been
reported concerning the chemical nature of fungal biolu-
minescence, and the two-stage process of the bioluminescence
of fungi remains a hypothesis supported only by indirect data.
Cross-reactions between cold and hot extracts from diverse
fungal species suggests a uniform bioluminescence mecha-
nism in higher fungi.[11,12]
[*] Dr. K. V. Purtov, Dr. V. N. Petushkov, Dr. N. S. Rodionova,
Dr. A. I. Petunin, Dr. V. S. Bondar, Dr. E. K. Rodicheva,
Dr. S. E. Medvedeva, Prof. J. I. Gitelson
Institute of Biophysics
Siberian Branch of the Russian Academy of Sciences
Akademgorodok, Krasnoyarsk 660036 (Russia)
E-mail: jigitelson@gmail.com
Dr. M. S. Baranov, Dr. K. S. Mineev, Z. M. Kaskova, A. S. Tsarkova,
Prof. A. S. Arseniev, Prof. S. Lukyanov, Dr. I. V. Yampolsky
Institute of Bioorganic Chemistry
Russian Academy of Sciences
Miklukho-Maklaya 16/10, Moscow 117997 (Russia)
E-mail: ivyamp@ibch.ru
Prof. Y. Oba, Y. Oba
We focused on assaying the fruiting bodies of several
nonluminous fungal species collected in the forests around
Krasnoyarsk (Russia) for the presence of a luciferin precur-
sor. For this, the enzymatic bioluminescence assay[13] was
carried out on cold extracts from the bioluminescent myce-
lium of Neonothopanus nambi. We found activity, corre-
sponding to a fungal luciferin precursor, to be present in the
hot aqueous extracts of five nonluminous forest species.
Moreover, the luciferin precursor content in the fruiting
bodies of these five nonluminous fungi appeared to be about
100 times higher than in the mycelium of well-known
luminous species, such as N. nambi and Mycena citricolor
(see Figure S1 in the Supporting Information). We focused on
Graduate School of Bioagricultural Sciences, Nagoya University
Nagoya 464-8601 (Japan)
Prof. S. Lukyanov, Dr. I. V. Yampolsky
Pirogov Russian National Research Medical University
Ostrovitianov 1, Moscow 117997 (Russia)
[**] We thank Dr. K. V. Antonov for LC-HRMS experiments and
appreciate the contributions of Yoshiki Suzuki, Shusei Kanie,
Masashi Naito, and Manabu Bessho (experiments on Mycena
chlorophos species). We thank Dr. Ekaterina Putintseva for graphical
design. This research was carried out using equipment provided by
the IBCH core facility (CKP IBCH). This work was supported by the
Russian Science Foundation grant 14-50-00131.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201501779.
8124
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 8124 –8128

Angewandte
Chemie
Figure 1. A) Chromatographic profile of the ethyl acetate extract from Pholiota squarrosa. B) Structures of the compounds corresponding to the
chromatographic peaks. The luminescence activities of 1–6 are shown (signal/background ratio).
the fruiting bodies of Pholiota squarrosa in further isolation
experiments.
HRMS spectra as well as the synthesis, see the Supporting
Information).
Hispidin is a well-known representative of a styrylpyrone
The ethyl acetate extract from the fruiting bodies of
Ph. squarrosa was concentrated and subjected to preparative
reversed-phase HPLC to yield six different compounds 1–6
(Figure 1). Compounds 1–6 showed pronounced activities in
the enzymatic bioluminescence assay developed by Oliveira
and Stevani.[13] The luminescence of 1–6 was 24000, 80, 6670,
40, 1300, and 1000, respectively (signal/background ratio of
the assay, see the Supporting Information for details). In
addition, two pairs (1,3 and 5,6) were found to undergo
tautomerization to produce equilibrium mixtures of tauto-
mers during further chromatography of isolated pure frac-
tions.
class of fungal and plant secondary metabolites.[14,15] To
confirm its role as a luciferin precursor in luminous fungi, we
isolated it from glowing mycelium of N. nambi. In these
experiments, we took advantage of the observation that
soaking the mycelium in distilled water overnight leads to
a dramatic increase in the specific activities of the hot and
cold extracts (up to 250- and 140-fold, respectively). Ethyl
acetate extraction of the hot aqueous extract prepared from
the soaked mycelium of N. nambi followed by HPLC led to
the isolation of two bioluminescent compounds with retention
times and UV spectra identical to those of hispidin (1) and its
cis isomer (3; Figure 2). Moreover, these two compounds
Chromatography was then carried out on all six com-
pounds dissolved in either [D6]DMSO or [D6]acetone, and
1
their NMR spectra (1H, 2D DQF-COSY, 2D H-13C HSQC,
and 2D 1H-13C HMBC) were recorded. These spectra in
combination with the HRMS spectra allowed their chemical
structures to be determined: the spectra of the pairs 1 and 3,
and 5 and 6 were completely identical and contained two sets
of signals corresponding to the double-bond isomers of
hispidin (1: trans, 3: cis) and bisnoryangonin (5: trans, and
6: cis). Compound 2 was identified as a homodimer of hispidin
(3,3’-bishispidinyl) and compound 4 as a heterodimer of
hispidin and bisnoryangonin (3-bisnoryangonyl-14-hispidin).
The structural assignment of 1, 3, 5, and 6 was also confirmed
by comparing the chromatograms and spectra of these
samples with those of commercially available hispidin
(Sigma) and synthetic bisnoryangonin (for the NMR and
Figure 2. A) RP HPLC chromatogram of the ethyl acetate extract from
the mycelium of Neonothopanus nambi soaked in distilled water.
B) Comparison of (A) with HPLC profiles of synthetic hispidin (blue)
and bisnoryangonin (green) under the same chromatographic condi-
tions. Both hispidin and bisnoryangonyn occur as a mixture of cis and
trans double-bond isomers.
Angew. Chem. Int. Ed. 2015, 54, 8124 –8128
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8125

.
Angewandte
Communications
showed tautomerization behavior, specific bioluminescence
activities, and HRMS spectra (see Figure S6) identical to
hispidin and its isomer, identified above as luciferin precur-
sor(s). Overall 0.5 mg of a mixture of tautomers 1 and 3 was
isolated from 10 g of the soaked mycelium.
The content of the luciferin precursor hispidin—even in
soaked specimens of N. nambi and other bioluminescent
species, such as M. citricolor and Panellus stipticus—was many
times lower, than that in non-bioluminescent fungi, for
example, Ph. squarrosa. This fact probably explains why
previous attempts to identify hispidin as a substrate of fungal
bioluminescence were unsuccessful.
We then tested the hypothesis of Stevani and co-workers
that all luminous fungi share a common substrate.[11] We
analyzed three other bioluminescent species by using chro-
matography coupled with an enzymatic assay for the presence
and specific activity of hispidin. All the analyzed specimens
(M. citricolor, P. stipticus, and Armillaria borealis) were found
to contain hispidin and its cis isomer, which showed lumines-
cence activity in the bioenzymatic assay with cold extracts
from N. nambi.
We also tested commercial hispidin for activity with the
cold extracts prepared from the mycelia of N. nambi, M. citri-
color, P. stipticus, and A. borealis upon addition of NADPH.
In all cases, we observed dose-dependent bioluminescence,
strongly suggesting that hispidin is a common luciferin
precursor of all higher luminous fungi. To confirm this
conclusion, we measured the luminescence characteristics of
the cold extract from the Mycena chlorophos fruiting body
upon addition of hispidin and NADPH. The intensity of the
light emission produced by the mixture of the cold extract
from M. chlorophos with hispidin (6.1 mm) and NADPH
(0.18 mm) was easily seen with the naked eye and comparable
to that of the live fruiting body of the same fungus (Fig-
ure 3A–C). The luminescence spectrum of an in vitro mixture
was identical to the in vivo bioluminescence spectrum of the
M. chlorophos fruiting body, and also with the fluorescence
emission spectrum of hispidin (Figure 3D).
In accordance with the previous studies of fungal biolu-
minescence, our results showed that two different enzymatic
activities within the cold extract were required to produce
light from hispidin and NADPH. The first enzyme utilized
hispidin and NADPH as substrates and produced true
luciferin, which is capable of luminescence when mixed with
the second enzyme (luciferase). No additional cofactors
except molecular oxygen were required for the reaction of
luciferin with luciferase. The two enzymes present in the cold
extract could be separated by two independent methods:
ultracentrifugation or gel filtration using a Superdex 75
column (see the Supporting Information for experimental
details).
Figure 3. Comparison of the A) in vitro luminescence of the cold
extract from the M. chlorophos fruiting body with the i) hot extract from
the same fungus and NADPH, ii) hispidin and NADPH, iii) NADPH.
Exposure 5 s, ISO 6400, F3.5, NIKON D5100. B) The same mixtures as
in (A) measured with a Berthold Centro LB960 luminometer at 10 min
accumulation. i) 650000 rlu, ii) 45000000 rlu, iii) 43000 rlu. C) In vivo
luminescence of the M. chlorophos fruiting body compared to the
in vitro luminescence of the cold extract from the M. chlorophos fruiting
body with hispidin and NADPH. D) Fluorescence spectra of hispidin in
methanol, with excitation at 379 nm (black line), in vivo biolumines-
cence spectrum of the M. chlorophos fruiting body (red line), and
in vitro luminescence of the cold extract from M. chlorophos with
hispidin and NADPH (blue line).
(tr = 17.2 min), reaching a maximum concentration 35 min
after the onset of the reaction (Figure 4). At that point the
reaction was stopped by acidifying to pH 2, and the com-
pound eluting at 17.2 min was isolated by HPLC (19 mg was
obtained from 32 mg of hispidin).
The new compound produced a bright NADPH-inde-
pendent luminescence when mixed with the cold extract of
N. nambi (Figure S8). We also found that the microsomal
fraction of the cold extract was responsible of this lumines-
cence, in full accordance with Airth’s hypothesis[2,4,5] and
findings by Stevani and co-workers[11–13] (Scheme 1), thus
confirming the functional role of this compound as the fungal
luciferin.
The UV/Vis spectra of luciferin were similar to those of
hispidin (Figure 5). The 1H NMR spectrum of fungal luciferin
revealed the same pattern of protons as in hispidin. However,
the disappearance of one signal corresponding to H-3 of
hispidin was observed in luciferin (Figure 6). The HRMS
spectrum of luciferin showed
m/z 263.0571, which corresponds to the formula C13H11O6
(calcd m/z 263.0550). This formula shows that fungal luciferin
a molecular ion with
+
Considering the likely role of hispidin as the fungal
luciferin precursor, we converted it enzymatic into luciferin.
For this purpose, commercial hispidin was incubated with
a partially purified and soluble NADPH-dependent enzyme
obtained by gel filtration of the cold extract from N. nambi
mycelium (ca. 35 kDa fraction, Superdex 75 column) in the
presence of NADPH. HPLC analysis of the reaction mixture
showed gradual appearance of one major new component
Figure 4. Enzymatic synthesis of fungal luciferin from hispidin cata-
lyzed by a 35 kDa fraction of the cold extract from N. nambi mycelium
in the presence of NADPH.
8126
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 8124 –8128

Angewandte
Chemie
Scheme 2. Mechanism of fungal bioluminescence.
Figure 5. UV/Vis spectra of hispidin and fungal luciferin.
hispidin proved the lack of both kinds of enzymatic activities
in these extracts.
contains one additional oxygen atom compared to hispidin.
All these data taken together allow unambiguous structural
assignment of fungal luciferin as 3-hydroxylated hispidin or
(E)-6-(3,4-dihydroxystyryl)-3,4-dihydroxy-2H-pyran-2-one
(Figure 6).
The biochemical action of the soluble NADPH-depen-
dent enzyme responsible for luciferin biosynthesis is hydrox-
ylation, in contrast to the previous theory that it would reduce
a luciferin precursor, a hypothesis that persisted for more than
50 years[2,16]. In fact, many oxygen-dependent hydroxylases
utilize NAD(P)H as a cosubstrate, because full reduction of
one oxygen molecule requires four electrons, while mono-
hydroxylation of one substrate molecule provides only two
electrons.[17] The data reported here suggest the mechanism of
fungal bioluminescence shown in Scheme 2.
Our data indicate that the well-described fungal and plant
secondary metabolite hispidin is a luciferin precursor in at
least four evolutionary distant genera of luminous fungi. The
reactivity of the identified luciferin, 3-hydroxyhispidin, with
luciferases from the same four luminous fungi coupled with
the evolutionary breadth of the considered fungi suggests that
most, if not all, luminous fungi utilize hispidin as the luciferin
precursor and 3-hydroxyhispidin as luciferin. As evident from
the occurrence of hispidin in both luminous and nonluminous
fungi, the fungal bioluminescence is defined by the presence
of hispidin-3-hydroxylase and luciferase rather than by the
ability to biosynthesize the luciferin precursor hispidin.
Keywords: bioluminescence · bioorganic chemistry ·
biosynthesis · luciferin · natural products
The moderate luminescence activities of hispidin ana-
logues 2 and 4 might be explained by their ability to undergo
analogous hydroxylation at the pyranone fragment followed
by reaction with luciferase. Our examination of the ability of
the cold extract of the fruiting bodies of Ph. squarrosa to
catalyze 3-hydroxylation of hispidin in the presence of
NADPH and to produce light upon addition of 3-hydroxy-
How to cite: Angew. Chem. Int. Ed. 2015, 54, 8124–8128
Angew. Chem. 2015, 127, 8242–8246
[1] E. N. Harvey in A History of Luminescence From the Earliest
Times Until 1900, American Philosophical Society, Philadelphia,
USA, 1957.
[2] R. L. Airth, W. D. McElroy,
J. Bacteriol. 1959, 77, 249 –
250.
[3] R. Dubois, C. R. Soc. Biol.
1885, 37, 559 – 562.
[4] “Characteristics of cell-free
fungal bioluminescence”:
R. L. Airth in Light and
Life (Eds.: W. D. McElroy,
B. Glass), Johns-Hopkins
Press, Baltimore, 1961,
pp. 262 – 273.
[5] R. L. Airth, G. E. Foerster,
Arch. Biochem. Biophys.
1962, 97, 567 – 573.
[6] “Purification and properties
of the active substance of
fungal luminescence”: S.
Kuwabara, E. C. Wassink
in Bioluminescence in Prog-
ress (Eds.: F. H. Johnson,
E. Y. Haneda), Princeton
Figure 6. Comparison of the 1H NMR spectra of hispidin (unlabeled peaks correspond to an admixture of
the cis isomer of hispidin) and enzymatically synthesized fungal luciferin.
University Press, Princeton,
1966, pp. 233 – 245.
Angew. Chem. Int. Ed. 2015, 54, 8124 –8128
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8127

.
Angewandte
Communications
[7] M. Endo, M. Kajiwara, K. Nakanishi, J. Chem. Soc. D 1970, 309 –
310.
[13] A. G. Oliveira, C. V. Stevani, Photochem. Photobiol. Sci. 2009, 8,
1416 – 1421.
[8] M. Isobe, D. Uyakul, T. Goto, J. Biolumin. Chemilumin. 1987, 1,
181 – 188.
[14] C. Beckert, C. Horn, J.-P. Schnitzler, A. Lehning, W. Heller, M.
Veit, Phytochemistry 1997, 44, 275 – 283.
[9] M. Isobe, D. Uyakul, T. Goto, Tetrahedron Lett. 1988, 29, 1169 –
1172.
[10] S. Hayashi, R. Fukushima, N. Wada, Biophysics 2012, 8, 111 –
114.
[15] I. K. Lee, B. S. Yun, J. Antibiot. 2011, 64, 349 – 359.
[16] O. Shimomura, Bioluminescence: Chemical Principles and
Methods, World Scientific Publishing, Singapore, 2006.
[17] P. R. Ortiz de Montellano, Chem. Rev. 2010, 110, 932 – 948.
[11] A. G. Oliveira, D. E. Desjardin, B. A. Perry, C. V. Stevani,
Photochem. Photobiol. Sci. 2012, 11, 848 – 852.
[12] C. V. Stevani, A. G. Oliveira, L. F. Mendes, F. F. Ventura, H. E.
Waldenmaier, R. P. Carvalho, T. A. Pereira, Photochem. Photo-
biol. 2013, 89, 1318 – 1326.
Received: February 24, 2015
Published online: June 11, 2015
8128
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 8124 –8128