Fast Assembly and High-Throughput Screening of Structure and Antioxidant Relationship of Carotenoids

Article abstract of DOI:10.1021/acs.orglett.8b03915

C₂₀ heptaenyl diphosphonate 4 was prepared for one-pot synthesis of carotenoids 1. Olefination with various aromatic aldehydes allowed fast assembly of the corresponding carotenoids. The SAR of carotenoids was investigated by high-throughput screening of ABTS and DPPH assays and their hierarchical clustering analysis. Antioxidant activity of carotenoids increased with the number of electron-donating substituents. Carotene 1a with multiple electron-donating substituents was most proficient, which showed better radical scavenging activities than β-carotene and lycopene.

Full text of DOI:10.1021/acs.orglett.8b03915

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Fast Assembly and High-Throughput Screening of Structure and
Antioxidant Relationship of Carotenoids
Dahye Kim, Gaosheng Shi, YunJi Kim, and Sangho Koo*
Department of Energy Science and Technology, Department of Chemistry, Myongji University, Myongji-Ro 116, Cheoin-Gu,
Yongin, Gyeonggi-Do, 17058, Korea
S
* Supporting Information
ABSTRACT: C20 heptaenyl diphosphonate 4 was prepared for one-pot synthesis of carotenoids 1. Olefination with various
aromatic aldehydes allowed fast assembly of the corresponding carotenoids. The SAR of carotenoids was investigated by high-
throughput screening of ABTS and DPPH assays and their hierarchical clustering analysis. Antioxidant activity of carotenoids
increased with the number of electron-donating substituents. Carotene 1a with multiple electron-donating substituents was
most proficient, which showed better radical scavenging activities than β-carotene and lycopene.
Scheme 1. Comparison of the New One-Pot Synthesis of
Carotenoids 1 with the Conventional Method
arotenoids (1) are structurally unique natural products
that generally contain nine conjugated CC double
C
bonds in the central chain with two terminal rings.1 This
structural motif is related with their energy-transferring activity
from chlorophylls in photosynthesis.2 Antioxidant activity by
scavenging reactive oxygen species (ROS) is another important
role of these red pigments.3 Rapid destructive reactions with
singlet oxygen or radical species explain the antioxidant
mechanism of the conjugated polyene chain.4 Given lipid
peroxidation, DNA mutation, cancers, and other aging
processes being induced by ROS,5 it is necessary to develop
powerful antioxidants to eliminate ROS generated during the
respiration process.
It was envisioned that the structural modification of the
terminal rings would significantly alter the effective con-
jugation and, thus, the reactivity of the conjugated polyene
chain toward ROS.6 An efficient synthetic method, as well as
high-throughput screening (HTS), is an essential element for
gaining information about the structure and activity relation-
ship (SAR) to excavate lead compounds for new drugs.7
Devising a highly efficient (robust and expeditious) synthetic
method of carotenoids with various terminal rings would allow
fast assessment of SAR of carotenoids for antioxidant activity.
A conventional synthetic method of carotenoids relies on the
olefination of C10 2,7-dimethyl-2,4,6-octatrienedial (2) with
C15 unit 3 (Scheme 1), which should be prepared in multiple
steps generally from the corresponding aldehydes (RCHO) by
sequential chain extension with C3 (e.g., acetone) and C2 (e.g.,
ethylene) units, followed by functionalization with the linchpin
group X, which can be selected according to the olefination
method among PPh3 for Wittig, P(O)OEt2 for Wadsworth-
Emmons, and SO2BT (benzothiazole) for Julia-Kocienski.8
Novel C20 heptaene 4 containing phosphonate groups was
devised to facilitate one-pot synthesis of various carotenoids 1
Received: December 7, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03915
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
from diversely substituted aromatic aldehydes. New C5
building block 5 with two linchpin groups of P(O)OEt2 and
SO2BT was selected for sequential olefination,9 and was
selectively coupled at the SO2BT site with C10 dial 2 to
produce the key C20 diphosphonate 4.
A 2-fold Julia-Kocienski olefination of C5 unit 5 with C10 dial
2 utilizing NaHMDS as a base smoothly progressed at −78 °C
to room temperature (rt) to give C20 diphosphonate 4 (6:1 E/
Z) in 64% yield (Scheme 3). Reactivity of the less-hindered
Preparation of C5 units 5 and 8 were described in Scheme 2.
Bromohydrin formation (NBS, H2O) from isoprene, followed
Scheme 3. Preparation of C20 Diphosphonate 4 and
Expeditious Synthesis of Carotenoids 1
Scheme 2. Preparation of C5 Units 8 and 5 with Diethyl
Phosphonate and Benzothiazolyl Sulfone Groups
allylic carbon was reinforced by the more-reactive BT-sulfonyl
group. Subsequent Wadsworth-Emmons olefination of the
purified C20 all-(E)-diphosphonate 4 with various aromatic
aldehydes with MeOK base in MeOH/toluene at reflux
produced the corresponding carotenoids 1 containing aromatic
rings in reasonable 23%−66% yields (recrystallized from THF/
MeOH, see Table 1). The configuration of the polyene chain
was carefully assigned to be either all-E or 9′-Z, or a mixture of
by allylic bromination (PBr3, cat. CuI) produced 1,4-
dibromide 6 in 70% yield with 4:1 E/Z selectivity. The order
of phosphorylation and sulfonylation of dibromide 6 dictated
the regiochemistry of bifunctional C5 unit, in which the
internal methyl group allowed facile allylation at the distal
carbon by steric bias. C5 Unit 8 was prepared first and tested
for feasibility of sequential olefination. Arbuzov reaction with
P(OEt)3 gave allylic phosphonate 7 in 62% yield (4:1 E/Z).
Sulfonylation by sulfide coupling (BT-SH, K2CO3) and
monoperphthalic acid oxidation (urea−H2O2, phthalic anhy-
dride) produced C5 unit 8 in 62% yield (6:1 E/Z). The
preliminary reaction of 8 with β-cyclocitral unfortunately gave
no olefination at the phosphate site but elimination of BT-
sulfone to produce dienyl phosphonate 9 in 24% yield. It
seemed that superior reactivity of sterically less-crowded allylic
position (by remote methyl group) was mismatched with less-
reactive phosphonate group.
The regio-isomeric C5 unit 5 was prepared from dibromide
6 by the above two-step sulfonylation to produce allylic BT-
sulfone 10 (23% yield, E-form by recrystallization), followed
by phosphorylation with P(OEt)3 in 30% yield. This sequence
of reactions suffered from low yields, and modification of
reaction conditions was necessary to improve the yields of
reactions. To further differentiate the reactivity of the two
allylic positions, 4-bromo-1-chloro-2-methylbut-2-ene (11)
was prepared in 72% overall yield (5:1 E/Z) from isoprene
by chlorohydrin formation (NCS, H2O, DMF) and allylic
bromination (PBr3, cat. CuI). The yield of the two-step
sulfonylation to give BT-sulfone 12 was improved up to 68%
(E-form by recrystallization). Arbuzov reaction was sluggish at
the allylic chloride site of BT-sulfone 12, and thus Finkelstein
reaction (NaI, acetone) was preceded to produce C5 unit 5 in
76% yield (4:1 E/Z). The E/Z ratio seemed to be deteriorated
during the Finkelstein reaction. The pure E-isomer was easily
purified by silica gel column chromatography and used for the
next sequential olefination reactions.
1
both by comparing their H NMR spectra with those of all-
(E)-carotene 1b and 9′-(Z)-carotene 1g (Table 1).10,11 The
numbering sequence followed the carotenoid tradition, instead
of the systematic IUPAC rule.12
It was found that the configuration of the polyene chain was
controlled mostly by steric and electronic effects of the
substituents in the terminal ring: ortho-methyl substitution
preferentially produced all-E form; on the other hand, para-
substituent favored 9′-Z form.10 It was another noteworthy
feature that furan gave all-E form, but thiophene produced 9′-Z
form exclusively. Eighteen carotenoids were quickly assembled
in this way, and HTS for their antioxidant activities were
performed by measuring EC50 values (mol/L) for scavenging
ABTS13 [2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic
acid)] and DPPH14 (1,1-diphenyl-2-picrylhydrazyl) radicals.
Antioxidant activities can be also evaluated by lipid
peroxidation, DNA damage, OH radical and O2 anionic radical
scavenging, Cu(II)-chelating, Fe(III)-reducing assays.13a ABTS
and DPPH assays were chosen for simple and fast HTS of the
carotenoid antioxidant activities.
HTS is a process by which a large number of compounds
can be tested in an automated fashion for biological activities.7
The HTS assays for antioxidant with the microplate reader of
96-well proved to be robust and reproducible as well as time-
saving. EC50 was defined as a measure of activity to be the
concentration of antioxidant necessary to decrease the initial
radical absorbance (A) to its 50% value. The measurements of
EC50 (mol/L) for ABTS and DPPH assays were performed
four times for each sample and the values of average and
standard deviation were reported in Table 1. There are
relatively good correlations between ABTS and DPPH assays
in the order of carotenoids’ radical scavenging activities.
Hierarchical clustering analysis was performed for the EC50
values of carotenoids in ABTS and DPPH assays with
Euclidian distance as a similarity measure and Ward’s linkage
B
DOI: 10.1021/acs.orglett.8b03915
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Yield and Stereochemistry (All-E/9′-Z) of Carotenoids 1 Prepared from C20 Diphosphonate 4 and Their EC50 Values
for ABTS and DPPH Assays
a
The carotenoids were grouped (background shading) according to the hierarchical clustering analysis (see Figure S-1 in the Supporting
b
Information). The ratio of all-E/9′-Z configuration of carotenoids was calculated based on the integration of 1H NMR peaks of the mixture (see
c
page S84 in the Supporting Information as an example). All-E-carotene 1p was obtained in the synthesis of 9′-Z-carotene 1l (see page S24 in the
Supporting Information).
to extract unbiased information about the structural relevance
to the carotenoids’ antioxidant activity.15 Eighteen carotenoids
were grouped into five tribes in this way, which was denoted by
background shading in Table 1 (also see the dendrogram of
Figure S-1 in the Supporting Information).
The first group (Table 1, entries 1−3) demonstrated the
strongest antioxidant activities, which showed the structural
features of ortho- and multiple electron-donating substituents
in the aromatic rings. It suggested that ortho-substitution
induced more to the all-E configuration of the polyene chain.10
The second-best group in radical scavenging activity contained
an electron-donating substituent at the para-position of the
aromatic ring, which favored 9′-Z configuration of the
carotenoid polyene chain (Table 1, entries 4, 5, 7, and 8).
Two odd carotenes in this group are those with furan (Table 1,
entry 6) and meta-dimethylbenzene (entry 9) as terminal rings,
which exist as all-E form.
The aromatic rings of carotenoids in the third group
possessed para-bromo or ortho-methyl substituents (Table 1,
entries 10 and 13). 9′-(Z)-Carotene with benzene rings also
C
DOI: 10.1021/acs.orglett.8b03915
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
belonged to this group (Table 1, entry 12). The correlation
between ortho-substituent and all-E configuration was con-
firmed again (Table 1, entry 13). 9′-Z Configuration was the
major for the other carotenoids, including the one with
thiophene as the terminal rings (Table 1, entry 11). 9′-(Z)-
Carotenoids with 2-naphthalene or meta-toluene rings, and all-
(E)-carotene with benzene rings comprised the fourth group.
The structural feature of the last (fifth) group of carotenoids
was obvious that they contained electron-withdrawing para-
substituted benzene rings, which exhibited the lowest radical
scavenging activities (Table 1, entries 17 and 18).
It can be concluded from the above hierarchical clustering
analysis of HTS on the antioxidant activity of the carotenoids
that ortho- and para-electron-donating aromatic rings increase
the radical scavenging activity (groups 1 and 2), while electron-
withdrawing ones decrease it (group 5). It is not easy to draw a
general activity profile on the E vs Z configuration of
carotenoids.13b Higher activities of the carotenoids in group
1, compared to those in group 2, can be ascribed not to the
configuration of the polyene, but to the multiple electron-
donating aromatic substituents. The ortho-substituents in
group 1 induced the polyene chain more to all-E configuration
by steric reason, whereas the para-substituents in group 2
favored 9′-Z configuration presumably by effective conjuga-
tion.10 All-(E)-carotene with furan rings is more potent than
9′-(Z)-carotene with thiophene rings for the same rationale
(Table 1, entries 6 and 11). The activity order for the methyl-
substituent position in toluene rings was as follows: para >
ortho > meta (Table 1, entries 8, 13, and 15). Higher radical
scavenging activity was observed for 9′-(Z)-carotene contain-
ing benzene rings than its all-(E)-counterpart (Table 1, entries
12 and 16).
novel carotene 1a exhibits slightly better antioxidant activity
than those natural carotenoids do. It seems that electron-
donating groups are more important than effective conjugation
in stabilizing the carotenoid radical species in these radical
scavenging assays.
In conclusion, versatile C20 heptaenyl diphosphonate 4 was
developed for one-pot olefination of carotenoids 1 with various
aromatic aldehydes, which diversified the terminal structures of
the carotenoids. Fast assembly of diverse carotenoids, HTS for
ABTS and DPPH radical scavenging activities, together with
hierarchical clustering analysis provided structure and anti-
oxidant-activity relationships for the carotenoids. Electron-rich
aromatic rings increased the radical scavenging activity. The
activity order of the methyl substituent in the toluene ring was
as follows: para > ortho > meta position. Configurational effect
of the polyene chain on the activity was found to be 9′-Z > all-
E for the carotene with unsubstituted benzene rings. The
strongest antioxidant activity was observed for 1a with the
aromatic rings of multiple electron-donating substituents,
which exhibited stronger activity than natural β-carotene and
lycopene. The SAR of carotenoids for antioxidant activity
provides valuable designing principles for the carotenoid-based
antioxidant drugs.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b03915.
Materials and methods for ABTS and DPPH assays,
hierarchical clustering analysis, experimental procedures
and analysis data for all new compounds (PDF)
HTS of the radical scavenging activities was repeated for the
representative carotenoids 1a, 1d, 1j, 1n, and 1q in the above
five groups, together with β-carotene and lycopene as positive
controls. Measurement of EC50 values was perfromed in
quadruplicate, and the average values were depicted as bar
graphs in Figure 1. The same activity trend was confirmed for
the above five groups. It was found that the natural carotenoids
are very good radical scavenging antioxidants, but most of all,
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: sangkoo@mju.ac.kr.
ORCID
Sangho Koo: 0000-0003-4725-1117
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Research
Foundation of Korea through the Basic Science Research
Program (No. NRF-2016R1A2B4007684) and partly by the
Korea Institute of Energy Technology Evaluation and Planning
(KETEP) and the Ministry of Trade, Industry & Energy
(MOTIE) of Korea (No. 20174010201160).
REFERENCES
■
(1) (a) Carotenoids; Briton, G., Liaaen-Jensen, S., Pfander, H., Eds.;
̈
Birkhauser: Basel, Switzerland, 1995. (b) Cane, D. E. Comprehensive
Natural Products Chemistry, Vol. 2; Barton, D., Nakanishi, K., Eds.;
Elsevier: Oxford, U.K., 1999.
(2) (a) Mackowski, S.; Wormke, S.; Brotosudarmo, T. H. P.; Jung,
C.; Hiller, R. G.; Scheer, H.; Brauchle, C. Biophys. J. 2007, 93, 3249−
3258. (b) Frank, H. A.; Brudvig, G. W. Biochemistry 2004, 43, 8607−
8615.
(3) (a) Krinsky, N. I. Annu. Rev. Nutr. 1993, 13, 561−587.
(b) Skibsted, L. H. J. Agric. Food Chem. 2012, 60, 2409−2417.
Figure 1. Comparison of antioxidant activities of the selected
carotenoids in Table 1 with those of β-carotene and lycopene by EC50
(mol/L) in ABTS and DPPH assays (see Table S-3 in the Supporting
Information).
D
DOI: 10.1021/acs.orglett.8b03915
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(4) (a) Burton, G. W. J. Nutr. 1989, 119, 109−111. (b) Krinsky, N.
I. Free Radical Biol. Med. 1989, 7, 617−635. (c) Palozza, P.; Krinsky,
N. I. Methods Enzymol. 1992, 213, 403−420. (d) Mortensen, A.;
Skibsted, L. H.; Sampson, J.; Rice-Evans, C.; Everett, S. A. FEBS Lett.
1997, 418, 91−97.
(5) Murphy, M. P. Biochem. J. 2009, 417, 1−13.
(6) (a) Miller, N. J.; Sampson, J.; Candeias, L. P.; Bramley, P. M.;
Rice-Evans, C. A. FEBS Lett. 1996, 384, 240−242. (b) Liao, F.-X.;
Hu, C.-H. Theor. Chem. Acc. 2013, 132, 1357.
(7) (a) Broach, J. R.; Thorner, J. Nature 1996, 384, 14−16.
(b) Major, J. J. Biomol. Screening 1998, 3, 13−17. (c) Herald, T. J.;
Gadgil, P.; Tilley, M. J. Sci. Food Agric. 2012, 92, 2326−2331.
(8) (a) Pommer, H.; Kuhn, R. Angew. Chem. 1960, 72, 911−915.
(b) Pommer, H. Angew. Chem., Int. Ed. Engl. 1977, 16, 423−429.
(c) Paust, J. Pure Appl. Chem. 1991, 63, 45−58. (d) Choi, J.; Oh, E.-
T.; Koo, S. Arch. Biochem. Biophys. 2015, 572, 142−150.
(9) (a) Cichowicz, N. R.; Nagorny, P. Org. Lett. 2012, 14, 1058−
1061. (b) Coleman, R. S.; Walczak, M. C. Org. Lett. 2005, 7, 2289−
2291.
̀
́
(10) Kim, M.; Jung, H.; Aragones, A. C.; Díez-Perez, I.; Ahn, K.-H.;
Chung, W.-J.; Kim, D.; Koo, S. Org. Lett. 2018, 20, 493−496.
(11) The ratio of all-E/9′-Z configuration of carotenoid was
1
calculated in H NMR spectrum by comparison of the integration
values for H10 (contributed solely from 9′-Z) and H12 (contributed
both from 9′-Z and all-E); see page S84 in the Supporting
Information for details.
(12) Britton, G. Methods Enzymol. 1985, 111, 113−149.
(13) (a) Li, X.; Lin, J.; Gao, Y.; Han, W.; Chen, D. Chem. Cent. J.
2012, 6, 140. (b) Zhang, Y.; Fang, H.; Xie, Q.; Sun, J.; Liu, R.; Hong,
Z.; Yi, R.; Wu, H. Molecules 2014, 19, 2100−2113.
́
(14) (a) Sanchez-Moreno, C.; Larrauri, J. A.; Saura-Calixto, F. J. Sci.
́
́
Food Agric. 1998, 76, 270−276. (b) Jimenez-Escrig, A.; Jimenez-
́
́
Jimenez, I.; Sanchez-Moreno, C.; Saura-Calixto, F. J. Sci. Food Agric.
2000, 80, 1686−1690. (c) Sharma, O. P.; Bhat, T. K. Food Chem.
2009, 113, 1202−1205.
(15) (a) Ward, J. H., Jr. J. Am. Stat. Assoc. 1963, 58, 236−244.
(b) Orłowska, M.; Pytlakowska, K.; Mrozek-Wilczkiewicz, A.; Musioł,
R.; Waksmundzka-Hajnos, M.; Sajewicz, M.; Kowalska, T. Acta
Chromatogr. 2016, 28, 207−221.
(16) Brahmana, H. R.; Katsuyama, K.; Inanaga, J.; Katsuki, T.;
Yamaguchi, M. Tetrahedron Lett. 1981, 22, 1695−1691.
(17) (a) Hand, E. S.; Belmore, K. A.; Kispert, L. D. Helv. Chim. Acta
1993, 76, 1939−1948. (b) Gao, G.; Wurm, D. B.; Kim, Y.-T.; Kispert,
L. D. J. Phys. Chem. B 1997, 101, 2038−2045.
(18) (a) Francis, G. W.; Luning, B.; Enzell, C. R.; Tricker, M. J.;
̈
Svensson, S. Acta Chem. Scand. 1972, 26, 2969−2971. (b) Linner, E.;
Eugster, C. H.; Karrer, P. Helv. Chim. Acta 1955, 38, 1869−1874.
E
DOI: 10.1021/acs.orglett.8b03915
Org. Lett. XXXX, XXX, XXX−XXX