An Efficient Multigram Synthesis of Hypericin Improved by a Low Power LED Based Photoreactor

Article abstract of DOI:10.1021/acs.oprd.7b00317

In this work, an improved synthesis process was developed for the multigram production of hypericin. An inexpensive and efficient low power Light Emission Diode (LED) based photoreactor was designed and employed to perform the protohypericin photocyclization reaction allowing its photoconversion in hypericin. This closed system overcomes safety issues related to scale-up hypericin preparation typically described in the literature which combines the use of open systems, organic solvents, and high-power light sources. The photoreactor designed allows a solution to, mainly, the intrinsic effect of hypericin photobleaching inherent to the protohypericin photocyclization reaction, implying an increase in the yield of the final product and consequently the final cost. Using a red-LED based photoreactor, a safety protocol was carried out in a 5-g scale hypericin preparation with quantitative yield.

Full text of DOI:10.1021/acs.oprd.7b00317

Subscriber access provided by READING UNIV
Full Paper
An efficient multigram synthesis of hypericin
improved by a low power LED based photoreactor
Renato Sonchini Gonçalves, Bruno Ribeiro Rabello, Gabriel Batista Cesar, Paulo
C. S. Pereira, Marcos Alessandro dos Santos Ribeiro, Eduardo C. Meurer, Noboru
Hioka, Celso Vataru Nakamura, Marcos Luciano Bruschi, and Wilker Caetano
Org. Process Res. Dev., Just Accepted Manuscript • Publication Date (Web): 28 Nov 2017
Downloaded from http://pubs.acs.org on November 28, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Organic Process Research & Development is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
An efficient multigram synthesis of hypericin improved by a low power
LED based photoreactor
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Renato S. Gonçalves*†, Bruno R. Rabello‡, Gabriel B. César†, Paulo C. S. Pereira†, Marcos A. S.
Ribeiro† Eduardo C. Meurerǁ, Noboru Hioka†, Celso V. Nakamura§, Marcos L. Bruschi§ and
,
Wilker Caetano†
† Department of Chemistry, State University of Maringá, Avenue Colombo, 5790, Maringá,
Paraná, 87020ꢀ900, Brazil.
‡ Instituto Federal de Educação, Ciência e Tecnologia Catarinense, 283, Concórdia, Santa
Catarina, 89703ꢀ720, Brazil.
§ Department of Pharmacy, State University of Maringá, Avenue Colombo, 5790, Maringá,
Paraná, 87020ꢀ900, Brazil.
ǁ Federal University of Parana, Jandaia do Sul, Paraná, 86900ꢀ000, Brazil.
ABSTRACT: In this work, an improved synthesis process was developed for the multigram
production of hypericin. An inexpensive and efficient low power Light Emission Diode (LED)
based photoreactor was designed and employed to perform the protohypericin photocyclization
reaction allowing its photoconversion in hypericin. This closed system overcomes safety issues
related to scaleꢀup hypericin preparation typically described in the literature which combine the
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 24
1
2
3
4
5
6
7
8
9
use of open systems, organic solvents and highꢀpower light sources. The photoreactor designed
allow to solve mainly the intrinsic effect of hypericin photobleaching inherent to protohypericin
photocyclization reaction, implying in the yield increased of the final product and consequently
in the final cost. Using a red LED based photoreactor a safety protocol was carried out to 5ꢀgram
scale hypericin preparation with quantitative yield.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS: Hypericin synthesis, LED, photoreactor, photobleaching effect, photocyclization
reaction, electronic absorpition spectroscopy, kinetic studies, gramꢀscale.
INTRODUCTION
Hypericin is a naturally occurring naphtodianthrone found in plants of the genus Hypericum,
commonly known as Saint John's Wort.1 Hypericin have attracted increasing attention due to
broad spectrum of pharmacological applications, among which antiꢀdepressive, antiviral, antiꢀ
inflammatory, and antiꢀtumoural activities have been reported.2ꢀ4 However, the difficulty in
obtaining hypericin in both significant amount and pure state direct from plant material imply in
its very higher commercial cost (1 mg ~ 95%, USD 815.81), unfeasibly its utilization on largeꢀ
scale. In order to overcome this limitation, different synthetic routes have been proposed for
hypericin including the first total synthesis comprising a 12 process steps, starting from 3,5ꢀ
dimethoxyphtalic anhydride and mꢀcresol.5ꢀ12 One year later, the synthesis of hypericin was
accomplished by a semiꢀsynthetic route starting from emodin anthrone, which is converted to
emodin bianthrone by an oxidative dimerization process followed by a baseꢀcatalyzed oxidation
to protohypericin and finally exposure to bright sunlight to yield hypericin with a low yield of
10%.13 This semiꢀsynthetic strategy was later adapted by Steglish et al. and adopted by others.14ꢀ
16
Starting from emodin, the procedure consist in a reductive coupling by a 3 week long
ACS Paragon Plus Environment
2

Page 3 of 24
Organic Process Research & Development
1
2
3
4
5
6
treatment with alkali and hydroquinone to obtained protohypericin, which after workup was
irradiated with sunlight to obtained hypericin in 29% yield.
7
8
9
In 1998, Mazur et al. improved this semiꢀsynthetic route employing an oxygen transfer
reagent (pyridine Nꢀoxide) and a conventional redox catalyst for the oxidative dimerization of
emodin to protohypericin. The photocyclization reaction of protohypericin was carried out with
employed of halogen lamp (500 W), which after a prolonged irradiation time yielded 60% of
hypericin.17 Interestingly, in the works reported later concerning the synthesis of hypericin and
its derivatives, were described the use of highꢀpower (400ꢀ1000 W) and multifrequential light
sources in the photocyclization step.18ꢀ29 Moreover, as far as we know, none of these works
considers the possibility of the hypericin photobleaching effect when the photocyclization step is
conducted under uncontrolled experimental conditions.30,31 The photobleaching effect is
characterized by photochemical changes on dyes molecular structure resulting of covalent bonds
disruption, which are integrant of chromophore and fluorophore groups. As a result of this
chemical disruption, the dye loses its characteristic color and/or its ability to emit fluorescence
for instance.25 As such, the present case concerning the hypericin formation whose electronic
absorption spectral show an expressive overlapping with the emission spectral of the light
source, the photobleaching effect is relatively pronounced. Therewith, the low hypericin yield
associated with its photobleaching effect and induced by the use of highꢀpower and
multifrequential light source can impact mainly on its largeꢀscale production cost. Therefore,
these are factors of great study interest, in the sense of seeking alternatives for the minimization
of the physicꢀchemical and photophysical parameters that may induce the photobleaching
process in these systems. In this work, we show that the overall yield of hypericin synthesis
provided by preceding works can be significantly improved if the protohypericin
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 24
1
2
3
4
5
6
7
8
9
photocyclization reaction is performed under lowꢀpower Light Emission Diode (LED) sources
that show wellꢀdefined spectral irradiance bands. Here, we described the design and application
of a LED based photoreactor to improve the multiꢀgram scale preparation of hypericin.
Specifically, our project is applied in the protohypericin photoconversion reaction, the final step
of the hypericin synthesis. Our project brings together innovative factors from earlier designs: (i)
The substitution of relative highꢀpower light sources (Watt) by lowꢀpower LEDs light sources
(mW); (ii) A compact and transparent flow cell in zigꢀzagging format maximizing the exposure
of the circulating sample to the radiation flux; (iii) A vertically mounted system in modular units
allowing an adjustment of the specific light dose (mW/cm2) and reaction time based on the initial
amount of reagent; (iv) Controlled recirculation of the reagent into the flow cell ensuring a
maximum protohypericin photoconversion (v) Coupling of an optical fiber to the output of the
sample stream, interfaced to an analysis device. All those parameters adjusted allow to optimized
the protohypericin photoconversion step, reducing the temporal exposure of hypericin to specific
light irradiation with relative lowꢀpower light dose, simultaneously modulating the photoreaction
and minimizing the photoinduced processes that lead to its photodegradation. Among the
advantages of our design is the fact that the photocyclization reaction can be performed in a
closed system, overcomes safety issues related to scaleꢀup hypericin preparation typically
described in the literature which combine the use of open systems, organic solvents and highꢀ
power light sources.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
4

Page 5 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Photoreactor Design. The prototype LED based photoreactor described in this study was
comprised by at least four or more photoꢀreactive modular units vertically arranged in series. The
modular unit consist of a threeꢀbay aluminum chassis (10 x 20 x 8 cm) supported by a multiple
heat sink cooling system. In the central bay is inserted the flow cell, which is consists of a
transparent Pyrex tube in zigꢀzagging format (3 mm intern diameter, 180 cm length). In both
faces of the modular unit sits two irradiation systems, composed by six LEDs blocks of 3.0 x 4.5
cm2. Each LEDs block consists of 36 x red 1210ꢀSMD (Surface Mounted Diode) LEDs of light
dose 12.5 mW/cm2 (with a maximum emission at 530 nm), positioned at 1.0 cm and parallel
from flow cell (Figure 1). For supply the protohypericin solution into photoreactor, the bottom of
the photoreactor is equipped with a Supelco PEEK fitting that attaches to 1/16” tubing that
connects to a metering pump. The protohypericin solution is pumped from feed flask (amber
glass) into the flow cell by down of the photoreactor. The flow rate of the metering pump can be
adjusted based on the initial concentration of reagent solution (protohypericin). At the top of the
photoreactor is fitted with a Supelco PEEK fitting that attaches to 1/16” tubing that connects to a
microꢀsplitter valve interfaced to a spectrophotometer and equipped with a continuous flow
quartz cuvette. The control valves allow to adjust the sample input on the spectrophotometer and
recirculate the reagent into the flow cell until its total photoconversion in hypericin. The S40
thread adaptors at the feed and collection flasks provide necessary air (O2) to be mixed with the
reagent solution before get in to the flow cell.
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. (A) Process flow diagram of the protohypericin photoconversion. (B) Showing the
photoreactor vertically arranged with four photoꢀreactive modular units interlocked. (C) Showing
the deconstructed modular unit. The flow cell is inserted between two glass slabs preventing the
intimate contact of the sample with the irradiation systems.
Synthesis of Hypericin. Naphtodianthrone hypericin was synthesized through adapted
literature procedures employing the emodin as starting material (Scheme 1).17ꢀ19 Significant
amount of emodin was accessed reproducing the optimized protocol for the isolation of
anthraquinones from Cortex Frangula.34 Briefly, dry bark of stem and branches of Rhamnus
frangula L. was tritured and macerated in MeOH at room temperature. The extracts were
concentrated, hydrolysed in acid medium and exhaustively extracted with DCM in a Soxhlet
apparatus. Isolation of anthraquinones from crude extract was performed by silica gel column
chromatography employed a mixture of CHCl3 : AcOEt with a gradient of 0ꢀ15% as mobile
phase. Elution with an increase of the solvent polarity (addition of 15% AcOEt) provide an
eluate of orange color. TLC analysis revealed that the fractions correspondent to its eluate were
1
sufficiently purified, and the isolated emodin (≥ 98%) was confirmed by H NMR analyse
(Figures S7 and S8).
ACS Paragon Plus Environment
6

Page 7 of 24
Organic Process Research & Development
1
2
3
4
5
6
Scheme 1. Optimized experimental conditions for the hypericin synthesis.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Next, the emodin was submitted to reduction reaction with tin(II) chloride dihydrate
(SnCl2.2H2O) and concentrated hydrogen chloride as reducing agents to afforded the emodin
anthrone. Unlike of the experimental conditions reported in the precedent works, we employed
optimized reaction parameters for the emodin reduction reaction. Thus, 3.2 and 100.0 mol equiv
o
of SnCl2.2H2O and HClconc respectively, at 120 C for 30 min were set to converted efficiently
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 24
1
2
3
4
1
emodin to emodin anthrone. The crude H NMR analyze reveal a yield of 87% of emodin
5
6
7
8
anthrone, which was used in the next step without previously purification (Figures S9ꢀS11).
Having stablished the emodin reduction reaction, we set out to synthesize protohypericin by
oxidative dimerization of emodin anthrone with reproduction of the procedure reported in the
patent literature.17 Emodin anthrone was treated with a combination of pyridine Nꢀoxide,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
pyridine, piperidine and ferrous sulfate at 100 C for 1 h to obtained the protohypericin.
Although, in this precedent work the authors did not have reported a purification procedure for
isolation of protohypericin from crude product, here we employed sizeꢀexclusion column
1
chromatography and ethanol as mobile phase for yielded 70% of pure protohypericin. The H
NMR spectroscopy analyze show a clear signature of protohypericin (Figure S12 and S13).
Finally, protohypericin solution in acetone was photoconverted to hypericin with excellent yield
by employing the low power redꢀLED photoreactor designed here, moreover the optimal LED
source used for the protohypericin photoconversion was set with base on the kinetical studies
1
showed with details in the next sessions. H and 13C NMR and highꢀresolution mass
spectroscopy analysis were performed to confirm unequivocally the structure of hypericin
(Figures S14ꢀS17).
Protohypericin Photoconversion. The photoconversion step of the protohypericin to
hypericin can be performed under irradiation with lowꢀpower light sources with wellꢀdefined
spectral irradiance bands. The main advantage of this procedure in face to previous reports that
use high power and multifrequential light sources is achieve the optimal experimental conditions
for this step towards to multiꢀgram hypericin preparation process. The use of lowꢀpower light
sources (LEDs) on the photoconversion stage was firstly inferred from the study published by
Poutaraud et al.32 The authors analysed empirically the influence of polychromatic light (LEDs
ACS Paragon Plus Environment
8

Page 9 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
of the first generation) in the in situ photosynthesis of the pigments hypericin and
pseudohypericin from St. John´s Wort and its different extracts. In this case, the optimal
experimental condition observed was achieved mainly upon irradiation with green LED light
source. However, for exclusive photoconversion process involving protohypericin solutions in a
pure state, instead of complex plant material containing different biomolecules classes, the use of
light source with different wavelength comprising the visible spectral range must be taken into
account aiming to achieve the optimal experimental conditions, therefore providing more
accurate information regarding specific physicalꢀchemistry and photochemical parameters
related with overall photoconversion process, on the other hands, used for the improved of the
prototype photoreactor.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Aiming the optimization of the photoreaction device, in this work the multifrequential
dependence on the protohypericin photocyclization reaction have been evidenced through
specific experiments employing emission light sources with distinct photonic parameters.
Specifically, intrinsic factors should be taken into account as wavelength emission and power
LED light source, light dose, time of irradiation, prodrug concentration, and the spectral overlap
between light source emission and electronic absorption spectra of protohypericin and hypericin.
Moreover, the inherent effects of the hypericin photobleaching also should be relevant leading to
the choice of desirable specific lowꢀpower light source. The optimized conditions for the
photoconversion of protohypericin did not result in a 20% photodegradation of hypericin in the
first three minutes of LED irradiation when compared to the usual light sources (mercury vaporꢀ
lamp).17ꢀ29 Even more significantly, twenty minutes of UV exposure the hypericin show an
expressive decreased of more than 90% of yield, as will be discussed below. However, the use of
LED light sources enables to obtained the photoproduct in excellent yield even to a prolonged
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 24
1
2
3
4
5
6
irradiation time, overlapping the overall yield of the hypericin synthesis reported in the preceding
works.
7
8
9
Concerning with the studies in micro scale, as a basis for photoreactor prototype design and
readjustment the protohypericin photocyclization reaction kinetic profile were assessed by
electronic absorption spectroscopic measurements. In a specific micro scale setup described in
the Experimental Session, when an 2.0 ꢁL protohypericin solution (1.6 x 10ꢀ5 mol.Lꢀ1) is
irradiated in few minutes, its initial purple color relative to the maximum absorpition band at λ550
begins to turn red, concomitantly with the arising in the electronic absorption spectra of the
hypericin absorption band at λ596, which become more pronounced as more hypericin molecules
are formed. In a first analyse, we prior investigated in micro scale the effect of the polychromatic
LEDs on the protohypericinꢀhypericin photoconversion efficiency (Figures S1ꢀS4). To our
delight, all reactions led to excellent yields (> 97%) independently on the LED type (Table 1,
entries 1ꢀ4). As earlier mentioned, the high photoconversion yields observed here are due mainly
to the minimized hypericin photobleaching effect.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Experimental conditions obtained to the protohypericin photoconversion process under
different light sources.
Yield (%)a
Entry Conditions
Blue LED
(463 nm, 10.2 mW/cm2), 20 min.
1
2
3
4
5
6
97
Green LED
(504 nm, 2.2 mW/cm2), 28 min
98
Orange LED
(593 nm, 1.0 mW/cm2), 22 min
98
Red LED
(629 nm, 0.6 mW/cm2), 74 min
Quantitative
Blue LED
(463 nm, 10.2 mW/cm2), 800 min
89
79
Mercuryꢀvapor lamp
(multifrequential of 400 W), 7 min
ACS Paragon Plus Environment
10

Page 11 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Mercuryꢀvapor lamp
(multifrequential of 400 W), 35 min
7
4
The optimal experimental condition is indicated in bold.
Yields were calculated based on epsilon values corresponding to the maximum absorption
bands of protohypericin and hypericin.
a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The subsequent experiment where performed under long irradiation time, above of 10 hours
of blue LED exposure (maximum emission at 530 nm) in order to investigated the relationship of
LED sources with the hypericin photobleaching effect. The choice of this light source was made
on the basis of its relatively low spectral overlap whit the main hypericin absorption band (λ596),
and also due to its higher light dose (10.2 mW/cm2) relative to the other LEDs systems studied in
this work. The Figure 2A show the electronic absorption spectra changes of the initial
protohypericin solution (1.6 x 10ꢀ5 mol.Lꢀ1) during its photoconversion to hypericin.
Interestingly, the kinetic profile monitoring of the hypericin absorption band at λ596 show a
decrease of only 8% of absorbance, and consequently in its yield, relative to the maximum
absorbance value achieved at 30 min through blue LED exposure (entry 5). This result
demonstrated that the rate of the hypericin photobleaching effect is minimum relative to the time
needed to reach its maximum yield irradiation with lowꢀpower LED sources.
However, at this work stage, we demonstrated the influence of the highꢀpower
multifrequential mercuryꢀvapor lamp sources on the hypericin photobleaching effect. Thus, we
performed kinetic studies of the protohypericin photoconversion with the same initial
protohypericin concentration (1.6 x 10ꢀ5 mol.Lꢀ1) similar to the previous microꢀscale studies. In
order to compare with the data described in the literature, we have monitored the protohypericin
photoconversion via electronic absorption under mercuryꢀvapor lamp irradiation exposure (400
W) for a reaction time of 30 min. The Figure 2B shows the complete hypericin formation at 7
min of UV exposure, in agreement with the reaction times reported in all the previous works (10ꢀ
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 24
1
2
3
4
5
6
7
8
9
30 min). However, at this photoreaction stage which the hypericin absorption band (λmax = 596
nm) display a maximum intensity, was observed a significant decrease of 20% in the hypericin
yield in comparison with the set optimal experimental condition (compare entry 4 with 6).
Furthermore, the photobleaching effect becomes still more pronounced for UV exposure for
a prolonged irradiation time (> 10 min). Under this extreme experimental condition, the
hypericin molecules shown a faster photodegradation (Figure 2C), and the mainly hypericin
absorption band show a decreased of 50%, visibly observed by the gradual discolored of the
solution. Surprisingly, under 30 min of UV exposure, the hypericin absorption band reached a
minimum value (0.19) leading a dramatic yield decrease of 96% (entry 7). This result makes it
clear that the pronounced photodegradation effect becomes inevitable with the employed of highꢀ
power and multifrequential light sources on the final stage of hypericin synthesis, even for a
controlled photoconversion reaction time.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. UVꢀvisible kinetic monitoring of photoconversion process of the initial protohypericin
solution in acetone (1.6 x 10ꢀ5 mol.Lꢀ1). (A) Irradiation with blue LED system (10.2 mW/cm2) for
a reaction time of 800 min. The decreased in hypericin yield of only 11% show a photobleaching
ACS Paragon Plus Environment
12

Page 13 of 24
Organic Process Research & Development
1
2
3
4
5
6
effect practically negligible; (B) Irradiation with mercuryꢀvapor lamp (400 W). The complete
hypericin formation was accomplished at 7 min of continuous irradiation with a significant
reduced yield of 79%; (C) Exceeding the irradiation time, the hypericin absorption band at λ596
show a rapidly decrease and reach a minimum value after 35 min of UV exposed.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To further demonstrate the relevance of this present design, and in according with previous
microꢀscale studies, a gramꢀscale supply experiments was performed in order to evaluated the
reproducibility to two lowꢀpower light sources in the preparation of the 5.0 g hypericin. To this
end, we employed one photoꢀreactive modular unit composed of 36 SMD red or green LEDs,
and the photoconversion process were monitored by realꢀtime UVꢀvisible spectroscopy analysis
(Figures S5 and S6). The data showed a highꢀefficiency photoconversion to both LEDs systems,
and with reproducibility of the optimized experimental conditions set in microꢀscale (Table 2).
The redꢀLED photoreactor was employed to photoconverted quantitatively an initial 3.3 x 10ꢀ3
mol.Lꢀ1 protohypericin concentration to hypericin, after 8 cycles of continuous irradiation with a
4.0 mW/cm2 light dose. However, only 3 cycles of continuous irradiation with the greenꢀLED
photoreactor of 11.3 mW/cm2 light dose, were necessary to the complete photoconversion of the
same protohypericin concentration, yielding 98% of hypericin, i.e. an irradiation time of
approximately 2.7 times lower than that obtained for the redꢀLED system. The high spectral
overlap of the green LED whit the main hypericin absorption band (λ596) lead to a possible
photobleaching effect, which results in this small difference in the hypericin yield observed
between the two LEDs systems. Consequently, on one hand the employ of the greenꢀLED
photoreactor in the production of hypericin in industrial scale could resulted in a significative
reduction of the energy consumption, but on the other, possible yield losses inherent to the
photodegradation process would inevitably lead to an increasing of final cost. From this point of
view independently of the LED light system used in the production of hypericin, the significant
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 24
1
2
3
4
5
6
7
8
9
benefits of the present design in comparison with the usual mercuryꢀvapor lamps, involve mainly
the higherꢀefficient in the protohypericinꢀhypericin photoconversion process. Additionally, the
optimization of the photonic parameters demonstrated here allows to achieved a safe and
reproductive scaleꢀup protocol for the hypericin synthesis with a lower power consumption, and
higher yield. Finally, although in this work we demonstrated the high efficiency of the LED
based photoreactor for a gramꢀscale supply of hypericin, the innovative flow continuously
system designed for this prototype allow that the hypericin can be prepared in a scale even higher
overcoming the limitations related to open irradiation systems previously reported, which require
a wide contact surface between the protohypericin solution and the irradiation light source,
limiting the largeꢀscale production of hypericin.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Experimental conditions obtained to the protohypericin photoconversion process under
red and greenꢀLED photoreactors.
Light dose
Number of
cycles
Reaction time
(min)
Isolated
Comments
LED
photoreactor
Wavelength
(nm)
(mW/cm2)
yields (%)
Low spectral overlap
with hypericin
Red
629
504
4.0
8
3
13,6
5,1
Quantitative
High spectral overlap
with hypericin:
Possible
Green
11.3
98
photobleaching
CONCLUSIONS
This work presents an improved synthesis process developed for a multigram production of
hypericin. The main problem presented in such process resides in a low yield and purity degree
hypericin production implying to high commercial cost of the final product. Therefore, in order
to optimize the overall hypericin process, especially concerning with its critical final stage
related with the protohypericin photocyclization reaction, an inexpensive and efficient low power
ACS Paragon Plus Environment
14

Page 15 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Light Emission Diode (LED) based photoreactor was also designed. Taking into account the
suitable photonic parameters like the wavelength emission and power LED light source, light
dose, time of irradiation, proꢀdrug concentration, and the spectral overlap overcoming, the
undesirable intrinsic hypericin photobleaching effect, and hypericin multigram synthesis can be
achieving in excellent yields nearby 98% for all kind of LEDs, dependent on all of these
photonic parameters. Finally, this closed system also overcomes safety issues related to scaleꢀup
hypericin preparation typically described in the literature which combine the use of open
systems, organic solvents and highꢀpower light sources. Furthermore, this system can be applied
to other photoreaction systems.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Materials. Reactants and reagents were purchased from commercial sources and used
without further purification, except for emodin, which was isolated from Cortex Frangula.34
Experimental Methods. Nuclear magnetic resonance spectra were recorded on a Bruker
AVANCE III HD 500 MHz spectrometer operating respectively at 500,133 and 125,03 MHz,
equipped with direct field gradient and probe of 5 mm, and at a constant temperature of 25o C.
All samples were prepared with an addition of 0.7 mL of deuterated solvents in 5ꢀ10 mg of
1
13
compounds. H and C NMR data are recorded in DMSOꢀd6, and CD3CN listed as residual
internal: DMSOꢀd5 (δ 2.50 and 40.0), CD2HCN (δ 1.39) and the spectra were referenced as the
TMS peak. Data are presented as follows: chemical shift (ppm), multiplicity (s = singlet, d =
doublet, dd = doublet of doublet, dq = doublet of quartet), and integration. All spectra were
processed using the Bruker TopSpin 3.1 software. High resolution mass spectroscopy spectrum
was recorded by direct infusion ESIꢀMS on a hybrid quadrupole orthogonal timeꢀofꢀflight mass
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 24
1
2
3
4
5
6
7
8
9
spectrometer, QTof HRMS (Waters Corporation, Milford, MA, USA), controlled by MassLynx
4.1 software. The electrospray ionization was performed in the negative ion mode (ESIꢀ) with
capillary voltage set to 2.5 kV, cone voltage to 35 V and source temperature to 150 ºC. The
desolvation gas was nitrogen, with flow of 700 L hꢀ1 and temperature of 400 °C. Data was
acquired from m/z 50 to 1000 in MS mode and MS2 experiment was acquired from m/z 50 to 600,
in which the collision energy alternated between low energy (2 V) and high energy (45 V).
Leucineꢀenkephalin (Waters Corporation, USA), C28H35N5O7 ([MꢀH]ꢀ of m/z 554.2771) was used
as reference at a concentration of 0.2 ng Lꢀ1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Electronic absorption spectroscopy analysis (UVꢀvisible) were recorded on a Beckman
Coulter DU800 UVꢀvisible Spectrophotometer at room temperature. Protohypericin
photocyclization kinetics analysis were measured through UVꢀvisible analysis. Aliquots of
protohypericin in acetone (1.6 x 10ꢀ5 mol.Lꢀ1) were added in quartz cuvettes with a 1 cm optical
pathway. The cuvettes were fixed in a support adapted with six lateral LEDs (three in each
lateral), arranged at 0.5 cm of the solution.33 The experiments were performed with four groups
of lowꢀpower 1210ꢀSMD LEDs sources: Blue, green, orange, and red (0.6 – 10.2 mW/cm2). The
absolute irradiance of LEDs was measured using an Ocean Optics spectrometer, model
USB4000ꢀUVꢀVIS.
Procedure for the synthesis of emodin anthrone. For a suspension of emodin (10.0 g, 37.0
mmol), in acetic acid (350 mL) kept in a roundꢀbottom flask, a solution of SnCl2.2H2O (52.3 g,
o
0.22 mmol) in HClconc (365 mL, 370 mmol) was added and the solution was heated to 120 C for
30 min under vigorous stirring. The solution was cooled, poured in water (2 L) and neutralized
with Na2CO3. The precipitate was filtered off, washed with distilled water and drying under
1
o
vacuum to yield 8.7 g (87%) of yellow/brown solid. C15H12O4, H NMR (DMSOꢀd6, 25 C, 500
ACS Paragon Plus Environment
16

Page 17 of 24
Organic Process Research & Development
1
2
3
4
5
6
MHz): δH 12.37 (s, 1H, OHꢀ1), 12.21 (s, 1H, OHꢀ8), 10.84 (s, 1H, OHꢀ3), 6,77 (m, 1H, ArꢀH4),
6.67 (m, 1H, ArꢀH5), 6.42 (dq, 1H, ArꢀH2), 6.23 (d, 1H, ArꢀH7), 4.30 (s, 2H, BenzꢀH10), 2.32
7
8
9
13
o
(dd, 3H, ArꢀCH3) ppm; C NMR (DMSOꢀd6, 25 C, 75 MHz): δC 191.35 (ꢀC=O), 165.24 (C6),
164.81 (C8), 161.91 (C1), 147.32 (C11), 145.23 (C14), 142.25 (C13), 120.15 (C4), 115.39 (C3),
113.08 (C12), 108.65 (C2), 107.64 (C5), 101.23 (C7), 32.51 (C9), 21.84 (ꢀCH3); UVꢀVis (EtOH)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
λ
max 250, 258, 271, 300 e 353 nm. The characterization data of emodin anthrone are in agreement
with the literature.35
Procedure for the synthesis of protohypericin. To a solution of emodin anthrone (8.00 g,
31.2 mmol) in 176 mL of pyridine/piperidine (10:1 v/v), were added 16.0 g (168 mmol) of
pyridine Nꢀoxide and 0.4 g of ferrous sulphate heptahydrate (1.44 mmol) and the mixture was
o
heated in the dark at 100 C for 1 h under vigorous stirring. After cooling, the solvent was
concentrated to approximately 5.0 mL and the solution was neutralized with 3% HCl solution.
The precipitate was filtered off, washed with water and then dried under vacuum for 24 h. Crude
product was purified by sizeꢀexclusion column chromatography (SEC) using Sephadex® LHꢀ20
1
matrix and EtOH as mobile phase to yield 5.60 g (70%) purple crystals. C30H18O8; H NMR
(DMSOꢀd6, 25 oC, 500 MHz): δH 14.40 (s, 2H, OHꢀ1, OHꢀ6), 12.89 (s, 2H, OHꢀ7, OHꢀ14), 7.24
(s, 2H, ArꢀH8, ArꢀH13), 6.78 (s, 2H, ArꢀH10, ArꢀH11), 6.33 (s, 2H, ArꢀH2, ArꢀH5), 2.09 (s, 6H,
ArꢀCH3) ppm; UVꢀVis (EtOH) λmax 369, 524 e 596 nm.
Procedure for the synthesis of hypericin. Protohypericin solution (5.00 g, 9.85 mmol) in
acetone (3 L) was charged into 5 L feed amber flask and pumped into the flow cell by down of
the photoreactor at a volumetric flow rate (1.25 x 10ꢀ4 m3/s). The solution was irradiated with
four photoꢀreactive modular units vertically arranged in series, each one composed by 36 x red
1210ꢀSMD (Surface Mounted Diode) LEDs of light dose 12.5 mW/cm2 (Honglitronic). The rate
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 24
1
2
3
4
5
6
7
8
photoconversion was monitored until the hypericin absorption band at λ596 nm reached a
maximum of intensity, correspondent with a quantitative photoconversion of protohypericin to
hypericin. The crude product was recrystallized from a mixture of hexane : MeOH (95:5 v/v) to
9
1
o
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
afford an blueꢀdark crystalline solid. Yield: 4.85 g, 97%. C30H16O8; H NMR (CD3CN, 25 C,
500 MHz): δH 14.75 (s, 2H, OHꢀ1, OHꢀ6), 14.13 (s, 2H, OHꢀ7, OHꢀ12), 7.31 (s, 2H, ArꢀH8, Arꢀ
H11), 6.55 (s, 2H, ArꢀH2, ArꢀH5), 2.71 (s, 6H, ArꢀCH3) ppm; UVꢀVis (EtOH) λmax 392, 480,
o
513, 552, 596. 13C NMR (CD3CN, 25 C, 500 MHz): δC 185.2, 175.7, 169.8, 162.8, 144.6,
128.4, 127.6, 122.7, 122.1, 120.5, 119.6, 109.9, 106.6, 103.5. The identification of molecular
formula of Hypericin (C30H16O8) was confirmed on its negative HRMS ion of m/z 503,0854
ꢀ
2
₋
(calcd for [M H] 503,0845). The MS experiment showed fragments ions characteristic for this
ꢀ
compound of m/z 487.0202 [M H CH4] , 458.0504 [M H CO2 H·]ꢀ, 433.0361
₋
₋
₋
₋
₋
ꢀ
₋
₋
₋
₋
₋
₋
₋
[M H CH2=C=O CO] ,
431.0601
[M H CO2 CO]
and
405.0425
ꢀ
₋
₋
₋
[M H CH2=C=O 2CO] . The characterization data of hypericin are in agreement with the
literature.35ꢀ39
ASSOCIATED CONTENT
AUTHOR INFORMATION
Corresponding Author
*email: rsgoncalves2@uem.br
Notes
The authors declare no competing financial interest.
Funding Sources
ACS Paragon Plus Environment
18

Page 19 of 24
Organic Process Research & Development
1
2
3
4
5
6
Fundação Araucária/PR, CNPQ (p 308142/2014 4), CAPES
Supporting Information Available.
7
8
9
Experimental procedures, characterization data and copies of spectra are available free for charge
via the Internet at
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
REFERENCES
1. Karioti, A.; Bilia, A. R. Int. J. Mol. Sci. 2010, 11, 562ꢀ594.
2. Joniova, J.; Rebic, M.; Strejckova, A.; Huntosova, V.; Stanicova, J.; Jancura, D.; Miskovsky,
P.; Bano, G. Biophys. J. 2017, 112, 966ꢀ975.
3. Do, M. H.; Kim, S. Y. Biomol. Ther. 2017, 25, 158ꢀ164.
4. Huntosova, V.; Novotova, M.; Nichtova, Z.; Balogova, L.; Maslanakova, M.; Petrovajova, D.;
Stroffekova, K. Toxicol. In Vitro 2017, 40, 184ꢀ195.
5. Brockmann, H. Angew. Chem. 1949, 61, 389ꢀ389.
6. Brockmann, H. Chem. Ber./Recl. 1949, 82, 348ꢀ357.
7. Brockmann, H.; Kluge, F. Naturwissenschaften 1951, 38, 141ꢀ141.
8. Brockmann, H.; Neeff, R. Naturwissenschaften 1951, 38, 47ꢀ47.
9. Brockmann, H.; Sanne, W. Naturwissenschaften 1953, 40, 461ꢀ461.
10. Brockmann, H.; Eggers, H. Angew. Chem., Int. Ed. Engl. 1955, 67, 706.
11. Brockmann, H.; Eggers, H. Angew. Chem., Int. Ed. 1955, 67, 706ꢀ706.
12. Brockmann, H.; Kluge, F.; Muxfeldt, H. Chem. Ber./Recl. 1957, 90, 2302ꢀ2318.
13. Brockmann, H.; Eggers, H. Chem. Ber./Recl. 1958, 91, 81ꢀ100.
14. Steglich, W.; Arnold, R. Angew. Chem., Int. Ed. Engl. 1973, 12, 79ꢀ79.
15. Banks, H. J.; Cameron, D. W.; Raverty, W. D. Aust. J. Chem. 1976, 29, 1509ꢀ1521.
16. Rodewald, G.; Arnold, R.; Griesler, J.; Steglich, W. Angew. Chem., Int. Ed. Engl. 1977, 16,
46ꢀ47.
17. Mazur Y.; Bock H.; Lavie D. CA 2 029 993. 1989.
18. Falk, H.; Schoppel, G. Monatsh. Chem. 1992, 123, 931ꢀ938.
19. Falk, H.; Meyer, J.; Oberreiter, M. Monatsh. Chem. 1993, 124, 339ꢀ341.
20. Kraus, G. A.; Zhang, W. Bioorg. Med. Chem. Lett. 1995, 5, 2633ꢀ2636.
21. Falk, H.; Vaisburg, A. F.; Amer, A. M. Monatsh. Chem.1995, 126, 993ꢀ1000.
22. Falk, H.; Sarhan, A.; Tran, H. T. N.; Altmann, R. Monatsh. Chem. 1998, 129, 309ꢀ318.
23. Altmann, R.; Falk, H.; Gruber, H. J. Monatsh. Chem. 1998, 129, 235ꢀ244.
24. Kowalik, E. G.; Zalkow, L. H. Org. Prep. Prep. Proced. Int. 2000, 32, 57ꢀ61.
25. Kim, S. W.; Park, J. H.; Yang, S. D.; Hur, M. G.; Kim, Y.; Chai, J.; Kim, Y. S.; Yu, K. H.
Bull. Korean Chem. Soc. 2004, 25, 1147ꢀ1150.
26. Motoyoshiya, J.; Masue, Y.; Nishi, Y.; Aoyama, H. Nat. Prod. Comm. 2007, 2, 67ꢀ70.
27. Falk, H. Monatch. Chem. 2008, 139, 991ꢀ993.
28. Tobia, A. J.; Cabana, B. E.; Vadlapatla, V.; Connolly, R. H. US20120245392A1. 2012.
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 24
1
2
3
4
5
6
29. Fang, H.; Hui, W.; Lin, C. Chin. J. Nat. Med. 2014, 12, 81ꢀ88.
30. Serra, F.; Terentjev, E. M. J. Chem. Phys. 2008, 128, 6.
31. Uzdensky, A. B.; Iani, V.; Ma, L. W.; Moan, J. J. Photochem. Photobiol., A 2002, 76, 320ꢀ
328.
7
32. Poutaraud, A.; Di Gregorio, F.; Tin, V.C.F.; Girardin, P. Planta Med. 2001, 67, 254ꢀ259.
33. Rabello, B. R.; Gerola, A. P.; Pellosi, D. S.; Tessaro, A. L.; Aparicio, J. L.; Caetano, W.;
Hioka, N. J. Photochem. Photobiol., A 2012, 238, 53ꢀ62.
34. Gonçalves, R.S; Silva, L.E.; Hioka, N.; Nakamura, C.V.; Bruschi, M.L.; Caetano, W. Nat.
Prod. Res. 2017, 1ꢀ4.
35. Kapinus, E. I.; Falk, H.; Tran, H. T. N. Monatch. Chem. 1999, 130, 623ꢀ635.
36. Piperopoulos, G.; Lotz, R.; Wixforth, A.; Schmierer, T.; Zeller, KꢀP. J. Chromatogr. B. 1997,
695, 309ꢀ316.
37. Brolis, M.; Gabetta, B.; Fuzzati, N.; Pace, R.; Panzeri, F.; Peterlongo, F. J. Chromatogr. A.
1998, 825, 9ꢀ16.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
38. Riedel, KꢀD.; Rieger, K.; MartinꢀFacklam, M.; Mikus, G.; Haefeli, W. E.; Burhenne, J. J.
Chromatogr. B. 2004, 813, 27ꢀ33.
39. Gadzovska, S.; Maury, S.; Ounnar, S.; Righezza, M.; Kascakova, S.; Refregiers, M.;
Spasenoski, M.; Joseph, C.; Hagège, D. Plant Physiol. Biochem. 2005, 43, 591ꢀ601.
ACS Paragon Plus Environment
20

Page 21 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TOC graphic
109x112mm (600 x 600 DPI)
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Optimized experimental conditions for the hypericin synthesis.
63x56mm (600 x 600 DPI)
ACS Paragon Plus Environment

Page 23 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. (A) Process flow diagram of the protohypericin photoconversion. (B) Showing the photoreactor
vertically arranged with four photo-reactive modular units interlocked. (C) Showing the deconstructed
modular unit. The flow cell is inserted between two glass slabs preventing the intimate contact of the sample
with the irradiation systems.
71x32mm (600 x 600 DPI)
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. UVꢀvisible kinetic monitoring of photoconversion process of the initially protohypericin solution in
acetone (1.6 x 10ꢀ5 mol.Lꢀ1). (A) Irradiation with blue LED system (10.2 mW/cm2) for a reaction time of 800
min. The decreased in hypericin yield of only 11 % show a photobleaching effect practically negligible; (B)
Irradiation with mercuryꢀvapor lamp (400 W). The complete hypericin formation was accomplished at 7 min
of continuous irradiation with a significant reduced yield of 79%; (C) Exceeding the irradiation time, the
hypericin absorption band at λ596 show a rapidly decrease and reach a minimum value after 35 min of UV
exposed.
89x92mm (600 x 600 DPI)
ACS Paragon Plus Environment