Bioactive products from singlet oxygen photooxygenation of cannabinoids

Article abstract of DOI:10.1016/j.ejmech.2017.11.043

Photooxygenation of Δ⁸ tetrahydrocannabinol (Δ⁸-THC), Δ⁹ tetrahydrocannabinol (Δ⁹-THC), Δ⁹ tetrahydrocannabinolic acid (Δ⁹-THCA) and some derivatives (acetate, tosylate and methyl ether) yi

Full text of DOI:10.1016/j.ejmech.2017.11.043

Accepted Manuscript
Bioactive products from singlet oxygen photooxygenation of cannabinoids
Ahmed Galal, Khaled M. Elokely, Vivek K. Yadav, Paulo Carvalho, Mohamed
Radwan, Desmond Slade, Waseem Gul, Shabana Khan, Olivia R. Dale, Afeef S.
Husni, Michael L. Klein, Stephen J. Cutler, Samir A. Ross, Mahmoud A. ElSohly
PII:
S0223-5234(17)30946-7
10.1016/j.ejmech.2017.11.043
EJMECH 9918
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 5 July 2017
Revised Date: 14 October 2017
Accepted Date: 17 November 2017
Please cite this article as: A. Galal, K.M. Elokely, V.K. Yadav, P. Carvalho, M. Radwan, D. Slade, W.
Gul, S. Khan, O.R. Dale, A.S. Husni, M.L. Klein, S.J. Cutler, S.A. Ross, M.A. ElSohly, Bioactive products
from singlet oxygen photooxygenation of cannabinoids, European Journal of Medicinal Chemistry
(2017), doi: 10.1016/j.ejmech.2017.11.043.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Bioactive products from singlet oxygen
photooxygenation of cannabinoids
Ahmed Galal╬*, Khaled M. Elokelyʘ,◊,┴, Vivek K. Yadavʘ,◊, Paulo Carvalho║, Mohamed
Radwan╬, Desmond Slade&, Waseem Gul╬,§, Shabana Khan╬#, Olivia R. Dale╬, Afeef S. Husni#,
Michael L. Kleinʘ,◊, Stephen J. Cutler#†, Samir A. Ross╬,#, Mahmoud A. ElSohly,§, Ω.*
╬ National Center for Natural Products Research, the University of Mississippi, University, MS
38677
ʘ Institute for Computational Molecular Science, Temple University, Philadelphia, PA 19122
◊ Department of Chemistry, Temple University, Philadelphia, PA 19122
┴ Department of Pharmaceutical Chemistry, Tanta University, Tanta 31527, Egypt
║ Feik School of Pharmacy, University of the Incarnate Word, San Antonio, TX, 78209
& MRI Global, 425 Volker Boulevard, Kansas City, Missouri
§ ElSohly Laboratories, Inc., 5 Industrial Park Drive, Oxford, MS 38655
# Department of of BioMolecular Sciences, University of Mississippi, University, MS 38677
Ω Department of Pharmaceutics, University of Mississippi, University, MS 38677
1

ACCEPTED MANUSCRIPT
†Current affiliation: University of South Carolina College of Pharmacy, Columbia, SC 29208
* Corresponding authors
KEYWORDS Cannabinoids, Photooxygenation, CB1 agonists, Antimicrobial, Anticancer,
Antileishmanial.
ABSTRACT: Photooxygenation of ∆8 tetrahydrocannabinol (∆8-THC), ∆9 tetrahydrocannabinol
(∆9-THC), ∆9 tetrahydrocannabinolic acid (∆9-THCA) and some derivatives (acetate, tosylate
and methyl ether) yielded 24 oxygenated derivatives, 18 of which were new and 6 were
previously reported, including allyl alcohols, ethers, quinones, hydroperoxides, and epoxides.
Testing these compounds for their modulatory effect on cannabinoid receptors CB1 and CB2 led
to the identification of 7 and 21 as CB1 partial agonists with Ki values of 0.043 µM and 0.048
µM, respectively and 23 as a cannabinoid with high binding affinity for CB2 with Ki value of
0.0095µM, but much less affinity towards CB1 (Ki 0.467µM). The synthesized compounds
showed cytotoxic activity against cancer cell lines (SK-MEL, KB, BT-549, and SK-OV-3) with
IC50 values ranging from 4.2 to 8.5 µg/mL. Several of those compounds showed antimicrobial,
antimalarial and antileishmanial activities, with compound 14 being the most potent against
various pathogens.
2

ACCEPTED MANUSCRIPT
Introduction:
Cannabinoids, from Cannabis sativa L., have been the focus of extensive chemical and
biological research due to their unique behavioral, psychotropic and other pharmacological
effects. The discovery that some of their biological activity could be translated into treatments
for a number of serious illnesses, such as glaucoma, depression, neuralgia, multiple sclerosis,
Alzheimer’s disease, alleviation of symptoms of HIV/AIDS and cancer [1-4] have given
momentum for further exploration of their chemical and biological properties. The discovery of
cannabinoid receptors CB1 and CB2 (with other possible receptors currently under investigation)
[5, 6] opened new possibilities for the design and exploration of cannabinoid structures. CB1
agonists exhibit analgesic properties, whereas CB1 antagonists and inverse agonists have shown
the potential to act as therapeutic agents against diabetes, drug dependence, and obesity. CB2
agonists have exhibited cytotoxicity and demonstrated potential for treatment of neuropathic
pain [7-9], suppression of inflammation [10] and attenuation of the severity of disease in animal
models of multiple sclerosis [11] and age-related illnesses [12-14].
In search of compounds with affinity for CB1 and CB2 cannabinoid receptors our group
decided to explore oxygenated derivatives of ∆9-THC and its isomer ∆8-THC. One of the earliest
references reporting the photooxygenation of ∆8-THC acetate via irradiation with UV light in
the presence of oxygen and using rose bengal as a photosensitizer [15] yielded three
hydroperoxides: (-)-8α- and (-)-8β-hydroperoxido-∆9,11-THC acetate, and (-)-9α-hydroperoxido-
∆7,8-THC acetate. More recently, other oxygenated derivatives of ∆9-THC and ∆8-THC have
been prepared, showing antibacterial [16] and anticancer effects [17-20], as well as
demonstrating some degree of affinity to cannabinoid receptors CB1 and CB2 [21, 22].
3

ACCEPTED MANUSCRIPT
This article is a continuation of our work [23] and describes the photooxygenation of ∆9-
tetrahydrocannabinol (∆9-THC - 1), ∆8-tetrahydrocannabinol (∆8-THC - 2), ∆9-
tetrahydrocannabinolic acid (∆9-THCA - 8) and some of their derivatives (Figure 1) under
different conditions. Meso-tetraphenylporphine was used as a photosensitizer in presence of
oxygen and irradiation with incandescent light, generating singlet oxygen (1O2*), which reacted
with the trisubstituted olefinic moiety to form oxygenated products. Six of those compounds (15,
23, 10, 12, 14 and 31) have been previously reported as minor oxygenated cannabinoids from
cannabis, serum metabolites of ∆9-THC, or products from non-photooxygenation reactions [15,
24-29].
These compounds were screened for various biological activities, including antimicrobial
(Staphylococcus aureus, methicillin-resistant Staphylococcus aureus [MRSA]), antifungal
(Cryptococcus neoformans, Candida glabrata, and Candida krusei), anticancer (cell lines SK-
MEL, KB, BT-549, and SK-OV-3), antimalarial (Plasmodium falciparum, D6 clone -
chloroquine-sensitive - and W2 clone - chloroquine-resistant) and antileishmanial (Leishmania
major), as well as their binding affinity towards cannabinoid receptors CB1 and CB2.
4

ACCEPTED MANUSCRIPT
Figure 1: Structures of ∆9-THC (1), ∆8-THC (2), ∆9-THC tosylate (3), ∆8-THC tosylate (4), ∆8-
THC methoxy (5), ∆9-THC acetate (6), ∆8-THC acetate (7), and ∆9-THCA (8).
Results and discussion
Chemistry
Photooxygenation of 1, 2, tosylates 3 and 4 [25], methoxy-∆8-THC 5, ∆9-THC acetate 6, ∆8-
THC acetate 7, and ∆9-THCA 8 using meso-tetraphenylporphine as a photosensitizer [30]
resulted in the formation of 24 derivatives. Six of them (15, 23, 10, 12, 14 and 31) have been
previously reported [15, 24-29] and the other 18 are, to the best of our knowledge, novel
compounds. One of our goals was to generate a large array of oxygenated derivatives in order to
correlate the position and nature of those functionalities with their biological activity. Initial
studies revealed that changes in polarity of the reactional solvent system led to the formation of
products with different patterns of oxygenation, and this knowledge was used to guide the choice
of solvents with variable polarities for our reactions. Reaction conditions and corresponding
products are summarized on Table 1.
Table 1. Reaction time, solvent systems and products of photooxygenation of THC derivatives.
Starting material Solvent system
Reaction time
11 h
8 h and 30 min
4 h and 30 min
4 h and 15 min
11 h and 30 min
Product(s)
9-12
13
14
15, 16, 17, and 18
17, 18, 19* and
Scheme
CH2Cl2/EtOH
1
2
2
3
3
hexanes/CH2Cl2 (4:1)
CH2Cl2/propanol (1:1)
CH2Cl2
1
2
CH2Cl2/EtOH
20
CH2Cl2/EtOH
CH2Cl2/EtOH
hexanes/ CH2Cl2
hexanes/ CH2Cl2
4 h and 15 min
6 h
4 h
8 h and 15 min. followed
by reduction with NaBH4
3 h
21, 22 and 23**
3
4
5
4
5
6
7
23-25
26-30
27
CH2Cl2/MeOH (1:1)
8
28
*product of reduction of 18; ** product of hydrolysis of 21
5

ACCEPTED MANUSCRIPT
Some of the products were subjected to further treatment under different conditions.
Compound 18, submitted to reduction with dimethyl sulfide for 22 hours, yielded the allylic
alcohol 19 (Scheme 2); epoxide 27, reduced with NaBH4 or NaHCO3/H2O/Adogen 464, afforded
compound 29, while reduction with Pd/C yielded compound 30 (Scheme 5). Attempted reduction
of compounds 21, 22 and hydroperoxides 24, 25, resulted in decomposition. Alkaline hydrolysis
of 21 yielded the known tertiary allylic alcohol 23 (Scheme 3), while attempted hydrolysis of 22
resulted also in decomposition.
6

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of compounds 9-14.
Reagents and conditions: (a) meso-tetraphenylporphine, O2, light, (b) CH2Cl2 /anhydrous EtOH
(1:2), 11 h; (c) hexanes/CH2Cl2 (4:1), 8 h and 30 min; (d) CH2Cl2 /n-propanol (1:1), 4 h and 30
min.
7

ACCEPTED MANUSCRIPT
Scheme 2. Synthesis of compounds 15-20.
Reagents and conditions: (a) meso-tetraphenylporphine, O2, light, (b) CH2Cl2, 4 h and 15 min;
(c) Me2S/22 h; (d) CH2Cl2/abs EtOH (1:1), 11 h and 30min.
X-ray analysis of compound 20 [31] (CCDC reference: 1442416), crystallized from ethyl
acetate: hexanes 1:9 producing needle-like crystals, allowed for confirmation of structure and the
establishment of its relative configuration (Figure 2).
8

ACCEPTED MANUSCRIPT
Figure 2. Plot of the molecular structure of compound 20, Displacement ellipsoids are drawn at
the 50% probability level
9

ACCEPTED MANUSCRIPT
Scheme 3. Synthesis of compounds 21-23.
OH
OOH
OH
OTs
c
11
9
O
C5H11
O
C5H11
8
OTs
10
23
21
1
10a
10b
7
a, b
6a
2
3
6
12
HOO
5
O
C5H11
OTs
4
13
4
O
C5H11
22
Reagents and conditions: (a) meso-tetraphenylporphine, O2, light, (b) CH2Cl2/anhydrous EtOH
(1:1), 4 h; (c) hydrolysis, KOH/EtOH, 75 min.
10

ACCEPTED MANUSCRIPT
Scheme 4. Synthesis of compounds 32, 24-26.
Reagents and conditions: (a) meso-tetraphenylporphine, O2, light, (b) CH2Cl2/abs EtOH (1:1), 6
h, (c) CH2Cl2/MeOH (1:1), 3 h.
Compounds 25 and 26 (Scheme 4) were obtained as separated compounds and exhibited
different chromatographic behavior on TLC, Rf = 0.42 (Hexanes- DCM-MeOH, 9:9:0.8) and Rf
0.36 (Hexanes - DCM-MeOH, 9:9:0.8) respectively. They also display different specific rotation
values and different 1H and 13C NMR shifts. However, none of the spectroscopic methods used
was capable of assigning with certainty the configuration of each isomer and we were unable to
obtain a crystalline sample for X-ray analysis.
=
11

ACCEPTED MANUSCRIPT
Scheme 5. Synthesis of compounds 27 - 31
Reagents and conditions: (a) meso-tetraphenylporphine, O2, light; (b) hexanes/CH2Cl2 (1:1),
10oC, 3 h, 45 min; (c) hexanes/CH2Cl2 (1:1), 7.7oC, 8 h; (d) NaBH4/MeOH, 6 h; (e)
NaBH4/MeOH, 2 h, 30 min; (f) NaHCO3/H2O/MeOH, Adogen 464, (g) Pd/C, H2, MeOH, 10 h.
Stereochemical assignments
Stereochemical assignments for derivatives 9, 13, 17, 18, 19, 21, 24 were determined on the
basis of NOESY correlations, as seen on Figure 3 and Table 2. The assignments for compound
23 were confirmed by comparison with published NMR data [15, 32].
12

ACCEPTED MANUSCRIPT
The orientation of the hydroxyl and ethoxyl functionalities at C-9 and C-10 of compound 11
were determined by comparison with the 13C values of the two diastereomers previously reported
[33].
OOH
OOH
H OTs
OH
H
OOH
H
H
OTs
OTs
OTs
10
H
H
10a
H
H
O
C5H11
O
C5H11
O
C5H11
O
C5H11
9
8
11
7
OEt
OOH
HO
H
O
O
O
O
O
O
H
H
O
C5H11
O
C5H11
O
C5H11
14
23
18
Figure 3. Relative configuration of compounds 9, 13, 17, 18, 19, 21, and 24, based on NOESY
correlations.
13

ACCEPTED MANUSCRIPT
Table 2. Stereochemical orientation of derivatives 9, 13, 17, 18, 19, 21, and 24., as determined
on the basis of NOESY correlations.
Compound Group analyzed
Orientation
Correlation
OEt at C-9
α
α
-oriented
-oriented
9
between Me-11 (δ 1.41) at C-9 and Me-13 (δ 1.04 -
previously established as
between H-8 (δ 4.14) and the
between H-10 (δ 4.63) and the
between H-10 (δ 4.14) and the β-oriented H-6a (δ 1.79)
between H-10 ( 4.06) and the β-oriented H-6a ( 1.74).
9.28) at C-9 and the α-
β
-oriented)
-oriented Me-12 (δ 1.52).
oriented H-10a (δ 2.07)
OH at C-8
13
17
18
19
21
α
HOO at C-10
HOO at C-10
OH at C-10
HOO at C-9
β-oriented
α
α
α
α
-oriented
-oriented
-oriented
δ
δ
between the proton of HOO (
oriented H-10 ( 3.52)
δ
δ
H-10
α
α
-oriented
-oriented
with H-10a (δ 2.85)
HOO at C-9
24
between Me-11 (δ 1.55 at C-9) and the
(δ 2.77)
β-oriented H-6a
2.2. Biological activity
2.2.1. Affinity to cannabinoid receptors
The control used in both binding and functional assays was the non-traditional cannabinoid,
CP 55,940[34]. The binding Ki for CP 55,940 at CB1 is 0.5 to 5nM, and the binding Ki for CP
55,940 at CB2 is 0.69 to 2.8nM. The functional Ki for agonism for this control at the CB1
receptor is 0.07 to 4nM, and 0.2 to 7.4nM for CB2.
Derivatives 13, 18, 19, 21, 27 and 29 have shown affinity for cannabinoid receptors in the low
micromolar and nanomolar range.
14

ACCEPTED MANUSCRIPT
Emax of compound 21 at 100
µM was 20% stimulation for CB1 and 90% for CB2. Compound
29, also at 100 M, exhibited an Emax of 20% stimulation for CB1 and 40% for CB2. Table 3
µ
presents the most representative values of binding affinity to CB1 and CB2 and Table 4 presents
the most representative results from functional assays.
Table 3: Results of CB1 and CB2 binding assays for compounds 7, 14, 18, 19, 21, 22, 23, 27, 29,
and 30. Error was monitored for each concentration point and displayed on the graphics
(supplementary information) with error bars.
Compound
CB1*
0.088/0.044
1.84/0.919
1.024/0.512
1.28/0.642
0.275/0.137
5.70/2.85
CB2*
0.316/0.158
4.07/2.034
7
14
18
0.851/0.426
1.11 (0.552)
0.421/0.211
2.06/1.03
19
21
22
0.93/0.47
0.019/0.0095
0.927/0.464
0.599/0.300
1.98/0.99
23
0.573/0.286
1.077/0.538
2.80/1.40
27
29
30
-/0.0005-0.005
-/0.00069-0.0028
CP 55,940
*Values are expressed as IC50/Ki in µM
15

ACCEPTED MANUSCRIPT
Table 4: Results of CB1 and CB2 functional assays for compounds 7, 21, and 23. The
radioligand used was [35S]-GTP-
ϒS, from Perkin Elmer. Error was monitored for each
concentration point and displayed on the graphics (supplementary information) with error bars.
Compound
CB1*
CB2*
0.087/0.043
0.097/0.048
0.387/0.193
-/0.0007-0.004
0.518/0.259
1.75/0.876
11.42/5.70
0.0002-0.0074
7
21
23
CP 55,940
*Values are expressed as IC50 / Ki in µM
Compounds 18 (C1 tosylate, C10 hydroperoxide), 19 (C1 tosylate, C10 hydroxyl) and 29 (C9
methoxy, C1 and C10 hydroxyl) showed a relatively high affinity for CB2 receptors, with IC50 of
the order of 0.5-0.6 µM whereas their affinity for CB1 receptors was at and slightly above 1 µM.
Addition of hydroxyl groups to C9 seems to improve affinity for both cannabinoid receptors,
but the state of the hydroxyl group at C1 (tosylated or free) seems to make a difference in
selectivity. Compound 21, a C9 hydroperoxide tosylate derivative, displayed good affinity for
the CB1 receptors with IC50 lower than 100 nM, and lower affinity for CB2. Compound 23, a diol
with a hydroxyl group at C9 and free hydroxyl at C1, displayed marked and selective binding
affinity for CB2 receptor with an IC50 lower than 20 nM, and lower affinity for CB1, with an IC50
of the order of 1
by an acetate, had the opposite profile, showing higher affinity for the CB1 receptors with IC50 of
the order of 0.5 M and lower affinity for CB2 receptors, with IC50 of the order of 1 M.
µM. Compound 27, a C9-C10 epoxide with the hydroxyl group at C-1 masked
µ
µ
16

ACCEPTED MANUSCRIPT
Functional assays on cannabinoid receptors
Cannabinoid receptor assays led to identification of derivatives 7, and 21 (Table 4) as CB1
partial agonists, with affinity values in the nanomolar level, and marginal affinity to CB2.
As previously mentioned, Emax of 21 (100
µM) was 20% for CB1 and 90% for CB2 and 29,
also at 100 M, exhibited an Emax of 20% for CB1 and 40% for CB2.
µ
The aforementioned results revealed that, in contrast to a previous study [35] reporting
complete loss of activity when the phenolic hydroxyl group at C-1 of THC is blocked,
photooxygenation of acetate and tosylate derivatives yielded oxygenated derivatives with
masked hydroxyl groups at C-1 which were found to retain affinity towards the cannabinoid
receptors.
1,4-Quinones 24-26 did not exhibit any level of affinity towards CB1 and CB2, presumably
indicating that this functionality may hinder receptor binding due to steric effect.
17

ACCEPTED MANUSCRIPT
2.2.2. Anticancer activity
Quinones 9, 10 [36], 14, and 25 exhibited anticancer activity against cell lines SK-MEL, KB,
BT-549, and SK-OV-3 with IC50 values ranging from 4.2 g/mL (14, against BT-549) to 8.65
g/mL (25, against SK-MEL) (Table 5). It is noteworthy to mention that cannabinoid quinone
µ
µ
derivatives prepared through KOH/EtOH oxidation [17] have been previously reported to
possess antitumor activity, with HU-331 [19] exhibiting its anticancer effect through a novel
mechanism of action as topoisomerase II inhibitor.
Table 5. Anticancer activity of compounds 9, 10, 14, 25 and 26, expressed as IC50 of growth
inhibition (µg/mL).
Cancer Cells
KB
Noncancer Cells
Compound
SK-MEL
BT-549
5.3 + 0.70
6.05 + 0.49
4.2 + 0.28
NA
SK-OV-3
NA
VERO
LLC-PK1
5.4 + 0.42
5.65 + 0.07
2.25 + 0.07
9.9 + 0.14
9.95 + 0.07
9
6.2 + 0.28
7.6 + 0.85
NT
NA
NA
5.95 + 0.78
NT
10
14
25
26
NA
5.25 + 0.35
NA
4.35 + 0.21
NA
4.1 + 0.42
NT
8.65 + 0.49
NA
NA
NA
NA
NT
SK-MEL
KB
BT-549
SK-OV -3
Vero
Human melanoma
Human epidermal carcinoma, oral
Ductal carcinoma, breast
Human ovary carcinoma
Monkey kidney fibroblasts
Pig kidney epithelial cells
LLC-PK1
Values are average of two determinations + std dev
NA = no activity up to 10
NT = not tested
µg/mL
18

ACCEPTED MANUSCRIPT
2.2.3. Antimicrobial activity
Compounds 9, 10, 11, 13, 14, 24-26, and 31 exhibited antimicrobial activity against pathogenic
bacteria Staphylococcus aureus, Methicillin-resistant Staphylococcus aureus (MRSA), and
pathogenic fungi Candida glabrata
,
Candida krusei, and Cryptococcus neoformans (Table 6).
8-THC, was found to be the most active anti-
Compound 14, a quinone derivative of
∆
cryptococcal agent and also the strongest antibacterial agent against MRSA with IC50 of 1.36
g/mL and MIC 2.50 g/mL. Compound 25 was the most potent agent against S. aureus with
IC50 0.91 g/mL and MIC 2.50 g/mL. Compounds 9, 10, 11, 13, 14, and 31 exhibited
considerable activity against C. neoformans S. aureus, and MRSA, without any effect on both
µ
µ
µ
µ
,
Candida species tested. Compound 24, despite not being the most active compound against
bacterial strains, exhibited inhibitory activity against all the organisms tested and was the most
active against both species of Candida
.
19

ACCEPTED MANUSCRIPT
Table 6. Antimicrobial activity of compounds 9, 10, 11, 13, 14, 24, 25, 26 and 31 expressed as
IC50 / MIC (µg/mL)
Compound
Candida
glabrata
Candida
krusei
Cryptococcus Staphylococcus
MRSA
neoformans
aureus
-/-
-/-
-/-
-/-
-/-
-/-
-/-
-/-
-/-
-/-
0.88/-
2.04/-
2.04/-
-/-
9
4.44/20.0
4.57/-
>20/-
4.86/-
10
17.07/-
5.53/10.0
1.36/2.50
2.63/5.0
5.78/10.0
15.3/-
11
1.84/5.00
0.70/2.50
0.93/2.50
1.40/2.50
2.05/5.0
8.34/-
2.03/2.50
1.35/2.50
1.30/2.50
0.91/2.50
2.41/5.0
10.71/-
-/-
13
14
6.54/10.0 5.77/10.0
24
-/-
-/-
-/-
-/-
-/-
17.0/-
20.0/-
-/-
25
26
-/-
31
Amphotericin B
Ciprofloxacin
-/-
1.36/2.50
-/-
-/-
-/-
0.11/0.25
0.12/0.25
20

ACCEPTED MANUSCRIPT
2.2.4. Antimalarial activity
Among the compounds tested against Plasmodium falciparum D6 (chloroquine-sensitive) and
W2 (chloroquine-resistant) strains, compound 14 exhibited the highest activity with IC50 of 0.16
µg/mL for D6 and IC50 of 0.20 µg/mL for W2 (Table 7).
Table 7. Antimalarial activity of compounds 1-10, 17, 19-21, represented as IC50 (µg/mL)
Compound
Chloroquine Artemisinin
9
11
14
20
25
28 30 31
P. falciparam
(D6 strain)
4.76 3.6 0.160 2..2
4.50 3.7 0.20 1.8
1.0 4.50 3.3 2.4
0.90 3.20 3.0 1.7
0.016
0.140
0.013
0.014
P. falciparam
(W2 strain)
D6: chloroquine-sensitive strain
W2: chloroquine-resistant strain
2.2.5. Antileishmanial activity
Compound 14, in addition to its antimicrobial and antimalarial effects, also displayed
pronounced antileishmanial effect against promastigotes of Leishmania donovani with IC50 0.06
µg/mL and IC90 of 0.13 µg/mL (Table 8). Those inhibitory concentrations are almost three times
lower than the standard compound Amphotericin B, placing compound 14 as a good candidate
for further studies of its antileishmanial properties.
21

ACCEPTED MANUSCRIPT
Table 8 Antileishmanial activity of compounds 9, 11, 13, 14, 17, 20, 24, 25, 26 and 29
presented as IC50 and IC90 (µg/mL)
Amphotericin
B
Compound
IC50
Pentamidine
9
11 13 14 17 20 24 25 26 29
2.1 3.1 35
1.0
2.0
0.16
0.5 3.0 0.6 0.06 3.1 4.5 0.7
1.3 0.13
6.5 22 1.2
11
8
>40
0.33
IC90
3.0 6.0
22

ACCEPTED MANUSCRIPT
2.3. Molecular Modeling
The two known subtypes of cannabinoid (CB) receptors CB1 and CB2 share approximately
44% identity throughout the entire protein sequence and roughly 74% of the seven
transmembrane (TM) regions. The structural similarities, principally in the ligand binding cavity,
led to non-selective behaviors of many CB modulators. Experimental crystal structure is not
available for CB2 receptor and homology models were built to be used in the study.
3D models were validated by inspecting dihedral angles, bond length, planarity and other
criteria of structural quality assessment. Molecular docking was performed to investigate the
binding pattern of our compounds. The effects of structural modifications of the phenolic
hydroxyl group at C1, aliphatic chain at C3, and aliphatic hydroxylation at C9 of classical CB
modulators are thoroughly studied showing their importance for CB activity. Several of the
active compounds lack some of these structural elements, and therefore we tried to understand
how these compounds interact with CB receptors.
Compounds 18, 19 and 29 showed better fitting in the active site of CB2 compared to CB1 as
implied by lower docking scores. Compounds 7, 21 and 23 showed docking scores of -7.4, -10
and -9.1 kcal/mol in CB1, and -7, -5.1 and -7.1 kcal/mol in CB2.
The interaction models of compound 21 in CB1 (Figures 4 and 5) demonstrated H-bonds with
Ser383 and His178,
amino acids in the binding pocket.
Compound 7 presented stacking with Phe170 and multiple hydrophobic interactions with
the amino acid residues of CB1, while compound 23 displayed strong H-bonding with Ser285,
stacking with Phe183 and Phe87, and several hydrophobic contacts with the surrounding amino
acids of CB2 (Figures 4 and 5).
π-π stacking with Phe170, and hydrophobic contacts with the surrounding
π-π
π
-
π
23

ACCEPTED MANUSCRIPT
We explored the stability of the docking poses of 7, 21 and 23 with molecular dynamics (MD)
simulations. The protein-ligand interactions were investigated throughout the course of MD
simulations. Protein structures were converged after a short MD period as calculated by the root
mean square deviation (RMSD) of backbone, side chains and heavy atoms (Figure 7), indicating
that the production stage was reached. The RMSD values, over 40 ns, showed a fluctuation
within 1-2 Å after the equilibration period confirming system stability.
24

ACCEPTED MANUSCRIPT
Figure 4. 3D interaction models of compounds 7 (CB1-7) and 21 (CB1-21) with CB1, and
compound 23 (CB2-23) with CB2. The protein is displayed as orange α-helices and green loops.
The ligands are shown as white sticks, and surrounding amino acids as lines.
Figure 5. 2D interaction models of compounds 7 (CB1-7), 21 (CB1-21) and 23 (CB2-23). H-
bonds are shown as purple lines.
π-π stacking is shown as green dashed lines with green spheres
at the ends. Hydrophobic interactions are displayed as solid green lines.
25

ACCEPTED MANUSCRIPT
Compound 21 demonstrated hydrophobic contacts with the surrounding amino acid residues in
the binding pocket of CB1. His178 forms a well-preserved H-bond with the sulfonyl oxygen of
the tosylate group (~56% of the simulation time) and the peroxy group shows intramolecular
hydrogen bond with the same sulfonyl oxygen (~50%). His178, Phe170 and Phe288 make
π-π
stacking (~23%, 44% and 18%, respectively) with the aromatic rings. Blocking the phenolic
hydroxyl group at C1 with toyslate group did not abolish the CB activity of compound 21. The
toyslate group offered favorable molecular region for CB interactions. The sulfonyl group acted
in part as the free phenolic hydroxyl group and formed strong H-bond with His178 and Ser383,
and the tosylate aromatic group formed π-π stacking with His178.
Hydrophobic contacts are very important for compound 7 with CB1. Phe108, Phe170 and
Leu193 display strong hydrophobic interactions with ligand atoms. Compound 23 strongly binds
to CB2 through H-bonds with His95 (~94%) and Ser285 (~99%), and hydrophobic contacts with
Phe87 (~19%) and Phe183 (~73%).
3. Conclusion
Photooxygenation of ∆ ∆ ∆
9-THC, 8-THC, 9- THCA and derivatives resulted in the formation of
24 oxygenated products with diversified functionalities, some of them previously reported as
minor constituents in Cannabis or its metabolites [37]. Change of reaction time and solvent
systems led to the formation of different products. Compounds 7 and 21 were recognized as
selective CB1 partial agonists, demonstrating that blockade of the C-1 hydroxyl function of the
cannabinoid structure does not necessarily abolish affinity towards cannabinoid receptors.
Cannabinoid receptor binding and functional assays also demonstrated that the introduction of
the 1,4-quinone moiety (compounds 9, 10, 14 and 24-26) led to loss of affinity towards
26

ACCEPTED MANUSCRIPT
cannabinoid receptors CB1 and CB2. Those same quinone derivatives, however, were the only
derivatives exhibiting anticancer and marked antimicrobial activity. Quinone 14 was the most
potent anti-cryptococcal and anti-MRSA agent, 25 was the best agent against S. aureus and
quinones 24-26 showed anticandidal activity. 9, 10, 11, 13, 14 and 31 showed antimicrobial
activity against C. neoformans, S. aureus, and MRSA, without any effect on the Candida species
tested.
Compounds 21, 23 and the quinone derivatives 9, 10, 14 and 24-26 bear promising
bioactivities warranting further pursuit focusing on improving yields and increasing selectivity of
the reactions.
4. Experimental protocols
4.1. Chemistry
Starting materials ∆ ∆ ∆
9-THC, 8-THC, and 9-THCA were isolated from Cannabis sativa[38]
grown in the Medicinal Plant Garden at the University of Mississippi, Mississippi, USA and
authenticated by Dr. Suman Chandra[39]. 1D and 2D NMR spectra were recorded in CDCl3 as a
solvent on a Bruker Avance DPX-400 spectrometer and on a Varian AS 400 spectrometer.
HRESIMS was obtained using a Bruker Bioapex FTMS in ESI mode. LRESIMS was obtained
using a 3200 Q Trap LC/MS/MS (Applied Biosystems MDS Sciex, Foster City, CA). TLC was
carried out on aluminum-backed plates precoated with silica gel F254 (20 x 20 cm, 200 µm, 60 Å,
Merck). Visualization was accomplished by spraying with fast blue or p-anisaldehyde [0.5 mL in
glacial acetic acid (50 mL) and H2SO4 (97%, 1 mL)] spray reagent followed by heating. Flash
silica gel (40-63 µm, 60 Å, Silicycle) and SiliaBond C18 silica gel (40-63 µm, 60 Å, 17% carbon
loading, Silicycle) were used for column chromatography.
27

ACCEPTED MANUSCRIPT
General experimental conditions
9-THC, 8-THC were converted to the tosylate [25], acetate [40, 41] esters or methyl ether
9-THCA were also subjected
∆
∆
[42] prior to photooxygenation, In addition, free cannabinoids and
∆
to photooxygenation. For the photooxygenation reactions, meso-tetraphenylporphine (1.0 mg)
was added to the appropriate THC derivative dissolved in a solvent or mixture of solvents. The
reaction mixture was irradiated with 500 W incandescent light for the appropriate amount of
time, with oxygen being gently bubbled into the solution and the temperature of the reaction bath
maintained at 10-13◦C. At the end of the reaction, the solvent was removed and the mixture
purified by column chromatography, unless otherwise specified.
Progress of the reactions was monitored by TLC. Free cannabinoids on the TLC plates were
visualized with fast blue, while tosylate derivatives were detected with p-anisaldehyde/H2SO4.
The identity of these compounds was deduced from spectral analysis including specific rotation,
NMR (1D and 2D), and HRESMS.
Compounds 10, 12, 14, 15, 23 and 31 along with their spectral data have been previously
published [15, 24-26].
(6aR,10aS)-10-ethoxy-9,10a-dihydroxy-6,6,9-trimethyl-3-pentyl-6a,7,8,9,10,10a-
hexahydro-6H-benzo[c]chromen-1-yl 4-methylbenzenesulfonate (16),
Following the general experimental conditions, 2 (800 mg, 1.17 mmol) dissolved in
dichloromethane (50 mL), was irradiated for 4h and 45 min to afford compound 16 (70 mg,
26
7.8%) as a resinous matter; Rf = 0.72 (Hexanes- EtOAc, 7:3); [
NMR (400 MHz, CHCl3, TMS) : 0.88 (distorted t, 3H, Me-5
Me-15), 1.32 (s, 3H, Me-11), 1.37 (s, 3H, Me-13), 2.43 (s, 3H, Me-4΄΄), 3.57 (m, 2H, H-14a and
α
]
D = 15.5 (
c
0.11, MeOH); 1H
δ
΄), 1.34 (s, 3H, Me-12), 1.26 (brs,
28