Synthesis and structure revision of dimeric tadalafil analogue adulterants in dietary supplements

Article abstract of DOI:10.1248/cpb.c17-00034

A number of phosphodiesterase 5 (PDE5) inhibitors approved by authorities have been used successfully in the treatment of erectile dysfunction. These medicines must be prescribed carefully due to their adverse effects, but they and their analogues are being illegally added to dietary supplements. These illegal dietary supplements pose a significant risk to public health. Several dimeric tadalafil analogues have been synthesized for use as reference standards in the inspection of functional foods that are mainly advertised as sexual enhancement products. During the course of this synthesis, 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate (HATU) was proven to be the reagent of choice for amide coupling to produce these dimeric tadalafil analogues. Moreover, the trans-isomer structures tentatively assigned for the isolated dimeric tadalafil analogues (bisprehomotadalafil and bisprecyclopentyltadalafil) found in dietary supplements are now revised to cis-isomer structures.

Full text of DOI:10.1248/cpb.c17-00034

498
Chem. Pharm. Bull. 65, 498–503 (2017)
Vol. 65, No. 5
Regular Article
Synthesis and Structure Revision of Dimeric Tadalafil Analogue
Adulterants in Dietary Supplements
Suresh Mandava,a Bogonda Ganganna,a Jungjoong Hwang,a Younchang Jang,a Jiho Hwang,a
Mallesham Samala,a Ki-Bbeum Kim,a Haeil Park,a Ji Hyun Lee,b Sun Young Baek,b and
,a
Jongkook Lee*
a College of Pharmacy, Kangwon National University; 1 Kangwondaehak-gil, Chuncheon, Gangwon-do 24341,
b
Republic of Korea: and Advanced Analysis Team, Toxicological Evalauation and Research Department, National
Institute of Food and Drug Safety Evaluation, Ministry of Food and Drug Safety; Cheongju, Chungcheongbuk-do
28159, Republic of Korea.
Received January 12, 2017; accepted March 1, 2017
A number of phosphodiesterase 5 (PDE5) inhibitors approved by authorities have been used success-
fully in the treatment of erectile dysfunction. These medicines must be prescribed carefully due to their
adverse effects, but they and their analogues are being illegally added to dietary supplements. These illegal
dietary supplements pose a significant risk to public health. Several dimeric tadalafil analogues have been
synthesized for use as reference standards in the inspection of functional foods that are mainly advertised as
sexual enhancement products. During the course of this synthesis, 1-[bis(dimethylamino)methylene]-1H-1,2,3-
triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate (HATU) was proven to be the reagent of choice for
amide coupling to produce these dimeric tadalafil analogues. Moreover, the trans-isomer structures tentative-
ly assigned for the isolated dimeric tadalafil analogues (bisprehomotadalafil and bisprecyclopentyltadalafil)
found in dietary supplements are now revised to cis-isomer structures.
Key words phosphodiesterase 5 inhibitor; bisprehomotadalafil; bisprecyclopentyltadalafil; bisprenortadalafil;
dietary supplement
Tadalafil (1, Cialis®), a phosphodiesterase 5 (PDE5) inhibi- the molecules or X-ray crystallography.
tor, has been widely used for the treatment of erectile dysfunc-
To reduce human exposure to these illegal products will
tion (ED). In spite of its excellent efficacy and safety profile, require a method for the rapid screening of dietary supple-
the administration of tadalafil (1) to patients who are concur- ments. Reference standards are needed to establish and oper-
rently using any form of organic nitrates or guanylate cyclase ate screening systems for detecting these dimeric tadalafil
stimulators is contraindicated due to its hypotensive effects.1–3) analogues. As a part of our ongoing efforts to protect public
Tadalafil (1) should also not be prescribed in patients with Ste- health from illegal drug analogues, we report herein the
vens–Johnson syndrome or exfoliative dermatitis because hy- synthesis and structural revision of several dimeric tadalafil
persensitivity reactions to this ED drug have been reported in analogues.
such patients.4) However, tadalafil (1) and its analogues have
been detected in a number of dietary supplements over the last
decade. These illegal products are mainly advertised as sexual
Results and Discussion
Since dimeric tadalafil analogues bearing two (pre)tadalafil
enhancement products and sold without any declarations on moieties were detected together with their monomeric ana-
the label of the pharmacological and toxicological effects aris- logues, we surmised that these dimeric tadalafil analogues
ing from the PDE5 inhibitory activity. Most of the tadalafil were generated from either stereoisomer of chloroacetamide
analogues detected in illegal dietary supplements were gener- 4 as by-products during the synthesis of their monomeric
ated by removal of an N-methyl group or its replacement by analogues and were not properly removed during purifica-
another group such as amino, ethyl, butyl, octyl or cyclopen- tion (Fig. 2). Presumably, dialkylation of amines 5a–c with
tyl.5–14) Epimeric tadalafil analogues5,15) and precursor-related chloroacetamide 4 (with or without subsequent intramolecu-
analogues of tadalafil (1) have also been identified.16–18) These lar amidation) might produce this type of dimeric tadalafil
analogues were presumably designed for use in adulteration to analogue. We initially attempted to improve the formation of
avoid identification by authorities. Recently, we found a new trans-bisprehomotadalafil (trans-2a) from the trans-isomer of
type of tadalafil analogues in dietary supplements, exemplified chloroacetamide 4 by modification of the reaction conditions
by 2a,b and 3, which bear two (pre)tadalafil moieties (Fig. 1). for tadalafil (1).21,22)
We have named these trans-bisprehomotadalafil, trans-bispre-
Contrary to our expectations, only trace amounts of trans-
cyclopentyltadalafil and bisprenortadalafil, respectively.5,19,20) bisprehomotadalafil (trans-2a) were obtained under a variety
The relative stereochemistry of these dimeric analogues of conditions, and the HPLC retention time for this product
was tentatively assigned based on no observation of nuclear was different from that of the previously isolated compound.
Overhauser effect (NOE) between the protons at stereogenic Thus, considering that the isolated bisprehomotadalafil (2a)
centers in the molecules and by comparison of their spectral might be a cis-isomer, we devised a new synthetic route
data with those of monomeric tadalafil derivatives. Their that produces both the cis- and trans-isomers of the dimeric
structures, thus, need to be confirmed clearly by synthesizing tadalafil analogues without forming the monomeric analogues
*To whom correspondence should be addressed. e-mail: jkl@kangwon.ac.kr
© 2017 The Pharmaceutical Society of Japan

Vol. 65, No. 5 (2017)
Chem. Pharm. Bull.
499
Fig. 1. Structures of Tadalafil (1) and Dimeric Tadalafil Analogues (2a,b and 3)
(Chart 1). We postulated that the coupling of diacid 8a with was in good agreement with those of the isolated compound
an excess of aminoester 9 and trans-9 would afford dimeric that was originally reported as a trans-isomer.19) Bisprecy-
tadalafil analogues 2a and trans-2a, respectively. Aminoesters clopentyltadalafil (2b), a cis-isomer, was also synthesized in
9 and trans-9 were prepared according to literature proce- moderate yield using HATU and proven to be identical to the
dures.21,22) Dialkylation of amines 5a,b with bromoacetate 6 isolated compound that had tentatively been assigned as the
produced diester 7a,b, from which the benzyl groups were trans-isomer (Table 1, Entry 11).5) These results were further
removed by hydrogenolysis to produce diacids 8a,b in high confirmed by performing extensive NOE experiments. In con-
yields.23) With these diacids 8a,b in hands, we moved next trast to 2a (isolated and synthesized) in CDCl3, NOE between
to couple them with amines 9 and trans-9. Initial attempts the protons at stereogenic centers in 2a were clearly observed
using N,N′-dicyclohexylcarbodiimide (DCC), 1-ethyl-3-(3- when acetone-d6 was used as an NMR solvent (Fig. S36). We
dimethylaminopropyl)carbodiimide (EDCI) [with hydroxy- also observed NOE between the protons at C-6 and C-12a
benzotriazole (HOBt), 1-hydroxy-7-azabenzotriazole (HOAt) when we carried out NOE experiments with a large amount of
or 4-dimethylaminopyridine (DMAP)] and benzotriazol-1-yl- 2b (synthesized) in acetone-d6. (Fig. S37).
oxytripyrrolidinophosphonium hexafluorophosphate (PyBOP)
We next sought to synthesize bisprenortadalafil (3). Cou-
produced only trace amounts of the dimeric analogues 2a and pling of aminoester 9 and the readily available diacid 10 using
trans-2a, with nearly quantitative recovery of the starting ma- HATU gave diamide 11 in 54% yield27) (Chart 2). Removal
terials 9 and trans-9 under various conditions. We also tried of the benzyloxycarbonyl (Cbz) group in 11 with catalytic
to couple amines 9 and trans-9 to an acid chloride generated hydrogenolysis under basic conditions concomitant with in-
from diacid 8a by using thionyl chloride or oxalyl chloride, tramolecular amidation of the resulting amine succeeded to
but we could not observe the formation of 2a and trans-2a.
give bisprenortadalafil (3) in 75% yield. The spectral data for
We ascribed these poor results to steric hindrance in the synthesized bisprenortadalafil (3) were in good accordance
amines 9 and trans-9 due to the bulky substituents in their with those for the isolated compound.20)
piperidine rings. We investigated several other amide coupling
reagents (Table 1). Significant amounts of dimeric tadalafil
Conclusion
analogues 2a and trans-2a were obtained when bromo-tris-
We have accomplished the synthesis of dimeric tadalafil
pyrrolidino-phosphonium hexafluorophosphate (PyBroP)24) analogues 2a,b and 3, and found that HATU is an effec-
and 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholin- tive coupling reagent for diamide formation from the steri-
ium chloride (DMT-MM) were used to activate diacid 8a cally hindered amine 9 during the course of the synthesis.
(Table 1, Entries 1–5).25) Further investigation revealed The structures of the isolated dimeric tadalafil analogues
that among the reagents examined, 1-[bis(dimethylamino)- 2a,b, which had tentatively been assigned as trans-isomers,
methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexa- were revised to cis-isomers. We believe that these synthesized
fluorophosphate (HATU) was the most efficient in producing dimeric tadalafil analogues will be highly useful as reference
these dimeric tadalafil analogues 2a and trans-2a in moderate standards to aid in inspections of illegally adulterated dietary
to good yields (Table 1, Entries 6–10).26) Contrary to the other supplements that contain this type of tadalafil analogue.
coupling reagents examined, amine 9 was almost completely
consumed by activation of diacid 8a with HATU. The spectral
data for the synthesized bisprehomotadalafil (2a), a cis-isomer,
Experimental
General Information All chemicals, solvents and re-

500
Chem. Pharm. Bull.
Vol. 65, No. 5 (2017)
Fig. 2. Proposed Mechanism for Generation of Tadalafil (1), trans-Tadalafil (trans-1) and Dimeric Tadalafil Analogues (2a,b, trans-2a,b, 3 and
trans-3)
agents were obtained from commercial suppliers and used extracted with CH2Cl2 (3×50mL). The combined organic lay-
without further purification. All reactions were conducted ers were washed with brine (50mL), dried over anhydrous
under a nitrogen or argon atmosphere in oven-dried glassware MgSO4, and concentrated under reduced pressure. The result-
with magnetic stirring. All the reactions were monitored by ing residue was purified by flash column chromatography
TLC using Merck pre-coated TLC plates (silica gel 60GF254, on silica gel (hexane–EtOAc, 9:1) to give aminoester 7a
1
0.25mm), and flash column chromatographic separations were (8.0g, 95%) as a colorless oil: H-NMR (400MHz, CDCl3)
carried out using silica gel (Merck Kieselgel 60, 230–400 δ: 7.41–7.28 (m, 10H), 5.14 (s, 4H), 3.61 (s, 4H), 2.79 (q,
mesh) packed in glass columns. 1H- and 13C-NMR spectra J=7.2Hz, 2H), 1.07 (t, J=7.2Hz, 3H); 13C-NMR (100MHz,
were recorded on Bruker Avance II 600MHz and Bruker CDCl3) δ: 171.1, 135.7, 128.5, 128.3, 66.2, 54.6, 48.3, 13.0; IR
DPX 400MHz spectrometers. The mass spectra were obtained (KBr, cm−1) 3424, 3100, 3012, 2969, 1694, 1472, 1404, 1363;
using JEOL JMS-700 and Waters Xevo G2 QTOF mass spec- high resolution (HR)-MS-electron ionization (EI): m/z Calcd
trometers. The IR spectra were recorded on a Frontier FT-IR for C20H23NO4 (M+) 341.1627. Found 341.1626.
spectrophotometer. Optical rotation data were collected by
using a Jasco P-2000 digital polarimeter.
2,2′-(Ethylazanediyl)diacetic Acid (8a) A mixture of
aminoester 7a (2.50g, 7.3mmol) and 5% Pd/C (350mg) in
Benzyl 2,2′-(Ethylazanediyl)diacetate (7a) To a stirred MeOH (50mL) was stirred under a hydrogen atmosphere for
suspension of bromoacetate 6 (11.2mL, 48.7mmol) and 12h at room temperature. The mixture was filtered through a
KHCO3 (5.5g, 54.9mmol) in N,N-dimethylformamide (DMF) pad of Celite, and washed with aqueous MeOH (50%, 50mL).
(100mL) was slowly added ethylamine 5a (70% aqueous so- The filtrate was concentrated in vacuo. Traces of water in
lution, 2.0mL, 22.1mmol) over 5min at 0°C. After 30min the resulting residue were removed by co-evaporation with
at 0°C, the mixture was stirred overnight at room tempera- isopropyl alcohol to yield diacid 8a (1.13g, 95%) as an off-
1
ture and concentrated in vacuo. Saturated aqueous NaHCO3 white solid: H-NMR (400MHz, D2O) δ: 3.99 (s, 4H), 3.38 (q,
(50mL) was added to the residue. The resulting mixture was J=7.2Hz, 2H)), 1.31 (t, J=7.2Hz, 3H); 13C-NMR (100MHz,

Vol. 65, No. 5 (2017)
Chem. Pharm. Bull.
501
Chart 1. Synthesis of Bisprehomotadalafils (2a and trans-2a) and Bisprecyclopentyltadalafil (2b)
Table 1. Coupling Reactions of Diacids 8a,b with Aminoester 9 or trans-9
Entry
Amine
Diacid
Reaction conditionsa,b)
Product
Yield (%)c)
1
2
trans-9
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8b
PyBroP, i-Pr2EtN, CH2Cl2, reflux, 48h
PyBroP, i-Pr2EtN, (CH2)2Cl2, reflux, 48h
PyBroP, i-Pr2EtN, CH2Cl2, reflux, 48h
PyBroP, i-Pr2EtN, (CH2)2Cl2, reflux, 48h
DMT-MM, NMM, CH2Cl2, rt, 48h
HATU, i-Pr2EtN, CH2Cl2, reflux, 62h
HATU, i-Pr2EtN, (CH2)2Cl2, reflux, 62h
HATU, i-Pr2EtN, CH2Cl2, reflux, 48h
HATU, i-Pr2EtN, (CH2)2Cl2, reflux, 48h
HATU, i-Pr2EtN, CH2Cl2, rt, 48h
trans-2a
trans-2a
2a
5
5
trans-9
3
9
54d)
4
9
2a
8
5
9
2a
47e)
10
32
12
11
68
60
6
trans-9
trans-2a
trans-2a
2a
7
trans-9
8
9
9
9
9
9
2a
10
11
2a
HATU, i-Pr2EtN, CH2Cl2, rt, 48h
2b
a) The reactions were conducted with diacids 8a,b (1.0 molar equiv.), amine 9 and trans-9 (2.0 molar equiv.), coupling reagents (3.0 molar equiv.) and bases (6.0 molar
equiv.). b) PyBroP, bromo-tris-pyrrolidino-phosphonium hexafluorophosphate; HATU, 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluoro-
phosphate; DMT-MM, 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride; NMM, N-methylmorpholine. c) Isolated. d) Based on recovered starting material
(70% conversion). e) Based on recovered starting material (50% conversion).
D2O) δ: 169.9, 55.6, 52.2, 9.0; IR (KBr, cm−1) 2970, 1746, 1455, 1455, 1152, 1001; HR-MS-EI: m/z Calcd for C23H27NO4 (M+)
1168; HR-MS-EI: m/z Calcd for C6H11NO4 (M+) 161.0688. 381.1940. Found 381.1939.
Found 161.0689. The diacid 8a was pure enough to provide
spectral data, and was subjected to the next step without fur- 8b was synthesized according to the procedure described for
2,2′-(Cyclopentylazanediyl)diacetic Acid (8b) Diacid
1
ther purification.
8a, and was obtained as a white solid (yield: 87%): H-NMR
Benzyl 2,2′-(Cyclopentylazanediyl)diacetate (7b) Ami- (400MHz, D2O) δ: 3.99 (s, 4H), 3.97–3.88 (m, 1H), 2.16–2.03
noester 7b was synthesized according to the procedure de- (m, 2H), 1.84–1.64 (m, 6H); 13C-NMR (100MHz, D2O) δ:
scribed for 7a, and was obtained as a pale yellow oil (yield: 170.2, 68.4, 55.6, 27.5 23.7; IR (KBr, cm−1) 3425, 3031, 2975,
94%): 1H-NMR (600MHz, CDCl3) δ: 7.41–7.28 (m, 10H), 2956, 1737, 1633, 1413, 1333, 1233; HR-MS-EI: m/z Calcd for
5.14 (s, 4H), 3.67 (s, 4H), 3.31–3.28 (m, 1H), 1.85–1.74 (m, C9H15NO4 (M+) 201.1001. Found 201.1001.
2H), 1.75–1.61 (m, 2H), 1.57–1.46 (m, 2H), 1.45–1.34 (m,
(1R,1′R,3R,3′R)-Dimethyl-2,2′-(2,2′-(ethylazanediyl)-
2H); 13C-NMR (150MHz, CDCl3) δ: 171.5, 135.7, 128.5, bis(acetyl))bis(1-(benzo[d][1,3]dioxol-5-yl)-2,3,4,9-tetra-
128.2, 66.2, 63.8, 54.0, 31.1, 23.5; IR (KBr, cm−1) 2956, 1747, hydro-1H-pyrido[3,4-b]indole-3-carboxylate) (2a) A mix-

502
Chem. Pharm. Bull.
Vol. 65, No. 5 (2017)
amine (0.68mL, 3.87mmol) and HATU (814mg, 2.14mmol)
in anhydrous dichloroethane (10mL) was stirred at room
temperature for 30min. Amine trans-9 (500mg, 1.42mmol)
was added and the reaction mixture was refluxed for 62h.
The mixture was poured into ice-cold water (5mL) and di-
luted with CH2Cl2 (50mL). The organic layer was separated,
washed with brine, dried over anhydrous MgSO4, and concen-
trated at reduced pressure. The residue was semi-purified by
flash column chromatography on silica gel (hexane–EtOAc,
1:1). The crude solid was recrystallized from isopropylalco-
hol to give trans-bisprehomotadalafil (trans-2a, 190mg, 32%)
1
as an off-white solid: [α]D25 +62.6 (c=0.05, CHCl3); H-NMR
(600MHz, CDCl3) δ: 8.25 (s, 2H), 7.50 (d, J=7.0Hz, 2H),
7.13 (d, J=24.8Hz, 6H), 6.81 (s, 2H), 6.71 (d, J=12.7Hz, 4H),
6.37 (s, 2H), 5.91 (s, 4H), 4.95 (s, 2H), [3.60 (s), 3.48 (brd,
J=11.9Hz), 3.19 (brd, J=12.6Hz) and 3.11 (brd, J=14.8Hz),
14H], 2.67 (s, 2H), 0.84 (s, 3H); 13C-NMR (150MHz, CDCl3)
δ: 173.0, 171.9, 148.2, 147.3, 136.4, 135.2, 133.5, 126.4, 122.1,
120.1, 119.7, 118.3, 111.4, 108.3, 107.3, 106.8, 101.2, 56.7, 56.0,
53.9, 52.3, 49.1, 22.5, 11.5; IR (KBr, cm−1) 3399, 2953, 2904,
1737, 1654, 1488, 1441, 1239, 1038; HR-MS-ESI (negative): m/z
Calcd for C46H42N5O10 [M−H]−824.2932. Found 824.2936.
(1R,1′R,3R,3′R)-Dimethyl-2,2′-(2,2′-(cyclopentylaza-
nediyl)bis(acetyl))bis(1-(benzo[d][1,3]dioxol-5-yl)-2,3,4,9-
tetrahydro-1H-pyrido[3,4-b]indole-3-carboxylate)
(2b)
Bisprecyclopentyltadalafil (2b) was synthesized according to
the procedure described for 2a, and was obtained as a white
solid (yield: 60%): [α]D25 +155.0 (c=0.1, CHCl3); 1H-NMR
(600MHz, (CD3)2CO) δ: 9.89 (s, 2H), 7.56 (d, J=7.7Hz,
2H), 7.34 (d, J=8.0Hz, 2H), 7.12 (t, J=7.4Hz, 2H), 7.06 (t,
J=7.4Hz, 2H), 6.90 (s, 2H), 6.85 (s, 2H), 6.68 (d, J=8.0Hz,
Chart 2. Synthesis of Bisprenortadalafil (3)
ture of diacid 8a (0.4g, 2.5mmol), diisopropylethylamine 2H), 6.53 (d, J=7.9Hz, 2H), 6.16 (d, J=6.5Hz, 2H), 5.94 (d,
(2.7mL, 15.5mmol) and HATU (3.0g, 7.9mmol) in anhydrous J=4.2Hz, 4H), 3.97 (d, J=14.7Hz, 2H), 3.61 (d, J=15.8Hz,
CH2Cl2 (50mL) was stirred at room temperature for 30min. 2H), 3.52 (d, J=14.7Hz, 2H), 3.41–3.30 (m, 1H), 3.16–3.11 (m,
Amine 9 (1.8g, 5.2mmol) was added and the reaction mix- 8H), 1.79 (s, 2H), 1.64–1.36 (m, 6H); 13C-NMR (150MHz,
ture was stirred for 48h at room temperature. The mixture (CD3)2CO) δ: 172.3, 171.6, 148.2, 147.8, 137.8, 135.5, 131.5,
was poured into ice-cold water (20mL) and diluted with 127.6, 123.6, 122.6, 119.8, 119.0, 112.0, 110.6, 108.3, 108.1,
CH2Cl2 (100mL). The organic layer was separated, washed 101.9, 64.8, 57.1, 52.6, 52.1, 51.8, 30.5, 27.3, 24.5, 24.3, 22.1;
with brine, dried over anhydrous MgSO4, and concentrated at IR (KBr, cm−1): 3396, 2951, 1740, 1650, 1487, 1439, 1236,
reduced pressure. The residue was semi-purified by flash col- 1039; HR-MS-ESI (negative): m/z Calcd for C49H46N5O10
umn chromatography on silica gel (hexane–EtOAc, 1:1). The [M−H]− 864.3245. Found 864.3265.
crude solid was recrystallized from isopropylalcohol to afford
Dimethyl-2,2′-(2,2′(((benzyloxy)carbonyl)azanediyl)-
bisprehomotadalafil (2a, 1.4g, 68%) as an off-white solid: bis(acetyl))(1R,1′R,3R,3′R)-bis(1-(benzo[d][1,3]dioxol-5-yl)-
1
[α]D25 −160.2 (c=0.1, CHCl3); H-NMR (600MHz, CDCl3) δ: 2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indole-3-carboxylate)
7.73 (s, 2H), 7.54 (d, J=7.2Hz, 2H), 7.25 (d, J=7.8Hz, 2H), (11) Carbamate 11 was synthesized according to the pro-
7.18 (t, J=7.3Hz, 2H), 7.13 (t, J=7.4Hz, 2H), 6.88 (s, 2H), cedure described for 2a and was obtained as an off-white
6.82 (s, 2H), 6.60 (d, J=7.9Hz, 2H), 6.55 (d, J=7.5Hz, 2H), solid (yield: 54%): [α]D25 −128.5 (c=0.2, CHCl3); 1H-NMR
5.86 (s, 4H), 5.62 (s, 2H), 3.80 (d, J=14.5Hz, 2H), 3.64 (d, (characterization of the distinguishable two rotamers was
J=15.7Hz, 2H), 3.35 (d, J=14.1Hz, 2H), 3.17 (s, 6H), 3.08 (dd, performed; 1:1.2 rotamer ratio; asterisks denote peaks of the
J=15.4, 6.7Hz, 2H), 2.79 (brs, 1H), 2.69–2.63 (m, 1H), 1.07 minor rotamer detected separately from those of the major ro-
(s, 3H); 13C-NMR (150MHz, CDCl3) δ: 171.0, 170.5, 147.5, tamer, 600MHz, CDCl3) δ: [8.07* (s), 7.83 (s), 2H)], 7.53 (dd,
147.1, 136.3, 133.5, 130.1, 126.4, 122.9, 122.4, 119.7, 118.6, J=12.4, 7.8Hz, 2H), 7.32–7.27 (m, 2H), 7.18 (d, J=5.0Hz, 4H),
111.0, 110.1, 107.8, 107.5, 101.0, 57.5, 52.1, 52.0, 51.4, 49.3, 7.15 (s, 5H), 6.86 (d, J=18.3Hz, 2H), 6.78 (s, 2H), 6.62–6.53
21.5, 11.9; IR (KBr, cm−1) 3403, 2951, 2899, 1738, 1645, 1488, (m, 4H), 5.89–5.79 (m, 4H), 5.07 (d, J=12.3Hz, 1H), 4.99 (d,
1440, 1237, 1038; HR-MS-electrospray ionization (ESI) (nega- J=12.3Hz, 1H), 4.80 (d, J=6.5Hz, 1H), 4.70–4.57 (m, 2H),
tive): m/z Calcd for C46H42N5O10 [M−H]− 824.2932. Found 4.53 (d, J=16.9Hz, 1H), 4.34 (m, 2H), [3.50 (d, J=15.8Hz) and
824.2944.
3.37* (d, J=14.8Hz), 2H], [3.14 (s), 3.00* (s), 6H], 2.81–2.75
(1R,1′R,3S,3′S)-Dimethyl-2,2′-(2,2′-(ethylazanediyl)- (m, 2H); 13C-NMR (100MHz, CDCl3) δ: 170.3, 170.1, 168.9,
bis(acetyl))bis(1-(benzo[d][1,3]dioxol-5-yl)-2,3,4,9-tetra- 168.6, 156.5, 147.4, 147.3, 147.1, 147.0, 136.4, 136.3, 135.8,
hydro-1H-pyrido[3,4-b]indole-3-carboxylate) (trans-2a)
A
133.3, 133.1, 129.6, 128.3, 128.0, 127.8, 126.3, 123.0, 122.5,
mixture of diacid 8a (115mg, 0.71mmol), diisopropylethyl- 122.3, 119.7, 119.5, 118.7, 118.6, 111.0, 110.9, 110.1, 110.0, 107.9,

Vol. 65, No. 5 (2017)
Chem. Pharm. Bull.
503
3) Bella A. J., Lee J. C., Carrier S., Benard F., Brock G. B., Can. Urol.
Assoc. J., 9, 23–29 (2015).
4) “Safety Information: Cialis (tadalafil) tablets.”: ‹http://www.fda.gov/
Safety/MedWatch/SafetyInformation/ucm203554.htm›, accessed on
8 Oct., 2016.
107.5, 107.5, 107.3, 101.01, 100.97, 68.1, 67.6, 52.2, 52.1, 52.0,
51.8, 50.1, 49.5, 23.9, 21.3, 21.1; IR (KBr, cm−1): 3394, 3316,
2949, 1742, 1700, 1658, 1487, 1439, 1237; HR-MS-ESI (nega-
tive): m/z Calcd for C52H44N5O12 [M−H]− 930.2986. Found
930.3016.
5) Lee J. H., Mandava S., Baek S. Y., Lee Y.-M., J. Pharm. Biomed.
Anal., 123, 1–9 (2016).
Methyl(1R,3R)-1-(benzo[d][1,3]dioxol-5-yl)-2-(2-((6R,
12aR)-6-(benzo[d][1,3]dioxol-5-yl)-1,4-dioxo-3,4,6,7,12,12a-
hexahydropyrazino[1′,2′:1,6]pyrido[3,4-b]indol-2(1H)-yl)-
acetyl)-2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indole-3-car-
boxylate (3) A mixture of dimer 11 (0.60g, 0.64mmol),
5% Pd/C (0.24g) and five drops of Et3N was stirred under
a hydrogen atmosphere at room temperature for 3d. The
mixture was filtered through a pad of Celite, and the filtrate
was concentrated at reduced pressure. The crude solid was
recrystallized from methanol to yield bisprenortadalafil (3,
0.37g, 75%) as an off-white solid: [α]D25 −61.0 (c=0.1, CHCl3);
1H-NMR (600MHz, (CD3)2CO) δ: 10.21 (s, 1H), 9.95 (s,
1H), 7.61 (d, J=7.6Hz, 1H), 7.58 (d, J=7.8Hz, 1H), 7.36 (d,
J=8.0Hz, 1H), 7.33 (d, J=7.9Hz, 1H), 7.13 (t, J=7.4Hz, 1H),
7.11–7.03 (m, 3H), 6.93 (s, 1H), 6.90 (s, 1H), 6.90–6.87 (m,
1H), 6.82 (s, 1H), 6.70 (d, J=8.1Hz, 1H), 6.68 (d, J=8.1Hz,
1H), 6.56 (d, J=7.6Hz, 1H), 6.33 (s, 1H), 5.94 (s, 2H), 5.89
(dd, J=9.5, 0.9Hz, 2H), 5.23 (d, J=6.8Hz, 1H), 4.88 (d,
J=16.2Hz, 1H), 4.59–4.52 (m, 2H), 4.40 (d, J=16.9Hz, 1H),
3.98 (d, J=16.9Hz, 1H), 3.68–3.60 (m, 2H), 3.18–3.11 (m, 5H);
13C-NMR (150MHz, (CD3)2CO) δ: 171.6, 168.7, 168.5, 168.1,
148.6, 148.4, 148.1, 147.7, 138.0, 137.9, 137.7, 135.1, 135.0,
131.2, 127.6, 127.4, 123.8, 122.8, 122.6, 121.0, 120.2, 120.0,
119.2, 119.1, 112.2, 112.1, 110.7, 108.7, 108.3, 108.2, 108.1,
106.6, 102.1, 102.0, 56.7, 56.6, 52.9, 52.5, 52.3, 52.1, 48.7, 24.0,
22.2; IR (KBr, cm−1) 3402, 2897, 1740, 1664, 1487, 1440,
1306, 1204; HR-MS-ESI (negative): m/z Calcd for C43H34N5O9
[M−H]− 764.2357. Found 764.2375.
6) Xu Y., Kee C.-L., Ge X., Low M.-Y., Koh H.-L., J. Pharm. Biomed.
Anal., 118, 235–241 (2016).
7) Huang Y.-C., Lee H.-C., Lin Y.-L., Li C.-Y., Tsai C.-F., Cheng H.-F.,
Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk As-
sess., 33, 179-185 (2016).
8) Kern S. E., Nickum E. A., Flurer R. A., Toomey V. M., Litzau J. J.,
J. Pharm. Biomed. Anal., 103, 99–103 (2015).
9) Lee J. H., Kim H. J., Noh E., Kim J. Y., Cho S. H., Do J.-A., Yoon
C.-Y., Cho S., Kim W. S., J. Pharm. Biomed. Anal., 103, 80–84
(2015).
10) Lee E.-S., Kim J. W., Lee J. H., Han K. M., Cho S., Hwang I., Han
S. Y., Chae K., Kim J., Food Addit. Contam. Part A Chem. Anal.
Control. Expo. Risk Assess., 30, 621–626 (2013).
11) Toomey V. M., Litzau J. J., Flurer C. L., J. Pharm. Biomed. Anal.,
59, 50–57 (2012).
12) Häberli A., Girard P., Low M.-Y., Ge X., J. Pharm. Biomed. Anal.,
53, 24–28 (2010).
13) Hasegawa T., Takahashi K., Saijo M., Ishii T., Nagata T., Kurihara
M., Haishima Y., Goda Y., Kawahara N., Chem. Pharm. Bull., 57,
185–189 (2009).
14) Kurita H., Mizuno K., Kuromi K., Suzuki N., Ueno C., Kamimura
M., Fujiwara A., Owada K., Ogo N., Yamamoto M., J. Health Sci.,
54, 310–314 (2008).
15) Venhuis B. J., Zomer G., Vredenbregt M. J., de Kaste D., J. Pharm.
Biomed. Anal., 51, 723–727 (2010).
16) Kern S. E., Lorenz L. M., Lanzarotta A., Nickum E. A., Litzau J. J.,
J. Pharm. Biomed. Anal., 128, 360–366 (2016).
17) Zhang G., Yu Y., Wu X., Li J., J. Pharm. Biomed. Anal., 94, 210–
214 (2014).
18) Hasegawa T., Saijo M., Ishii T., Nagata T., Haishima Y., Kawahara
N., Goda Y., Shokuhin Eiseigaku Zasshi, 49, 311–315 (2008).
19) Lee J. H., Kim H. J., Mandava S., Hwang J., Park H. J., Cho S.,
Baek S. Y., Lee J., J. Pharm. Biomed. Anal., 115, 352–358 (2015).
20) Lee J. H., Park H. N., Ganganna B., Jeong J. H., Park S.-K., Lee
J., Baek S. Y., Food Addit. Contam. Part A Chem. Anal. Control.
Expo. Risk Assess., 33, 945–952 (2016).
21) Daugan A., Grondin P., Ruault C., Le Monnier de Gouville A.-C.,
Coste H., Linget J. M., Kirilovsky J., Hyafil F., Labaudinière R., J.
Med. Chem., 46, 4533–4542 (2003).
Acknowledgments This research was supported by a
Research Grant (15182MFDS523) from the Ministry of Food
and Drug Safety (MFDS) in Korea and a 2015 Research Grant
(D1000298-01-01) from Kangwon National University. We
thank the Central Laboratory of Kangwon National University
for providing us with technical assistance on the spectroscopic
experiments.
22) Xiao S., Shi X.-X., Xing J., Yan J.-J., Liu S.-L., Lu W.-D., Tetrahe-
dron Asymmetry, 20, 2090–2096 (2009).
23) Pulido D., Albericio F., Royo M., Org. Lett., 16, 1318–1321 (2014).
24) Frérot E., Coste J., Pantaloni A., Dufour M.-N., Jouin P., Tetrahe-
dron, 47, 259–270 (1991).
Conflict of Interest The authors declare no conflict of
interest.
Supplementary Materials The online version of this ar-
ticle contains supplementary materials.
25) Kunishima M., Kawachi C., Monta J., Terao K., Iwasaki F., Tani S.,
Tetrahedron, 55, 13159–13170 (1999).
References
26) Carpino L. A., J. Am. Chem. Soc., 115, 4397–4398 (1993).
27) Khattab S. N., El-Faham A., El-Massry A. M., Mansour E. M. E.,
Abd El-Rahman M. M., Lett. Pept. Sci., 7, 331–345 (2000).
1) Ventimiglia E., Capogrosso P., Montorsi F., Salonia A., Expert
Opin. Drug Saf., 15, 141–152 (2016).
2) Ryu J. K., Cho K. S., Kim S. J., Oh K. J., Kam S. C., Seo K. K.,
Shin H. S., Kim S. W., World J. Mens Health, 31, 83–102 (2013).