Synthesis, cytotoxicity and QSAR study of N-tosyl-1,2,3,4- tetrahydroisoquinoline derivatives

Article abstract of DOI:10.1007/s12272-013-0111-9

1-Substituted-N-tosyl-1,2,3,4-tetrahydroisoquinoline analogs (4a-4l) were synthesized using the modified Pictet-Spengler reaction and evaluated for cytotoxicity. All tetrahydroisoquinolines displayed cytotoxicity against MOLT-3 cell lines, except for p-me

Full text of DOI:10.1007/s12272-013-0111-9

Arch. Pharm. Res.
DOI 10.1007/s12272-013-0111-9
RESEARCH ARTICLE
Synthesis, cytotoxicity and QSAR study of N-tosyl-1,2,3,4-
tetrahydroisoquinoline derivatives
•
Ratchanok Pingaew Apilak Worachartcheewan
•
•
Chanin Nantasenamat Supaluk Prachayasittikul
•
•
Somsak Ruchirawat Virapong Prachayasittikul
Received: 1 September 2012 / Accepted: 25 March 2013
Ó The Pharmaceutical Society of Korea 2013
Abstract 1-Substituted-N-tosyl-1,2,3,4-tetrahydroisoquin-
oline analogs (4a–4l) were synthesized using the modified
Pictet–Spengler reaction and evaluated for cytotoxicity. All
tetrahydroisoquinolines displayed cytotoxicity against
MOLT-3 cell lines, except for p-methoxy analog 4d. Inter-
estingly, the o-hydroxy derivative 4k was shown to be the
most potent cytotoxic againstHuCCA-1, A-549 and MOLT-3
cell lines. The lowest IC50 value of 1.23 lM was observed for
MOLT-3 cells. Trimethoxy analog 4f exerted the most potent
activity against HepG2 with an IC50 of 22.70 lM, which is
lower than the reference drug, etoposide. QSAR studies
showed that total symmetry index (Gu), 3D-MoRSE (Mor31v
and Mor32u) and 3D Petitjean index (PJI3) were the most
important descriptors accounting for the observed cytotox-
icities. The most potent cytotoxic compound (4k) against
MOLT-3 had the highest Gu value, correspondingly the
inactive p-methoxy analog (4d) had the lowest Gu value. On
the other hand, the highest molecular mass compound (4f)
was shown to be the most potent cytotoxic against HepG2
cells. The studies disclose that tetrahydroisoquinolines 4f and
4k are potentially interesting lead pharmacophores that
should be further explored. The QSAR models provided
insights into the physicochemical properties of the investi-
gated compounds.
Keywords Cytotoxicity ꢀ Isoquinoline ꢀ Multiple linear
regression ꢀ Pictet–Spengler reaction ꢀ QSAR ꢀ
Sulfonamide
Introduction
Isoquinoline is an important class of heterocyclic scaffold
which is frequently found in many natural sources (Bentley
1998). Natural and synthetic isoquinoline alkaloids exhibit
a variety of pharmacological properties including antican-
cer, antimicrobial, antimalarial and anti-HIV activities
R. Pingaew (&)
Department of Chemistry, Faculty of Science, Srinakharinwirot
University, Bangkok 10110, Thailand
e-mail: ratchanok@swu.ac.th
´
(Bermejo et al. 2002; Cui et al. 2006; Gonzalez et al.
A. Worachartcheewan ꢀ C. Nantasenamat ꢀ S. Prachayasittikul
Center of Data Mining and Biomedical Informatics, Faculty of
Medical Technology, Mahidol University, Bangkok 10700,
Thailand
2007a; Iwasa et al. 2001; Pingaew et al. 2013; Saitoh et al.
2006; Scott and Williams 2002; Siengalewicz et al. 2008).
Sulfonamide moiety is present in diverse biologically
active molecules (Scozzafava et al. 2003). Aryl/heteroaryl
sulfonamides are known to exert antitumor properties.
Some of the modified analogs such as E7010 (ABT-751),
HMN-214, T138067 and T900607 have been evaluated in
clinical trials (Hu et al. 2008). Various biological studies of a
series of 1,2,3,4-tetrahydroisoquinolines based sulfonamide
have been previously documented. They have been revealed
to act as phenylethanolamine N-methyltransferase inhibitors
(Grunewald et al. 2005), matrix metalloproteinase inhibitors
A. Worachartcheewan ꢀ C. Nantasenamat ꢀ
V. Prachayasittikul (&)
Department of Clinical Microbiology and Applied Technology,
Faculty of Medical Technology, Mahidol University,
Bangkok 10700, Thailand
e-mail: mtvpr@mahidol.ac.th
S. Ruchirawat
Chulabhorn Research Institute and Chulabhorn Graduate
Institute, Bangkok 10210, Thailand
123

R. Pingaew et al.
General procedure for the synthesis of N-sulfonyl-
1,2,3,4-tetrahydroisoquinolines (4a–4l)
(Ma et al. 2004; Matterand Schwab 1999; Matter et al. 2002),
carbonic anhydrase inhibitors (Gitto et al. 2009, 2010),
´
bradykinin-1 antagonists (Huszar et al. 2008) as well as
A mixture of sulfonamide 3 (0.67 mmol) and appropriate
aldehyde (0.72 mmol) in formic acid or trifluoroacetic acid
(15 mL) was stirred under reflux for 0.5–27 h and then
allowed to cool to room temperature. The reaction mixture
was added to 30 mL of ice cool water and the product was
extracted with CH2Cl2 (2 9 30 mL). Combined extracts
were washed with saturated aqueous NaHCO3, dried (anh.
Na2SO4) and evaporated to dryness under reduced pres-
sure. The crude product was purified by silica gel column
chromatography.
potential ligands of the delta opioid receptor (Barn et al.
2001; Bourdier et al. 2007). Considering the interesting
bioactivities of the isoquinolines and sulfonamides has
prompted us to design and synthesize a series of 1,2,3,4-
tetrahydroisoquinoline derivatives bearing the sulfonamide
entity through the use of simple and efficient method.
The Pictet–Spengler reaction which involves the cycli-
zation of iminium ion derived from the condensation of
b-arylethylamine derivatives with aldehydes, is one of the
most important strategies for the construction of tetrahy-
droisoquinoline (Cox and Cook 1995; Larghi et al. 2005).
However, this method has a limitation regarding its pre-
requisite of electron donating substituents. Modifications to
the original approach by means of using the masked car-
bonyls and electron withdrawing groups, such as acyl or
sulfonyl moieties, bound to the starting b-arylethylamine
have been developed (Ito and Tanaka 1977; Lukanov et al.
1987; Orazi et al. 1986; Silveira et al. 1999, 2003, 2006). In
this article, we employed the modified Pictet–Spengler
reaction for the synthesis of a series of N-tosyl-1,2,3,4-
tetrahydroisoquinolines, by exploring the effects of various
substituents at the C-1 position of the tetrahydroisoquino-
line system. It is expected that not only the sulfonyl group
affects the cyclization by increasing the electrophilicity of
an iminium intermediate, but the sulfonyl group could also
be a useful feature governing the bioactivity. Although
some N-sulfonyltetrahydroisoquinoline analogs have been
previously known, however their cytotoxic activities have
apparently not yet been published. In this regard, the
cytotoxicities of N-sulfonyl-1,2,3,4-tetrahydroisoquinolines
were investigated. To provide insights into the effect that
substituents at the C-1 position has on their cytotoxicities,
the quantitative structure–activity relationship (QSAR) of
the compounds was investigated.
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-isoquinoline (4a)
85 % yield. mp 144–145 °C [144–145 °C (Ito and Tanaka
1977)]. 1H NMR (300 MHz, CDCl3): d (ppm) 2.40 (s, 3H,
ArCH3), 2.82 (t, J = 5.8 Hz, 2H, C4), 3.30 (t, J = 5.8 Hz,
2H, C3), 3.79, 3.80 (2 s, 6H, 2 9 OCH3), 4.14 (s, 2H, C-1),
6.48 (s, 1H, C5-ArH), 6.52 (s, 1H, C8-ArH), 7.30 (d,
J = 8.0 Hz, 2H, C30-ArH2), 7.69 (d, J = 8.0 Hz, 2H, C20-
ArH2); 13C NMR (75 MHz, CDCl3): d (ppm) 21.5, 28.4,
43.8, 47.2, 55.9, 109.0, 111.3, 123.4, 125.0, 127.7, 129.7,
133.3, 143.6, 147.7, 147.9. IR (UATR) cm-1: 1612, 1518,
1342, 1161, 1116. TOF-MS m/z: 348.1271 (Calcd for
C18H22NO4S: 348.1270).
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-methylisoquinoline (4b)
89 % yield. mp: 150–151 °C [150 °C (Hazebroucq 1966)].
1H NMR (300 MHz, CDCl3): d (ppm) 1.43 (d, J = 6.7 Hz,
3H, CHCH3), 2.34 (s, 3H, ArCH3), 2.36–2.68 (m, 2H, C4),
3.30–3.41, 3.75–3.92 (m, 2H, C3), 3.77, 3.81 (2 s, 6H,
2 9 OCH3), 5.03 (q, J = 6.7 Hz, 1H, C1), 6.42, 6.49 (2 s,
2H, C5-ArH, C8-ArH), 7.17 (d, J = 7.8 Hz, 2H, C30-
ArH2), 7.64 (d, J = 7.8 Hz, 2H, C20-ArH2); 13C NMR
(75 MHz, CDCl3): d (ppm) 21.4, 23.3, 27.3, 38.5, 51.7,
55.9, 56.0, 109.5, 111.4, 124.7, 126.9, 129.5, 129.7, 138.1,
143.0, 147.6, 147.8. IR (UATR) cm-1: 1612, 1518, 1323,
1154, 1125. TOF-MS m/z: 362.1423 (Calcd for
C19H24NO4S: 362.1421).
Materials and methods
Column chromatography was carried out using silica gel 60
(70–230 mesh ASTM). Analytical thin-layer chromatog-
raphy (TLC) was performed with silica gel 60 F254 alu-
1
minium sheets. H- and 13C-NMR spectra were recorded
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-phenylisoquinoline (4c)
on a Bruker AVANCE 300 NMR spectrometer (operating
at 300 MHz for 1H and 75 MHz for 13C). FTIR were
obtained using a universal attenuated total reflectance
attached on a Perkin–Elmer Spectrum One spectrometer.
Mass spectra were recorded on a Bruker Daltonics
(microTOF). Melting points were determined using a
Griffin melting point apparatus and were uncorrected.
92 % yield. mp: 122–123 °C. 1H NMR (300 MHz,
CDCl3): d (ppm) 2.36 (s, 3H, CH3), 2.40–2.65 (m, 2H, C4),
3.15–3.30, 3.71–3.90 (m, 2H, C3), 3.78, 3.84 (2 s, 6H,
2 9 OCH3), 6.19 (s, 1H, C1), 6.46 (s, 2H, C5-ArH, C8-
ArH), 7.13 (d, J = 8.0 Hz, 2H, C300-ArH2), 7.19–7.31 (m,
123

Cytotoxicity of N-tosyl-1,2,3,4-tetrahydroisoquinolines
5H, C20-ArH2, C30-ArH2, C40-ArH), 7.58 (d, J = 8.2 Hz,
2H, C200-ArH2); 13C NMR (75 MHz, CDCl3): d (ppm)
21.4, 26.1, 38.7, 55.8, 55.9, 58.8, 110.7, 111.2, 125.6,
126.0, 127.0, 127.6, 128.2, 128.8, 129.3, 137.9, 141.5,
143.0, 147.5, 148.2. IR (UATR) cm-1: 1611, 1516, 1336,
1157, 1116. TOF-MS m/z: 424.1583 (Calcd for
C24H26NO4S: 424.1577).
(UATR) cm-1: 1590, 1518, 1323, 1158, 1115. TOF-MS
m/z: 514.1912 (Calcd for C27H32NO7S: 514.1894).
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(4-hydroxyphenyl)-isoquinoline
(4g)
34 % yield. mp: 146–147 °C. 1H NMR (300 MHz,
CDCl3): d (ppm) 2.31 (s, 3H, CH3), 2.25–2.55 (m, 2H, C4),
3.10–3.22, 3.65–3.80 (m, 2H, C3), 3.73, 3.78 (2 s, 6H,
2 9 OCH3), 6.10 (s, 1H, C1), 6.39, 6.40 (2 s, 2H, C5-ArH,
C8-ArH), 6.70 (d, J = 8.5 Hz, 2H, C30-ArH2), 7.02 (d,
J = 8.5 Hz, 2H, C20-ArH2), 7.08 (d, J = 8.1 Hz, 2H, C300-
ArH2), 7.54 (d, J = 8.1 Hz, 2H, C200-ArH2); 13C NMR
(75 MHz, CDCl3): d (ppm) 21.4, 26.0, 38.4, 55.8, 56.0,
58.4, 110.7, 111.2, 115.1, 125.9, 127.0, 129.3, 130.2,
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(4-methoxyphenyl)-
isoquinoline (4d)
35 % yield. mp: 119–120 °C. 1H NMR (300 MHz, CDCl3):
d (ppm) 2.36 (s, 3H, CH3), 2.40–2.65 (m, 2H, C4), 3.12–3.30,
3.70–3.90 (m, 2H, C3), 3.78, 3.80, 3.84 (3 s, 9H,
3 9 OCH3), 6.16 (s, 1H, C1), 6.45 (s, 2H, C5-ArH, C8-ArH),
6.81 (d, J = 8.0 Hz, 2H, C300-ArH2), 7.13 (d, J = 7.7 Hz,
4H, C20-ArH2, C30-ArH2), 7.59 (d, J = 8.2 Hz, 2H, C200-
ArH2); 13C NMR (75 MHz, CDCl3): d (ppm) 21.4, 26.1,
38.5, 55.3, 55.8, 55.9, 58.3, 110.7, 111.2, 113.5, 125.9, 127.0,
129.3, 130.0, 133.8, 138.0, 142.9, 147.5, 148.1, 159.1. IR
(UATR) cm-1: 1609, 1509, 1336, 1157, 1118. TOF-MS m/z:
454.1694 (Calcd for C25H28NO5S: 454.1683).
133.5, 137.9, 143.1, 147.4, 148.1, 155.4. IR (UATR) cm-1
:
3428, 1611, 1512, 1320, 1154, 1116. TOF-MS m/z:
440.1538 (Calcd for C24H26NO5S: 440.1532).
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(4-hydroxy-3-methoxyphenyl)-
isoquinoline (4h)
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(3-hydroxy-4-methoxyphenyl)-
isoquinoline (4e)
57 % yield. mp: 125–126 °C. 1H NMR (300 MHz,
CDCl3): d (ppm) 2.35 (s, 3H, CH3), 2.40–2.64 (m, 2H, C4),
3.15–3.30, 3.80–3.90 (m, 2H, C3), 3.78, 3.83, 3.84 (3 s,
9H, 3 9 OCH3), 6.11 (s, 1H, C1), 6.42–6.48 (m, 3H, C5-
ArH, C8-ArH, C60-ArH), 6.76 (d, J = 8.2 Hz, 1H, C50-
ArH), 6.92 (d, J = 1.7 Hz, 2H, C20-ArH), 7.13 (d,
J = 8.2 Hz, 2H, C300-ArH2), 7.59 (d, J = 8.2 Hz, 2H, C200-
ArH2); 13C NMR (75 MHz, CDCl3): d (ppm) 21.4, 26.1,
38.5, 55.8, 55.9, 58.7, 110.7, 111.1, 111.5, 113.5, 121.8,
125.7, 126.0, 127.0, 129.3, 133.5, 138.1, 143.0, 145.2,
146.5, 147.4, 148.1. IR (UATR) cm-1: 3432, 1611, 1513,
1321, 1156. TOF-MS m/z: 470.1636 (Calcd for
C25H28NO6S: 470.1632).
54 % yield. mp: 171–172 °C. 1H NMR (300 MHz, CDCl3): d
(ppm) 2.30 (s, 3H, CH3), 2.00–2.60 (m, 2H, C4), 3.12–3.30,
3.70–3.90 (m, 2H, C3), 3.73, 3.78, 3.83 (3 s, 9H, 3 9 OCH3),
6.06 (s, 1H, C1), 6.39, 6.40 (2 s, 2H, C5-ArH, C8-ArH),
6.63–6.76 (m, 3H, C20-ArH, C50-ArH, C60-ArH), 7.07 (d,
J = 8.1 Hz, 2H, C300-ArH2), 7.54 (d, J = 8.1 Hz, 2H, C200-
ArH2). 13C NMR (75 MHz, CDCl3): d (ppm) 21.4, 26.1, 38.6,
55.8, 55.9, 56.0, 58.4, 110.2, 110.7, 111.2, 115.1, 120.8,
125.9, 127.0, 129.2, 134.9, 138.0, 142.9, 145.2, 146.0, 147.5,
148.1. IR(UATR)cm-1:3433, 1596, 1509,1318,1157, 1113.
TOF-MS m/z: 470.1629 (Calcd for C25H28NO6S: 470.1632).
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(4-hydroxy-3,5-
dimethoxyphenyl)-isoquinoline (4i)
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(3,4,5-trimethoxyphenyl)-
isoquinoline (4f)
62 % yield. mp: 148–149 °C. 1H NMR (300 MHz,
CDCl3): d (ppm) 2.31 (s, 3H, CH3), 2.38–2.65 (m, 2H, C4),
3.15–3.28, 3.70–3.85 (m, 2H, C3), 3.72, 3.74, 3.80 (3 s,
6H, 4 9 OCH3), 5.48 (s, 1H, C1), 6.35 (s, 2H, C20-ArH2),
6.41, 6.42 (s, 2H, C5-ArH, C8-ArH), 7.10 (d, J = 8.2 Hz,
2H, C300-ArH2), 7.55 (d, J = 8.2 Hz, 2H, C200-ArH2). 13C
NMR (75 MHz, CDCl3): d (ppm) 21.4, 26.3, 38.7, 55.8,
56.0, 56.3, 58.9, 105.9, 110.8, 111.1, 125.6, 126.1, 127.0,
129.3, 132.7, 134.3, 138.1, 143.0, 146.7, 147.4, 148.1. IR
(UATR) cm-1: 3435, 1612, 1515, 1319, 1156, 1110. TOF-
MS m/z: 500.1729 (Calcd for C26H30NO7S: 500.1738).
85 % yield. mp: 174–175 °C. 1H NMR (300 MHz,
CDCl3): d (ppm) 2.31 (s, 3H, CH3), 2.38–2.65 (m, 2H, C4),
3.15–3.30, 3.70–3.90 (m, 2H, C3), 3.69, 3.76, 3.80, 3.81
(4 s, 15H, 5 9 OCH3), 6.05 (s, 1H, C1), 6.36, 6.42, 6.44
(3 s, 4H, C5-ArH, C8-ArH, C20-ArH2), 7.10 (d, J = 8.0 Hz,
2H, C300-ArH2), 7.56 (d, J = 8.2 Hz, 2H, C200-ArH2); 13C
NMR (75 MHz, CDCl3): d (ppm) 21.4, 26.3, 38.9, 55.8,
56.0, 56.1, 58.9, 60.8, 106.2, 110.8, 111.2, 125.4, 126.1,
127.0, 129.3, 137.2, 138.1, 143.0, 147.4, 148.2, 152.9. IR
123

R. Pingaew et al.
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(4-bromophenyl)-
isoquinoline (4j)
Cytotoxic assay
The cytotoxic assay was performed as previously described
by Tengchaisri and co-workers (Tengchaisri et al., 1998).
Briefly, cell lines suspended in RPMI 1640 containing
10 % fetal bovine serum (FBS) were seeded at 1 9 104
cells (100 lL) per well in a 96-well plate, and incubated in
humidified atmosphere, 95 % air, 5 % CO2 at 37 °C. After
24 h, additional medium (100 lL) containing the test
compound and vehicle was added to a final concentration
of 50 lg/mL, 0.2 % dimethyl sulfoxide (DMSO), and
further incubated for 3 days. After that, the cells were fixed
with EtOH–H2O (95:5, v/v), stained with crystal violet
solution, and lysed with a solution of 0.1 N HCl in MeOH.
The absorbance was measured at 550 nm. The number of
surviving cells was determined from the absorbance. Eto-
poside and/or doxorubicin were used as the reference
drugs.
84 % yield. mp: 120–121 °C. 1H NMR (300 MHz,
CDCl3): d (ppm) 2.32 (s, 3H, CH3), 2.30–2.60 (m, 2H, C4),
3.05–3.20, 3.65–3.85 (m, 2H, C3), 3.74, 3.80 (2 s, 6H,
2 9 OCH3), 6.09 (s, 1H, C1), 6.37, 6.41 (2 s, 2H, C5-ArH,
C8-ArH), 7.06 (d, J = 8.5 Hz, 2H, C20-ArH2), 7.09 (d,
J = 8.5 Hz, 2H, C30-ArH2), 7.37 (d, J = 8.0 Hz, 2H, C300-
ArH2), 7.54 (d, J = 8.2 Hz, 2H, C200-ArH2). 13C NMR
(75 MHz, CDCl3): d (ppm) 21.5, 26.1, 38.8, 55.8, 55.9,
58.2, 110.5, 111.2, 121.8, 125.0, 126.0, 127.0, 129.4,
130.5, 131.4, 137.7, 140.7, 143.2, 147.6, 148.3. IR (UATR)
cm-1: 1611, 1516, 1339, 1158, 1117. IR (UATR) cm-1
:
1611, 1516, 1339, 1158, 1117. TOF-MS m/z: 502.0676
(Calcd for C24H25BrNO4S: 502.0682).
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(2-hydroxyphenylisoquinoline
(4k)
Computational analysis
Data set
52 % yield. mp: 96–97 °C. 1H NMR (300 MHz, CDCl3): d
(ppm) 2.30 (s, 3H, CH3), 2.30–2.60 (m, 2H, C4),
3.17–3.32, 3.77–3.92 (m, 2H, C3), 3.75, 3.85 (2 s, 6H,
2 9 OCH3), 6.27 (s, 1H, C1), 6.32, 6.33 (2 s, 2H, C5-ArH,
C8-ArH), 6.46 (d, J = 7.6 Hz, 1H, C30-ArH), 6.70 (t,
J = 7.6 Hz, 1H, C50-ArH), 7.05 (d, J = 7.6 Hz, 1H, C60-
ArH), 7.08 (d, J = 8.2 Hz, 2H, C300-ArH2), 7.20 (t,
J = 7.6 Hz, 1H, C40-ArH), 7.61 (d, J = 8.2 Hz, 2H, C200-
ArH2). 13C NMR (75 MHz, CDCl3): d (ppm) 21.5, 25.1,
38.4, 54.7, 55.8, 55.9, 110.3, 110.9, 117.9, 119.7, 124.2,
125.7, 126.4, 127.1, 129.5, 129.9, 130.0, 136.4, 143.9,
147.7, 148.3, 155.4. IR (UATR) cm-1: 3405, 1611, 1517,
1322, 1151, 1119. TOF-MS m/z: 440.1531 (Calcd for
C24H26NO5S: 440.1526).
The IC50 (lM) values of cytotoxicity against HepG2 and
MOLT-3 cell lines were converted to pIC50 by taking the
negative logarithm to the base of 10 of the IC50 values. The
molecular structures of the 12 compounds investigated in
this study are presented in Fig. 1.
Quantum chemical and molecular descriptors calculation
Molecular structures of the compounds were drawn using
GaussView (Dennington II et al. 2003). Geometric opti-
mization was initially performed at the semi-empirical
method AM1 (Austin Model 1) and finally optimized at the
DFT level using Becke’s three-parameter hybrid method
using the Lee–Yang–Parr correlation functional (B3LYP)
together with the 6–31 g(d) basis set in the Gaussian 03 W
package (Frisch et al. 2004). Quantum chemical descriptors
were obtained from the optimized structure, which inclu-
ded: total energy (Etotal), highest occupied molecular orbital
energy (EHOMO), lowest unoccupied molecular orbital
energy (ELUMO), total dipole moment (l) of the molecule,
electron affinity (EA), the ionization potential (IP), energy
difference of HOMO and LUMO (EHOMO - ELUMO),
Mulliken electronegativity (v), hardness (g), electrophilic-
ity (x), softness (S), electrophilic index (xi), highest neg-
ative charge (Q-), highest positive charge (Q?) and mean
absolute atomic charge (Qm) (Karelson et al. 1996; Parr
et al. 1978, 1999; Parr and Pearson 1983; Thanikaivelan
et al. 2000). Furthermore, an additional set of 3,224
descriptors were obtained from the Dragon software (Talete
2007) using the optimized structure as input data. The
1,2,3,4-Tetrahydro-6,7-dimethoxy-2-[(4-
methylphenyl)sulfonyl]-1-(3-pyridyl)-isoquinoline (4l)
94 % yield. 1H NMR (300 MHz, CDCl3): d (ppm) 2.31 (s,
3H, CH3), 2.40–2.60 (m, 2H, C4), 3.05–3.20, 3.80–3.90
(m, 2H, C3), 3.71, 3.78 (2 s, 6H, 2 9 OCH3), 6.17 (s, 1H,
C1), 6.36, 6.44 (2 s, 2H, C5-ArH, C8-ArH), 7.12 (d,
J = 7.9 Hz, 2H, C300-ArH2), 7.45–7.60 (m, 3H, C50-ArH,
C200-ArH2), 7.97 (d, J = 7.7 Hz, 1H, ArH), 8.35 (s, 1H,
ArH), 8.63 (d, J = 4.1 Hz, 1H, ArH). 13C NMR (75 MHz,
CDCl3): d (ppm) 21.4, 26.0, 39.5, 55.9, 56.0, 56.5, 110.0,
111.6, 122.6, 125.5, 126.2, 127.0, 129.7, 136.7, 140.8,
141.8, 143.8, 143.9, 144.0, 148.1, 149.0. IR (UATR) cm-1
:
1688, 1611, 1517, 1338, 1158, 1115. TOF-MS m/z:
425.1536 (Calcd for C23H25N2O4S: 425.1530).
123

Cytotoxicity of N-tosyl-1,2,3,4-tetrahydroisoquinolines
Scheme 1 Synthesis of
N-sulfonyl-1,2,3,4-
tetrahydroisoquinolines (4a–4l)
MeO
MeO
MeO
MeO
Me
O O
S
Cl
a
+
O
S
Me
NH2
HN
1
2
O
3
b
5
8
4
6
7
MeO
MeO
Me
3
O
S
N
1
R
O
4
Reagents and conditions: (a) Na2CO3, CH2Cl2
(b) paraformaldehyde, HCOOH, reflux for 4a; RCHO, TFA, reflux for 4b-4l
Fig. 1 Molecular structures of
N-sulfonyl-1,2,3,4-
tetrahydroisoquinolines (4a–4l)
Me
MeO
MeO
Me
MeO
MeO
Me
MeO
MeO
O
S
O
S
O
S
N
N
N
N
O
O
O
4a
4b
4c
MeO
MeO
Me
Me
MeO
MeO
MeO
MeO
Me
O
S
O
S
O
S
N
N
O
O
O
OH
MeO
OMe
OMe
OMe
OMe
4f
4d
4e
MeO
MeO
Me
MeO
MeO
Me
MeO
MeO
Me
O
S
O
S
O
S
N
N
N
O
O
O
OMe
MeO
OMe
OH
OH
OH
4g
4h
4i
MeO
MeO
Me
MeO
MeO
Me
MeO
MeO
Me
O
S
O
S
O
S
N
N
N
O
O
HO
O
N
Br
4k
4j
4l
molecular descriptors from Dragon can be classified into 22
categories: 48 Constitutional descriptors, 119 Topological
descriptors, 47 Walk and path counts, 33 Connectivity
indices, 47 Information indices, 96 2D autocorrelation, 107
Edge adjacency indices, 64 Burden eigenvalues, 21 Topo-
logical charge indices, 44 Eigenvalue-based indices, 41
Randic molecular profiles, 74 Geometrical descriptors, 150
RDF descriptors, 160 3D-MoRSE descriptors, 99 WHIM
descriptors, 197 GETAWAY descriptors, 154 Functional
group counts, 120 Atom-centred fragments, 14 Charge
descriptors, 29 Molecular properties, 780 2D binary fin-
gerprints and 780 2D frequency fingerprints.
Multivariate analysis
The independent variables are comprised of quantum
chemical and molecular descriptors while the dependent
variables are the cytotoxicity pIC50 values. Multivariate
123

R. Pingaew et al.
analysis was performed using multiple linear regression
(MLR) using Waikato Environment for Knowledge Anal-
ysis (Weka), version 3.4.5. (Witten et al. 2011).
the model (Q2), F value and critical F value (Alves et al.
2001; Nantasenamat et al. 2008a, 2009, 2010).
Multiple linear regression
Results and discussion
The MLR models were calculated to obtain the following
Chemistry
equation:
X
Y ¼ B0 þ
BnXn
ð1Þ
4-Methyl-N-(2-(3,4-dimethoxyphenyl)ethyl)-benzenesul-
fonamide 3 was readily prepared by the condensation of
homoveratrylamine 1 with p-toluenesulfonyl chloride 2 in
methylene chloride catalyzed by sodium carbonate as shown
in Scheme 1 (Pingaew et al. 2013). With the availability of
the starting benzenesulfonamide 3 in hand, we then inves-
tigated the condensation of key sulfonamide 3 with various
where Y is the pIC50 values of the compounds, B0 is the
intercept, and Bn are the regression coefficients of the
descriptors Xn.
Data sampling
¨
aldehydes under Bronsted acid condition (i.e. HCOOH,
Leave-one-out cross-validation (LOO-CV) was used as a
data sampling approach to separate the data set as a
training set and testing set. LOO-CV is a practical and
reliable method for testing significance. Briefly, one data
sample was left out and used as the testing set and the
remaining N - 1 samples were used as the training set. In
this respect, each member of the data set had a chance to be
used as the testing set to predict the dependent variable (Y).
CF3COOH). The reactions readily furnished the cyclization
product, N-tosyl-1,2,3,4-tetrahydroisoquinoline analogs (4)
in moderate to good yields (34–94 %) as shown in Table 1.
The results showed that the synthesis of N-sulfonyl-1,2,3,4-
tetrahydroisoquinolines depended on the reactivity of alde-
hyde. The more reactive aldehyde such as formaldehyde,
could be obtained in mild formic acid (entry 1) instead of the
strong acid, TFA, which was employed in the cases of other
aldehydes (entries 2–12). The formation of the desired iso-
Statistical analysis
1
quinolines (4) was confirmed by H NMR spectra which
showed the absence of the NH proton, while the singlet of
methine proton at d in the range of 5–7 ppm was displayed
indicating that the cyclized products (4c–4l) were formed. In
addition, the singlet of methylene proton at d 4.14 ppm and
the quartet of methine proton at d 5.03 ppm were observed
Sample outliers were identified using 2 standardized
residual as the cutoff criteria where values exceeding the
cutoff will be removed. Standardization of the residuals
(difference of actual and predicted values) was calculated
according to the following equation:
xij ꢁ xꢀ
j
stn
ij
x
¼
ð2Þ
P
N
i¼1
2
ðxij ꢁ xꢀ Þ =N
j
Table 1 Chemical yield of the N-sulfonyl-1,2,3,4-tetrahydroiso-
quinolines (4a–4l)
stn
where x is the standardized value, xij is the value of each
ij
sample, xꢀ is the mean of each descriptors and N is the
j
Entry
Isoquinoline
R
Time
(h)
Yield
(%)
sample size of the data set.
Calculated descriptors from Dragon that are constant or
near-constant were implied to be redundant and were,
therefore, removed using the Unsupervised Forward
Selection algorithm of UFS, version 1.8. The UFS algo-
rithm was previously described by Whitley et al. (2000)
and Nantasenamat et al. (2005). Furthermore, significant
descriptors were selected using stepwise regression as
performed by SPSS Statistics 18.0 (SPSS Inc., USA). Inter-
correlation among the descriptors was calculated using
Pearson’s correlation coefficient as implemented in SPSS
Statistics 18.0 (SPSS Inc., USA).
1a
2
4a
4b
4c
4d
4e
4f
H
1
85
89
92
35
54
85
34
57
62
84
52
94
Me
0.5
2
3
C6H5
4
4-OMe-C6H4
3-OH,4-OMe-C6H3
3,4,5-(OMe)3-C6H3
4-OH-C6H4
3-OMe,4-OH-C6H3
3,5-(OMe)2,4-OH-C6H3
4-Br-C6H4
27
5
5
6
12
27
5
7
4g
4h
4i
8
9
8
10
11
12
4j
2
4k
4l
2-OH-C6H4
3-C5H4N
27
5
Evaluation of the predictive ability of the QSAR models
were assessed using the following statistical parameters:
correlation coefficient (r), square of correlation coefficient
(R2), root mean square error (RMSE), predictive power of
TFA, reflux for 4b–4l
a
HCOOH, reflux for 4a
123

Cytotoxicity of N-tosyl-1,2,3,4-tetrahydroisoquinolines
for 1-unsubstituted isoquinoline 4a and 1-methylisoquino-
line 4b, respectively. The structures of all obtained com-
pounds were further supported by 13C NMR, 2D NMR, IR as
well as HRMS.
enhances its absorption to cells. Comparable result was
also observed for heterocyclic pyridyl group at C-1 as seen
for analog 4l. Among the tested analogs, p-methoxy com-
pound (4d, R = p-OMeC6H4) was shown to be an inactive
cytotoxic (IC50 [ 110.09 lM). Remarkable cytotoxicity
was found when R displayed the p-OHC6H4 (4g,
IC50 = 20.06 lM) and p-BrC6H4 (4j, IC50 = 22.69 lM)
substituents. However, unsubstituted phenyl analog (4c)
exerted higher cytotoxic effect (IC50 15.49 lM) than the
substituted p-OH (4g) and p-Br (4j) compounds. Intro-
duction of di-OMe groups to 3- and 5-positions of the
analog 4g furnished compound 4i with lower cytotoxicity
(IC50 40.61 lM) than the parent compound 4g. It was
notable that 3,4,5-trimethoxy analog 4f exhibited greater
activity (IC50 9.24 lM) than the trisubstituted 3,5-dime-
thoxy-4-hydroxyphenyl analog 4i. It is likely that the tri-
methoxy analog 4f bears appropriate lipophilic moiety for
cell absorption when compared to 3,4,5- trisubstituted
phenyl analog 4i. It was also observed that substituents at
3- and 3,5-positions of 4-hydroxyphenyl analog 4g pro-
vided analogs 4h and 4i, respectively, with decreasing
cytotoxicities. Conversely, the inactive 4-OMe analog (4d)
was changed to be active compounds 4e and 4f when the
corresponding 3- and 3,5-substituents were placed on the
phenyl group of 4d. Significant results were observed when
the OH group was substituted on the ortho-position of the
phenyl group as seen for the most potent analog 4k.
Obviously, the o-OH analog 4k had 16-fold higher
Cytotoxic activity
For an assessment of the cytotoxicity (Tengchaisri et al.
1998), the synthesized isoquinoline analogs 4a–4l (Fig. 1)
were evaluated against HuCCA-1, HepG2, A549 and
MOLT-3 human cancer cell lines as summarized in
Table 2. All of the tested compounds displayed inhibition
activities against MOLT-3 cell lines, except for p-methoxy
analog 4d. The analog 4d also exhibited no cytotoxicity
against other tested cell lines. All the investigated cells
were inhibited by analogs 4k and 4l. Interestingly, the
analog 4k was shown to be the most potent cytotoxic
against HuCCA-1, A549 and MOLT-3 cells. In particular,
compound 4k had the lowest IC50 value of 1.23 lM against
MOLT-3 cell. On the other hand, trimethoxy analog 4f was
the most potent compound (IC50 = 22.70 lM) against
HepG2 cell and was stronger than the reference drug,
etoposide. These cytotoxics contain the common core
structure of N-tosyl-1,2,3,4-tetrahydroisoquinoline, but
with different substituents (R) at the C-1 position. It was
noted that high cytotoxicity (MOLT-3) was achieved when
R was aromatic (phenyl) as compared to alkyl (CH3) and H
substituents. This could be due to hydrophobic effect that
Table 2 Cytotoxic activity of compounds 4a–4l against four cancer cell lines
Compound
Cytotoxic activity (IC50, lM)a
HuCCA-1
HepG2
A549
MOLT-3
4a
[143.62
[124.26
[117.88
[110.09
[106.35
[97.24
[143.62
[143.62
[124.26
[117.88
[110.09
[106.35
[97.24
65.58 0.83
55.81 0.48
15.49 1.08
[110.09
4b
111.83 7.07
[117.88
4c
4d
[110.09
4e
[106.35
36.67 0.42
9.24 0.23
20.06 0.21
41.20 0.81
40.61 0.28
22.69 0.05
1.23 0.07
12.42 0.33
0.041 0.004
ND
4f
22.70 0.58
79.52 2.65
85.08 7.07
79.97 9.50
[99.59
4g
[113.60
[106.35
[99.97
[113.60
[106.35
[99.97
4h
4i
4j
[99.59
[99.59
4k
43.17 2.82
72.92 1.41
ND
44.69 4.72
49.39 3.60
27.18 5.29
0.42 0.03
56.80 5.46
117.61 1.41
ND
4l
Etoposideb
Doxorubicinb
0.64 0.21
0.70 0.03
Cancer cell lines are HuCCA-1 human cholangiocarcinoma cancer cells, HepG2 human hepatocellular carcinoma cell line, A549 human lung
carcinoma cell line and MOLT-3 human lymphoblastic leukemia cell line
ND not determined
a
The assays were performed in triplicate
b
Etoposide and/or doxorubicin were used as the reference drugs
123

R. Pingaew et al.
cytotoxicity than the p-OH compound 4g. This could be
assumed that o-OHC6H4 at C-1 is very important for the
cytotoxic activity against MOLT-3 cells. It is reasonable to
explain that OH group at the o-position may participate in
the formation of intramolecular H-bond with the N-sulfo-
nyl group of the target molecule 4k.
method to further select important descriptors using the
SPSS software.
The significant descriptors for predicting the cytotoxic
activity against MOLT-3 cell lines as obtained from step-
wise MLR are displayed in Table 3. Two descriptors such
as Mor32u (3D-MoRSE—signal 32/unweighted) and Gu
(G total symmetry index/unweighted) were found to be
responsible for cytotoxic activity against MOLT-3 cell.
The descriptors were used to construct QSAR model using
MLR according to Eq. (1). The QSAR equation (model 1)
and model evaluations are displayed in Table 4. It was
found that the r and RMS values obtained from the training
set were 0.8899 and 0.2229, respectively, while the r and
RMS values of the LOO-CV testing set were 0.7541 and
0.3247, respectively. Their predicted values of cytotoxicity
are presented in Table 3.
QSAR modeling
Compounds that act as cytotoxics against cancer cell lines,
particularly, MOLT-3 and HepG2 were selected for the
study. Molecular structures of the compounds were con-
structed and optimized for calculating quantum chemical
and molecular descriptors using Gaussian and Dragon
software, respectively. Successful utilization of these
descriptors in the development of QSAR models have
previously been reported (Nantasenamat et al. 2007a, b,
2008b; Prachayasittikul et al. 2010; Suksrichavalit et al.
2009; Suvannang et al. 2011; Worachartcheewan et al.
2009, 2011, 2012). Descriptors were generated using
DRAGON to yield 3,224 variables where constant or near-
constant descriptors were removed to obtain a set of 1,292
descriptors. The Dragon descriptor together with the set of
15 descriptors from Gaussian software were standardized
according to Eq. (2) to further remove redundant descrip-
tors using the UFS software. This resulted in a set of 11
descriptors which comprises of (i) Molecular properties:
GVWAI-80, (ii) 2D autocorrelation indices: GATS6v, (iii)
Geometrical descriptors: PJI3, (iv) WHIM descriptors: Gu,
(v) 2D frequency fingerprints: F06[N–O], (vi) Atom-cen-
tred fragments: C-024 (vii) GETAWAY descriptors: R5e?
and R4u?, (viii) 3D-MoRSE descriptors: Mor31v and
Mor32u and (ix) quantum chemical descriptors: LUMO.
These descriptors were then subjected to stepwise MLR
The results showed that compounds (4g, 4j, 4k) with
high Gu values had high cytotoxic activity against MOLT-
3 cells. The analog 4k had the highest Gu value of 0.192
and was found to be the most potent compound. On the
other hand, the inactive compound 4d had the lowest Gu
value (0.167). It was observed that trimethoxy analog 4f
had the highest Mor32u value (-0.025) while exhibiting
high cytotoxicity (IC50 9.24 lM) against MOLT-3,
albeit at a relatively lower level than compound 4k
(IC50 1.23 lM). Recently, it has been reported that the
cytotoxicity of thiopyridines against MOLT-3 showed
good correlation with atomic mass descriptor, GATS4 m
(Worachartcheewan et al. 2011b).
Gu is a descriptor accounting for the G total symmetry
´
index/unweighted (Gonzalez et al. 2007b; Zhang et al.
2009). This descriptor involves the information, shape and
other physicochemical properties of a molecular structure.
In this study, the highest Gu value was observed in com-
pound 4k which could be attributed to an intramolecular
hydrogen bonding of SO2 group and o-OHC6H4 group at
the C-1 position. Such interaction possibly arranges the
molecule to be in a highly symmetry position as compared
to the p-OH analog 4g. Mor32u and Mor31v are 3D-
MoRSE (3D Molecular Representation of structures based
on Electron diffraction) (Gosav et al. 2007; Habibi-Yan-
gjeh et al. 2009) descriptors that are related to the atomic
properties (e.g. atomic van der Waals volumes) to the
cytotoxic effect. The QSAR studies demonstrated that tri-
methoxy analog 4f, which has the highest molecular weight
also had the highest Mor32u value. The results suggested
that the highest inhibition against MOLT-3 cells required
the molecule to have high symmetry index (Gu) which
would correspondingly lead to appropriate interaction with
Table 3 The important descriptors and predicted cytotoxic activity
of compounds 4a–4l against MOLT-3 cell lines
Compound Gu
Mor32u Experimental
activity (pIC50
Predicted
activity (pIC50)
)
4a
4b
4c
4d
4e
4f
0.171 -0.211
0.177 -0.305
-1.817
-1.559
-1.374
-1.813
-1.963
-1.462
-0.832
-1.214
-1.791
-1.916
-1.046
-0.727
-1.120
-1.747
-1.190
-2.042
-1.564
-0.966
-1.302
-1.615
-1.609
-1.356
-0.089
-1.094
0.169 -0.171
0.167 -0.272
0.176 -0.305
0.176 -0.025
0.181 -0.343
0.176 -0.439
0.175 -0.447
0.183 -0.345
0.192 -0.265
0.179 -0.226
4g
4h
4i
´
the target site of action (Gonzalez et al. 2007b).
4j
The three descriptors including PJI3 (3D Petitjean shape
index), Mor32u and Mor31v (3D-MoRSE—signal
31/weighted by atomic van der Waals volumes), which
4k
4l
123

Cytotoxicity of N-tosyl-1,2,3,4-tetrahydroisoquinolines
Table 5 The important descriptors and predicted cytotoxic activity
of compounds 4a–4l against HepG2 cell lines
Compound PJI3 Mor32u Mor31v Experimental Predicted
activity
(pIC50
activity
(pIC50)
)
4a
4b
4c
4d
4e
4f
0.854 -0.211 0.269
0.900 -0.305 0.352
0.859 -0.171 0.278
0.868 -0.272 0.369
0.866 -0.305 0.227
0.869 -0.025 0.223
0.869 -0.343 0.344
0.873 -0.439 0.258
0.867 -0.447 0.344
0.867 -0.345 0.293
0.863 -0.265 0.28
0.859 -0.226 0.396
[-2.157
-2.094
–
-2.218
–
[-2.071
[-2.042
[-2.027
-1.356
–
–
-1.448
-1.845
-1.880
-1.988
–
4g
4h
4i
-1.901
-1.930
-1.903
4j
[-1.998
-1.650
4k
4l
-1.646
-1.676
-1.694
were obtained from stepwise MLR, were important vari-
ables for predicting the cytotoxicity against HepG2 as
displayed in Table 5. The descriptors used in the MLR
analysis to obtain the QSAR equation (model 2) are shown
in Table 4. It was observed that five compounds (4a, 4c–4e
and 4j) were inactive against HepG2 cell lines, thus seven
active compounds (4b, 4f–4i and 4k-4l) were used to
perform the MLR. We found that the r and RMS values
obtained for the training set were 0.9924 and 0.0277,
respectively, while the r and RMS values for the LOO-CV
testing set were 0.9563 and 0.0726, respectively. Their
predicted cytotoxic activity against HepG2 is shown in
Table 5.
It was found that the most potent compound 4f that
inhibited HepG2 cells had the highest Mor32u value of
-0.025, but had the lowest Mor31v value of 0.223. All the
investigated compounds had comparably high PJI3 values
(0.854–0.900), particularly, the 1-methyl analog 4b
(R = CH3) had the highest PJI3 value of 0.900. In the case
of compound 4a (R = H), it had the lowest PJI3 value of
0.854.
A compound with high molecular weight tends to have
high hydrophobic property that enhances its cell penetra-
tion ability as in the case of compound 4f which acts as the
Table 6 Inter-correlation matrix of the three significant descriptors
for predicting the pIC50 values of HepG2
Descriptors
PJI3
Mor32u
Mor31v
PJI3
1
Mor32u
Mor31v
-0.1285
0.0569
1
-0.3440
1
123

R. Pingaew et al.
-.2
-.4
-1.2
-1.4
-1.6
-1.8
-2.0
-2.2
-2.4
(a)
(b)
-.6
-.8
-1.0
-1.2
-1.4
-1.6
-1.8
-2.0
-2.2
-2.5
-2.0
-1.5
-1.0
-.5
0.0
-2.4
-2.2
-2.0
-1.8
-1.6
-1.4
-1.2
Experimental activity (MOLT-3 pIC
)
50
Experimental activity (HepG2 pIC
)
50
Fig. 2 Plot of the experimental and predicted anticancer activities
(pIC50) against MOLT-3 (a) and HepG2 (b) cell lines for the training
set (open square regression line is represented as dotted line) and the
leave-one-out cross-validated testing set (filled square regression line
is represented as solid line)
most potent cytotoxic against HepG2. On the other hand,
the analog 4f had the lowest atomic van der Waals volumes
(Mor31v). This could be due to the fact that the molecule
had small size or volume which makes it appropriate for
interacting with target cells.
molecular structures with their cytotoxicities. The most
potent cytotoxic (4k) against MOLT-3 had the highest Gu,
correspondingly the inactive compound (4d) had the lowest
Gu value. On the other hand, the highest molecular mass
compound (4f) was shown to be the most potent cytotoxic
against HepG2 cells. The studies demonstrated that tetra-
hydroisoquinolines 4f and 4k are interesting potential lead
pharmacophores that should be further explored. Further-
more, the obtained QSAR models are useful in the design
and synthesis of novel bioactive compounds.
To evaluate the inter-correlation of the molecular
descriptors, the Pearson’s correlation coefficient was cal-
culated in SPSS. The results showed that low inter-corre-
lation value of -0.166 were observed for descriptors from
the QSAR model of MOLT-3. Descriptors from the QSAR
model of HepG2 also presented low inter-correlation as
displayed in Table 6. Outlier compounds were not detected
when using the 2 standardized residual as a cutoff criteria
according to Eq. (2). The performance of the QSAR model
gave R2 and Q2 values of 0.7919 and 0.5687, respectively,
for MOLT-3 and 0.9849 and 0.9145, respectively, for
HepG2 (Table 4). Such values indicated that the models
provided good predictive performance as the R2 values
were greater than 0.6 and the Q2 values were higher than
0.5 (Nantasenamat et al. 2009, 2010). Plots of the experi-
mental versus predicted activities of the compounds as
anticancer agents against MOLT-3 and HepG2 cell lines
are shown in Fig. 2a, b.
Acknowledgments We gratefully acknowledge the research grant
supported by Srinakharinwirot University (B.E. 2555). This project is
supported by Office of the Higher Education Commission and Mah-
idol University under the National Research Universities Initiative.
R.P. sincerely thanks the financial support from the Science and
Technology Research Grant of the Thailand Toray Science
Foundation.
References
Alves, C.N., J.C. Pinheiro, A.J. Camargo, M.M.C. Ferreira, R.A.F.
Romero, and A.B.F. da Silva. 2001. A multiple linear regression
and partial least squares study of flavonoid compounds with anti-
HIV activity. Journal of Molecular Structure (Theochem) 541:
81–88.
Barn, D.R., W.L. Caulfield, J. Cottney, K. McGurk, J.R. Morphy, Z.
Rankovic, and B. Roberts. 2001. Parallel synthesis and biolog-
ical activity of a new class of high affinity and selective d-opioid
ligand. Bioorganic & Medicinal Chemistry 9: 2609–2624.
Bentley, K.W. 1998. The isoquinoline alkaloids. Amsterdam:
Harwood Academic Publishers.
In conclusion, the synthesis of 1-substituted-N-tosyl-
1,2,3,4-tetrahydroisoquinoline analogs has been accom-
plished by the modified Pictet–Spengler reaction of phe-
nylethylbenzene sulfonamide with various aldehydes. Most
tetrahydroisoquinoline analogs displayed cytotoxicity
against MOLT-3 cell lines, in which the o-OH analog 4k is
the most potent compound. The trimethoxy analog 4f is the
most potent cytotoxic against HepG2 cells. QSAR studies
revealed that the total symmetry index (Gu), 3D-MoRSE
(Mor31v and Mor32u) and 3D Petitjean index (PJI3) were
the most significant descriptors for correlating the
´
Bermejo, A., I. Andreu, F. Suvire, S. Leonce, D.H. Caignard, P.
Renard, A. Pierre, R.D. Enriz, D. Cortes, and N. Cabedo. 2002.
´
Syntheses and antitumor targeting G1 phase of the cell cycle of
benzoyldihydroisoquinolines and related 1-substituted isoquino-
lines. Journal of Medicinal Chemistry 45: 5058–5068.
123

Cytotoxicity of N-tosyl-1,2,3,4-tetrahydroisoquinolines
Bourdier, T., G. Poisnel, M. Dhilly, J. Delamare, J. Henry, D.
´
Hazebroucq, G. 1966. 2,3,4,5-Tetrahydro-1H-3-benzazepin-1-ones
and hexahydroimidazoisoquinoleines. Annales de Chimie 1:
221–254.
Hu, L., Z.-R. Li, J.–.D. Jiang, and D.W. Boykin. 2008. Novel diaryl or
heterocyclic sulfonamides as antimitotic agents. Anti-Cancer
Agents. Medicinal Chemistry 8: 739–745.
Debruyne, and L. Barre. 2007. Synthesis and biological evalu-
ation of N-substituted quinolinimides, as potential ligands for
in vivo imaging studies of d-opioid receptors. Bioconjugate
Chemistry 18: 538–548.
Cox, E.D., and J.M. Cook. 1995. The Pictet–Spengler condensation:
A new direction for an old reaction. Chemical Reviews 95:
1797–1842.
Cui, W., K. Iwasa, H. Tokuda, A. Kashihara, Y. Mitani, T. Hasegawa,
Y. Nishiyama, M. Moriyasu, H. Nishino, M. Hanaoka, C. Mukai,
and K. Takeda. 2006. Potential cancer chemopreventive activity
of simple isoquinolines, 1-benzylisoquinolines, and protoberbe-
rines. Phytochemistry 67: 70–79.
´
´
¨
Huszar, J., Z. Timar, K.K. Szalai, G. Keseru, F. Fu¨lop, and B. Penke.
2008. Novel bradykinin-1 antagonists containing a (1,2,3,4-
tetrahydro-isoquinolin-1-yl)acetic acid scaffold. European Jour-
nal of Medicinal Chemistry 43: 1552–1558.
Ito, K., and H. Tanaka. 1977. Syntheses of 1,2,3,4-tetrahydroisoqui-
nolines from N-sulfonyl-phenethylamines and aldehydes. Chem-
ical & Pharmaceutical Bulletin 25: 1732–1739.
Dennington II, R., T. Keith, J. Millam, K. Eppinnett, W.L. Hovell,
and R. Gilliland. 2003. GaussView, Version 3.09. Shawnee
Mission: Semichem, Inc.
Iwasa, K., M. Moriyasu, Y. Tachibana, H.S. Kim, Y. Wataya, W.
Wiegrebe, K.F. Bastow, L.M. Cosentino, M. Kozuka, and K.H.
Lee. 2001. Simple isoquinoline and benzylisoquinoline alkaloids
as potential antimicrobial, antimalarial, cytotoxic, and anti-HIV
agents. Bioorganic & Medicinal Chemistry 9: 2871–2884.
Karelson, M., V.S. Lobanov, and A.R. Katritzky. 1996. Quantum-
chemical descriptors in QSAR/QSPR studies. Chemical Reviews
96: 1027–1044.
Larghi, E.L., M. Amongero, A.B.J. Bracca, and T.S. Kaufman.
2005. The intermolecular Pictet–Spengler condensation with
chiral carbonyl derivatives in the stereoselective syntheses of
optically active isoquinoline and indole alkaloids. ARKIVOC
12: 98–153.
Lukanov, L.K., A.P. Venkov, and N.M. Mollov. 1987. Application of
the intramolecular a-amidoalkylation reaction for the synthesis
of 2-arylsulfonyl-1,2,3,4-tetrahydroisoquinolines. Synthesis
1987: 204–206.
Ma, D., W. Wu, G. Yang, J. Li, J. Li, and Q. Ye. 2004.
Tetrahydroisoquinoline based sulfonamide hydroxamates as
potent matrix metalloproteinase inhibitors. Bioorganic & Medic-
inal Chemistry Letters 14: 47–50.
Frisch, M.J., G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, J.A. Montgomery Jr, T. Vreven, K.N. Kudin, J.C.
Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B.
Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene,
X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C.
Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev,
A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K.
Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G.
Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas,
D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V.
Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W.
Chen, M.W. Wong, C. Gonzalez, and J.A. Pople. 2004. Gaussian
03, Revision C.02. Wallingford: Gaussian, Inc.
Gitto, R., S. Agnello, S. Ferro, L. De Luca, D. Vullo, J. Brynda, P.
Mader, C.T. Supuran, and A. Chimirri. 2010. Identification of
3,4-dihydroisoquinoline-2(1H)-sulfonamide as potent carbonic
anhydrase inhibitors: Synthesis, biological evaluation, and
enzyme–ligand X-ray studies. Journal of Medicinal Chemistry
53: 2401–2408.
Matter, H., M. Schudok, W. Schwab, W. Thorwart, D. Barbier, G.
Billen, B. Haase, B. Neises, K.-U. Weithmann, and T.
Wollmann. 2002. Tetrahydroisoquinoline-3-carboxylate based
matrix-metalloproteinase inhibitors: Design, synthesis and struc-
ture–activity relationship. Bioorganic & Medicinal Chemistry
10: 3529–3544.
Gitto, R., S. Ferro, S. Agnello, L. De Luca, G. De Sarro, E. Russo, D.
Vullo, C.T. Supuran, and A. Chimirri. 2009. Synthesis and
evaluation of pharmacological profile of 1-aryl-6,7-dimethoxy-
Matter, H., and W. Schwab. 1999. Affinity and selectivity of matrix
metalloproteinase inhibitors: A chemometrical study from the
perspective of ligands and proteins. Journal of Medicinal
Chemistry 42: 4506–4523.
Nantasenamat, C., C. Isarankura-Na-Ayudhya, T. Naenna, and V.
Prachayasittikul. 2007a. Quantitative structure-imprinting factor
relationship of molecularly imprinted polymers. Biosensors &
Bioelectronics 22: 3309–3317.
Nantasenamat, C., C. Isarankura-Na-Ayudhya, T. Naenna, and V.
Prachayasittikul. 2008a. Prediction of bond dissociation enthalpy
of antioxidant phenols by support vector machine. Journal of
Molecular Graphics and Modelling 27: 188–196.
Nantasenamat, C., C. Isarankura-Na-Ayudhya, T. Naenna, and V.
Prachayasittikul. 2009. A practical overview of quantitative
structure–activity relationship. EXCLI Journal 8: 74–88.
Nantasenamat, C., C. Isarankura-Na-Ayudhya, and V. Prac-
hayasittikul. 2010. Advances in computational methods to
predict the biological activity of compounds. Expert Opinion
on Drug Discovery 5: 633–654.
3,4-dihydroisoquinoline-2(1H)-sulfonamides. Bioorganic
Medicinal Chemistry 17: 3659–3664.
&
´
Gonzalez, J.F., E. de la Cuesta, and C. Avendan˜o. 2007a. Synthesis
and cytotoxic activity of pyrazino[1,2-b]-isoquinolines, 1-(3-
isoquinolyl)isoquinolines, and 6,15-iminoisoquino[3,2-b]-3-ben-
zazocines. Bioorganic & Medicinal Chemistry 15: 112–118.
´
´
Gonzalez, M.P., C. Teran, M. Teijeira, and A.M. Helguera. 2007b.
QSAR studies using radial distribution function for predicting
A1 adenosine receptors agonists. Bulletin of Mathematical
Biology 69: 347–359.
Gosav, S., M. Praisler, and D.O. Dorohoi. 2007. ANN expert system
screening for illicit amphetamines using molecular descriptors.
Journal of Molecular Structure 834–836: 188–194.
Grunewald, G.L., F.A. Romero, and K.R. Criscione. 2005.
3-Hydroxymethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetra-
hydroisoquinoline inhibitors of phenylethanolamine N-methyl-
transferase that display remarkable potency and selectivity.
Journal of Medicinal Chemistry 48: 134–140.
Habibi-Yangjeh, A., E. Pourbasheer, and M. Danandeh-Jenagharad.
2009. Application of principal component-genetic algorithm-
artificial neural network for prediction acidity constant of
various nitrogen containing compounds in water. Monatshefte
fuer Chemie 140: 15–27.
Nantasenamat, C., C. Isarankura-Na-Ayudhya, N. Tansila, T. Naenna,
and V. Prachayasittikul. 2007b. Prediction of GFP spectral
properties using artificial neural network. Journal of Computa-
tional Chemistry 8: 1275–1289.
Nantasenamat, C., T. Naenna, C. Isarankura-Na-Ayudhya, and V.
Prachayasittikul. 2005. Quantitative prediction of imprinting
factor of molecularly imprinted polymers by artificial neural
123

R. Pingaew et al.
network. Journal of Computer-Aided Molecular Design 19:
509–524.
Synthesis of 1-thiosubstituted tetrahydroisoquinolines and car-
bon–carbon bond formation via sulfonyl iminium ions generated
from N,S-sulfonyl acetals. Tetrahedron Letters 44: 6137–6140.
Silveira, C.C., A.S. Vieira, and T.S. Kaufman. 2006. Thiophenol-
mediated improvement of the Pictet–Spengler cyclization of N-
tosyl-b-phenethylamines with aldehydes. Tetrahedron Letters
47: 7545–7549.
Suksrichavalit, T., S. Prachayasittikul, C. Nantasenamat, C. Isarank-
ura-Na-Ayudhya, and V. Prachayasittikul. 2009. Copper com-
plexes of pyridine derivatives with superoxide scavenging and
antimicrobial activities. European Journal of Medicinal Chem-
istry 44: 3259–3265.
Suvannang, N., C. Nantasenamat, C. Isarankura-Na-Ayudhya, and V.
Prachayasittikul. 2011. Molecular docking of aromatase inhib-
itors. Molecules 16: 3597–3617.
Talete srl. 2007. Dragon for Windows (Software for Molecular
Descriptor Calculations), version 5.5. Milano: Talete srl.
Tengchaisri, T., R. Chawengkirttikul, N. Rachaphaew, V. Reutrakul,
R. Sangsuwan, and S. Sirisinha. 1998. Antitumor activity of
triptolide against cholangiocarcinoma growth in vitro and in
hamsters. Cancer Letters 133: 169–175.
Nantasenamat, C., T. Piacham, T. Tantimongcolwat, T. Naenna, C.
Isarankura-Na-Ayudhya, and V. Prachayasittikul. 2008b. QSAR
model of the quorum-quenching N-acyl-homoserine lactone
lactonase activity. Journal of Biological Systems 16: 279–293.
Orazi, O.O., R.A. Corral, and H. Giaccio. 1986. Synthesis of fused
heterocycles: 1,2,3,4-tetrahydroisoquinoline ring homologues
via sulphonamidomethylation. Journal of the Chemical Society,
Perkin Transactions 1: 1977–1982.
Parr, R.G., R.A. Donnelly, M. Levy, and W.E. Palke. 1978.
Electronegativity: The density functional viewpoint. Journal of
Chemical Physics 68: 3801–3807.
Parr, R.G., and R.G. Pearson. 1983. Absolute hardness: Companion
parameter to absolute electronegativity. Journal of the American
Chemical Society 105: 7512–7516.
Parr, R.G., L.V. Szentpaly, and S. Liu. 1999. Electrophilicity index.
Journal of the American Chemical Society 121: 1922–1924.
Pingaew, R., S. Prachayasittikul, S. Ruchirawat, and V. Prac-
hayasittikul. 2013. Synthesis and cytotoxicity of novel N-
sulfonyl-1,2,3,4-tetrahydroisoquinoline
thiosemicarbazone
derivatives. Medicinal Chemistry Research 22: 267–277.
Prachayasittikul, S., O. Wongsawatkul, A. Worachartcheewan, C.
Nantasenamat, S. Ruchirawat, and V. Prachayasittikul. 2010.
Elucidating the structure–activity relationships of the vasorelax-
ation and antioxidation properties of thionicotinic acid deriva-
tives. Molecules 15: 198–214.
Saitoh, T., K. Abe, M. Ishikawa, M. Nakatani, S. Shimazu, N. Satoh,
F. Yoneda, K. Taguchi, and Y. Horiguchi. 2006. Synthesis and
in vitro cytotoxicity of 1,2,3,4-tetrahydroisoquinoline deriva-
tives. European Journal of Medicinal Chemistry 41: 241–252.
Scott, J.D., and R.M. Williams. 2002. Chemistry and biology of the
tetrahydroisoquinoline antitumor antibiotics. Chemical Reviews
102: 1669–1730.
Scozzafava, A., T. Owa, A. Mastrolorenzo, and C.T. Supuran. 2003.
Anticancer and antiviral sulfonamides. Current Medicinal
Chemistry 10: 925–953.
Siengalewicz, P., U. Rinner, and J. Mulzer. 2008. Recent progress in
the total synthesis of naphthyridnomycin and lemonomycin
tetrahydroisoquinoline antibiotics (TAAs). Chemical Society
Reviews 37: 2676–2690.
Thanikaivelan, P., V. Subramanian, J. Raghava Rao, and B. Unni
Nair. 2000. Application of quantum chemical descriptor in
quantitative structure activity and structure property relationship.
Chemical Physics Letters 323: 59–70.
Whitley, D.C., M.G. Ford, and D.J. Livingstone. 2000. Unsupervised
forward selection: A method for eliminating redundant variables.
Journal of Chemical Information and Computer Sciences 40:
1160–1168.
Witten, I.H., E. Frank, and M.A. Hall. 2011. Data mining: Practical
machine learning tools and techniques, 3rd ed. San Francisco:
Morgan Kaufmann.
Worachartcheewan, A., C. Nantasenamat, C. Isarankura-Na-Ayudhy-
a, S. Prachayasittikul, and V. Prachayasittikul. 2011. Predicting
the free radical scavenging activity of curcumin derivatives.
Chemometrics and Intelligent Laborary Systems 109: 207–216.
Worachartcheewan, A., C. Nantasenamat, T. Naenna, C. Isarankura-
Na-Ayudhya, and V. Prachayasittikul. 2009. Modeling the
activity of furin inhibitors using artificial neural network.
European Journal of Medicinal Chemistry 44: 1664–1673.
Worachartcheewan, A., S. Prachayasittikul, R. Pingaew, C. Nanta-
senamat, T. Tantimongcolwat, S. Ruchirawat, and V. Prac-
hayasittikul. 2012. Antioxidant, cytotoxicity, and QSAR study of
1-adamantylthio derivatives of 3-picoline and phenylpyridines.
Medicinal Chemistry Research 21: 3514–3522.
Silveira, C.C., C.R. Bernardi, A.L. Braga, and T.S. Kaufman. 1999.
Pictet–Spengler condensation of N-sulfonyl-b-phenethylamines
with a-chloro-a-phenylselenoesters. New synthesis of 1,2,3,4-
tetrahydroisoquinoline-1-carboxylates. Tetrahedron Letters 40:
4969–4972.
Zhang, H., Q.Y. Chen, M.L. Xiang, C.Y. Ma, Q. Huang, and S.Y.
Yang. 2009. In silico prediction of mitochondrial toxicity by
using GA-CG-SVM approach. Toxicology in Vitro 23: 134–140.
Silveira, C.C., C.R. Bernardi, A.L. Braga, and T.S. Kaufman. 2003.
Thioorthoesters in the activated Pictet–Spengler cyclization.
123