1312
A. K. Saluja et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1310–1314
Table 1
that compounds with small, compact substituents possess-
ing lower molecular weights, and an increased number of
hydrogen acceptor atoms were more effective against hCA
IX isoforms over hCA II (e.g., 6a, 4a–c, 6d, and 10–12). Again
the 1,3,5-triazine ring could be substituted by the isostruc-
tural pyridine, pyrimidine or diazine rings without loss of
hCA IX inhibitory activity or selectivity for inhibiting the
tumor-associated over the cytosolic isoforms (Table 1).
(iv) hCAXII was also potently inhibited by the new compounds
reported here, with inhibition constants ranging from 2.1
to 33.9 nM. Most of the compounds, such as 6a, 4a–c, 6e,
and 11 showed a very good activity with KIs in the low nano-
molar range, of 2.1–4.1 nM, comparable with AZM (KI of
5.7 nM). It was observed that the disubstituted compounds
with bulkier moieties, such as aniline-morpholine, 3-methyl
aniline- and morpholine, or dimorpholino, were less active
against hCA XII, a situation already encountered at the inhi-
bition of the other CA isozymes discussed here, and probably
due to their too bulky nature and difficulties to effectively
bind within the restricted space of the enzyme active site
cavity. The monosubstituted compounds showed better
hCA IX/hCA XII inhibition activity compared to the disubsti-
tuted derivatives incorporating similar scaffolds.
Inhibition data of human carbonic anhydrase isozyme I, II, IX and XII with synthesized
compounds by a stopped-flow, CO2 hydration assay method15
No
R
R0
KIa (nM)
hCA I hCA II hCA IX hCA XII
4a
4b
4c
4d
4e
Cl
Cl
Cl
Cl
Cl
NHPh
87
18.0
18.3
40.7
67.1
5.9
5.1
1.2
21.4
21.2
4.2
4.3
2.1
8.6
25.4
3-NHC6H4Me
HOC2H4N(Me)
H2N(CH2)3NH
4-F-3-
ClC6H3NH
NHPh
3-NHC6H4Me
NH2
O(CH2CH2)2N
t-BuNH
354
1070
1375
2540 130
5a
5b
6a
6b
6c
6d
6e
10
11
12
AAZ
EZA
O(CH2CH2)2N
O(CH2CH2)2N
NH2
O(CH2CH2)2N
t-BuNH
430
447
90
583
410
2365
4350
421
87
34.1
20.1
10.2
2.4
34.1
19.3
5.9
18.3
10.7
3.9
16.4
15.1
3.9
33.9
18.5
12.3
4.1
11.5
5.4
6.7
21.9
21.2
12.5
54.3
72.6
75.9
24.2
15.6
21.4
12.1
8.0
HOC2H4N(Me) HOC2H4N(Me)
H2N(CH2)3NH
H2N(CH2)3NH
—
—
—
—
—
110
250
25.2
8.5
25.0
34.3
5.7
22.1
a
Mean from three different assay, by a stopped flow technique (errors were in
the range of 5–10% of the reported values).
(v) All the synthesized compounds showed better inhibition of
the transmembrane tumor associated isoforms hCA IX and
hCA XII over cytosolic isoforms hCA I and hCA II with good
selectivity profile and specificity. For example, 4c showed a
selectivity ratio of 891 for inhibiting hCA IX over hCA I,
and of 33.9 for inhibiting hCA IX over hCA II. The selctivity
ratios for inhibiting hCA XII over hCA I and II were in the
same range like that of hCA IX (selectivity ratio of 509 for
inhibiting hCA XII over hCA I and of 19 over hCAII), making
this compound highly selective for the inhibition of the
tumor-associated over the cytosolic CA isoforms.
(3-amino propyl amino)-1,3,5-triazin-2-ylamino)benzene sul-
fonamide (compound 6e), which had a KI value of 4.35 lM
against this isoform and possesses the rather bulky, long
1,3-propylene diamine arms. The bulkiness present in this or
structurally related compounds (e.g., 6d, 4d and 4e) probably
interferes with the effective binding within the restricted space
of the hCA I active site, leading to a lower activity of these com-
pounds, with inhibition constants in the micromolar range
(Table 1).
(ii) The synthesized compounds showed effective hCA II inhibi-
tory activity with KIs ranging between 12.5 and 130 nM,
having thus a better inhibition profile against this isoform
compared to hCA I (Table 1). Some of the new compounds,
such as 6a, 4a, 4b, 5b, 6b, 10, 11 and 12 showed quite potent
hCA II inhibition, with KIs in the range of 12.5–24.2 nM, com-
parable to the standard drugs AAZ and EZA. They incorpo-
rate both the 1,3,5-triazine or the isostructural rings
(pyridine, pyrimidine or diazine) in their molecules, as well
as amino, methoxy, chloro, anilino and morpholino groups.
Derivatives 5a, 6c, 4c, 6d, 4d, 6e and 4e showed a weaker
hCA II inhibitory activity, with inhibition constants ranging
between 34.1 and 130 nM (Table 1). They seem to incorpo-
rate bulkier moieties compared to the previous ones, of the
tert-butylamino, hydroxyethylamino-methyl or 3-chloro-4-
fluoroanilino. Probably the most important structural
feature interfering with the effective binding is again the
bulkiness of the moieties substituting the 1,3,5-triazine ring,
as the derivatives with compact or less bulky such moieties
showed a better activity compared to those incorporating
one or two bulky functionalities.
(iii) The transmembrane tumor associated isoform hCA IX was
effectively inhibited by the synthesized compounds with
KIs in the range of 1.2–34.1 nM. Compounds 6a, 4a–c, 6d
and 11 showed inhibition constants KI ranging from 1.2 to
5.9 nM, which is better than standard drugs AZM and EZM.
The active compounds incorporated –Cl or –NH2 substitu-
ents at position 2 of the triazine nucleus whereas com-
pounds 5a–b and 6b, having morpholine substitution at
position 2 of triazine ring showed less potent hCA IX inhib-
itory activity. Compound 6b, with the worst activity was
equipotent with EZM and incorporated two morpholine
moieties with the 1,3,5-triazine nucleus. It was observed
In order to have a hint on the binding orientation of these inhib-
itors within the active site of the enzyme, docking of some of the
new compounds was performed using the GOLD suite program.9
The PDB ID 3IAI, of a triazinyl-substituted benzenesulfonamide
bound to hCA II reported earlier by our and McKenna’s group10
was downloaded from RCSB protein data bank and used for the
docking experiments. Protein and ligand files were prepared using
DockPrep tool in CHIMERA software, which included addition of
charges, addition or removal of hydrogens, repairing of residues
by adding missing chains, etc. Binding site was defined at
X = 70.5457, Y = 51.4075, Z = 11.8884 by selecting all atoms within
12 Å, considering the co-crystallized bound ligand. Docking was
carried out using Genetic algorithm and validated using the co-
crystallized bound ligand, acetazolamide. After validation, all the
compounds were docked into the binding pocket keeping the same
value of coordinates and positioned all the ligands in the same ori-
entation as acetazolamide (Fig. 1).
As for all other hCA II–sulfonamide adducts investigated so far,
it was observed that the nitrogen atom of the sulfonamide 6b (a
potent hCA II inhibitor, KI of 12.5 nM) coordinates to Zn(II) ion
and participates in a network of hydrogen bonds involving residues
Thr199 and Glu106. An oxygen function of sulfonamide group also
hydrogen bonds to the –NH moiety of Thr199. Additionally, the
polar amino acids Gln67, Asn62, Ser65 were found to be in close
contact with the inhibitor 6b (Fig. 1a), thus stabilizing the
enzyme–inhibitor adduct. In the docking analysis of compound
4e (the least potent against hCA II in the series reported here, KI
of 130 nM, Fig. 1b), it was observed that due to presence of bulky,
fluorophenyl functionality at the 1,3,5-triazine ring, the orientation
and alignment of the inhibitor was not perfectly fitted within the
hCA II active site. The scaffold of the inhibitor was found