Exploring the potential of 3′-O-carboxy esters of thymidine as inhibitors of ribonuclease A and angiogenin

Article abstract of DOI:10.1016/j.bmc.2008.01.003

In this study, compounds with a carboxy ester in lieu of the phosphate ester at the 3′-position have been employed to inhibit the ribonucleolytic activity of ribonuclease A (RNase A). Phosphates at the 3′-position of pyrimidine bases are well-known inhibitors of the protein. We have investigated the inhibition of RNase A by 3′-O-carboxy esters of thymidine. The compounds behave as competitive inhibitors with inhibition constants ranging from 42 to 95 μM. The mode of inhibition has also been confirmed by ¹H NMR studies of the active site histidines of RNase A. Docking studies have further substantiated the experimental results. The compounds are also found to inhibit the ribonucleolytic activity of angiogenin, a homologous protein and potent inducer of blood vessel formation.

Full text of DOI:10.1016/j.bmc.2008.01.003

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 2819–2828
Exploring the potential of 30-O-carboxy esters of thymidine
as inhibitors of ribonuclease A and angiogenin
Kalyan S. Ghosh, Joy Debnath, Palash Dutta, Bijaya K. Sahoo and Swagata Dasgupta*
Department of Chemistry, Indian Institute of Technology, Kharagpur 721302, India
Received 13 October 2007; revised 2 January 2008; accepted 4 January 2008
Available online 10 January 2008
Abstract—In this study, compounds with a carboxy ester in lieu of the phosphate ester at the 30-position have been employed to
inhibit the ribonucleolytic activity of ribonuclease A (RNase A). Phosphates at the 30-position of pyrimidine bases are well-known
inhibitors of the protein. We have investigated the inhibition of RNase A by 30-O-carboxy esters of thymidine. The compounds
behave as competitive inhibitors with inhibition constants ranging from 42 to 95 lM. The mode of inhibition has also been con-
1
firmed by H NMR studies of the active site histidines of RNase A. Docking studies have further substantiated the experimental
results. The compounds are also found to inhibit the ribonucleolytic activity of angiogenin, a homologous protein and potent indu-
cer of blood vessel formation.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
the ribonucleolytic site of angiogenin, which is identical
to RNase A, is comprised of His 13, Lys 40, and His
114.8 A comparative schematic diagram of the nucleo-
tide binding site of RNase A and angiogenin is given
in Figure 1. The active site homology may be exploited
by targeting RNase A as a suitable model to help the de-
sign of antiangiogenic compounds.
Inhibitors of mammalian ribonucleases (RNases) that
cleave ribonucleic acid (RNA) have often been utilized
to limit the undesirable biological function of many
ribonuclease type proteins. Several RNase homologs,
including angiogenin,1 eosinophil-derived neurotoxin
(EDN),2 and bovine seminal RNase A3, utilize their
ribonucleolytic activities to bring forth effectual physio-
logical effects, which are oftentimes related to certain
disease processes.4 Apparently, to exhibit the angiogenic
activity (growth of new blood vessels) of angiogenin its
ribonucleolytic activity is essential, even though it is of
lower magnitude (ꢀ105 lower) than bovine pancreatic
ribonuclease A (RNase A).5 Thus inhibitors of these
proteins may be considered as potential lead compounds
for the development of drug candidates.
Many nucleoside and nucleotide-based small molecule
inhibitors of ribonucleases have been identified.9,10
These nucleotide-based inhibitors have close resem-
blance to the natural substrate of RNase A and thus
act as substrate mimics. Most of them are derivatized
with the phosphate ester with an adenosine 50-pyrophos-
phate derivative, pdUppA-30-p being the most potent
inhibitor known.11 Replacement of the phosphate esters
by 30-amino groups proved to be a successful venture
which has already been published in this journal.12
The crystal structure of the complex of 30-N-piperi-
The active site of RNase A, a convenient model system
of this superfamily, is conventionally described in terms
of multiple subsites for binding phosphate, base, and ri-
bose moieties of the RNA substrate.6,7 The phosphate
binding subsite of RNase A where phosphodiester bond
cleavage occurs is designated as P1, which contains the
catalytic residues His 12, Lys 41, and His 119. Similarly
dine-4-carboxyl-30-deoxy-ara-uridine with RNase
A
was also subsequently reported in this journal. The
structure confirmed the interaction with active site resi-
dues despite the absence of the phosphate group.13 We
have also reported the inhibition of RNase A with the
major green tea polyphenols, that act as a noncompeti-
tive inhibitors.14,15 These compounds show distinctly
different behavior than the compounds reported here.
This study has also been followed by the inhibition of
RNase A by copper complexes of the two gallate con-
taining polyphenols present in green tea.16
Keywords: Ribonuclease A; Angiogenin; Thymidine-30-O-carboxy es-
ter; Competitive inhibition; FT-IR; Circular dichroism; Docking.
*
Corresponding author. Tel.: +91 3222 283306; fax: +91 3222
255303; e-mail: swagata@chem.iitkgp.ernet.in
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.01.003

2820
K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
Figure 1. Key residues of active site of (A) RNase A and (B) human angiogenin.
In this study, the phosphate group at the 30-position has
been substituted with a carboxylate ester group. It might
be assumed that the negative charge center on the car-
bonyl oxygen atom of the ester moiety would be able
to perturb the ionic environment of the active site by
interacting with protonated histidine or lysine residues.
It is known that the pKa values of histidine residues of
the P1 subsite change from 5.22/6.78 for free enzyme
to 6.30/8.10 for the enzyme–substrate complex.17 A per-
turbation in the active site area would, therefore, be
capable of disrupting the catalytic activity, which would
tantamount to inhibition of the enzyme. The com-
pounds synthesized have been shown to inhibit the ribo-
nucleolytic activity of RNase A both qualitatively and
quantitatively. Changes in the local pKa of the active site
histidines His 12 and His 119 were also determined. FT-
IR and CD studies were used to monitor secondary
structural changes occurring due to the interaction of
the compounds with RNase A. Furthermore, this study
is pertinent since these compounds would be expected to
serve as potent inhibitors of the ribonucleolytic activity
of angiogenin that in turn is expected to affect its angio-
genic activity.5
Figure 2. (A) Agarose gel-based assay for the inhibition of RNase A.
2. Results and discussion
Lane 1: tRNA; Lane 2: RNase A and tRNA; Lane 3: 0.6 mM
compound 5, RNase A and tRNA; Lane 4: 0.6 mM compound 6,
RNase A, and tRNA; Lane 5: 0.6 mM 30-TMP, RNase A and tRNA;
Lane 6: 0.6 mM 30-CMP, RNase A, and tRNA. (B) Comparative
ribonucleolytic inhibitory activities of RNase A and angiogenin by 5
(black) and 6 (gray) 30-TMP (dotted) and 30-CMP (white).
In this study we have checked the inhibition of RNase A
by 30-O-carboxy ester modified thymidines compared to
30-TMP and 30-CMP, reported inhibitors of RNase A.9
The inhibition of the ribonucleolytic activity of RNase
A was initially checked by an agarose gel-based assay,
where the degradation of tRNA by RNase A was mon-
itored with and without the compounds (Fig. 2A). The
most intense band observed in lane 1 is due to the pres-
ence of the control tRNA. The faint intensity of the
band in lane 2 is due to the degradation of tRNA by
RNase A. The differential intensity of the bands in lanes
3, 4, 5, and 6 qualitatively indicates the degree of RNase
A inhibition by the synthesized compounds 5 and 6, 30-
TMP, and 30-CMP, respectively. These results show that
though the 30-phosphate ester of thymidine is a better
inhibitor, compounds 5 and 6 containing 30-O-carboxyl-
ate esters also act as effective inhibitors of RNase A. We
have also chosen 30-CMP as a very well-known inhibitor
of RNase A used as the standard inhibitor in our previ-

K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
2821
ous studies.12,14 Intensity observations qualitatively
indicate that compound 5 is a more potent inhibitor
than 30-CMP whereas the extent of inhibition of 6 and
30-CMP is comparable.
inhibitor than both 5 and 6. The order of inhibition con-
stant values obtained for 5 and 6 correlates well with the
results obtained from the agarose gel and precipitation
assays.
The inhibition of the ribonucleolytic activity of RNase
A by 5 and 6 was further assessed quantitatively by pre-
cipitation assay. Considering identical concentrations
(0.22 mM) for the compounds, we find that 5 and 6 in-
hibit RNase A by 38% and 22%, respectively, whereas
with 30-TMP and 30-CMP the inhibition is 45% and
24%, respectively (Fig. 2B). The concentrations of com-
pounds 5 and 6 at which 50% inhibition of RNase A oc-
curs (IC50), are 197 and 341 lM, respectively (IC50 for
30-TMP and 30-CMP are 167 and 312 lM, respectively).
To further determine the nature of inhibition and the
inhibition constants, kinetic experiments were con-
ducted. The inhibition constant values obtained for 5
and 6 were 42 1 and 95 3 lM, respectively. Eadie–
Hofstee plots (Fig. 3) indicate that these two compounds
behave as competitive inhibitors (convergence on the y-
axis). The reported Ki values for 30-TMP and 30-CMP
are 13.2 and 103 lM.18 This indicates that compound
5 is a better inhibitor than 30-CMP whereas compound
6 is comparable to 30-CMP, however 30-TMP is a better
Competitive inhibition of an enzyme is suggestive of a
direct interaction of the inhibitors with the active site
residues. In the reported hydrolytic mechanism of
RNase A, in the transphosphorylation step, His 12 acts
as a general base to deprotonate the 20-oxygen atom of
the substrate, which then attacks the phosphorus atom
to form a pentavalent transition state.7,10,19 This species,
which is stabilized by Lys 41, collapses to yield the cyclic
20,30-phosphate intermediate. His 119 serves as a general
acid to protonate the oxygen atom of the leaving group.
This mechanism was confirmed by site-directed muta-
genesis studies.19–21 Considering the acid–base role of
the active site histidines, one can follow the pKa changes
Figure 4. pH dependence of the histidine C-2 proton signals of (A)
only RNase A, (B) RNase A with compound 5, and (C) RNase A with
compound 6. Chemical shifts are shown for histidine residues His 12
(•), His 105 (j), His 119 (m).
Figure 3. Eadie–Hofstee plot for inhibition of RNase A by (A)
compound 5: 6.7 lM (m), 3.3 lM (j), and 0 lM (•); (B) compound 6:
62 lM (m), 30 lM (j), and 0 lM (•).

2822
K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
in the presence of inhibitors to monitor their behavior.
From the H NMR titration curve of the histidine C-2
1
proton (Fig. 4A), the observed pKa values for His 12,
His 105, and His 119 are 5.88, 6.59, and 6.22, respec-
tively, for free RNase A. When compound 5 and 6 com-
plex with RNase A (Fig. 4B and C, respectively), the
pKa values shift to 6.01, 6.62 and 6.77 for compound 5
and 6.23, 6.53, and 6.39 for compound 6, respectively.
Perturbation of the pKa values of His 12 and His 119
indicate that compounds 5 and 6 go to the active site
of RNase A. We observe that the pKa values for His
105 remain practically unaltered (<1%) which is also
similar to the reported change of pKa for His 105 in case
of 20-CMP22 and 30-CMP.23 The pKa value for His 119
shifts by ꢀ9% in the presence of compound 5. The in-
crease in the pKa value indicates that the presence of
the negative charge center has a direct influence on the
environment of the imidazole ring of His 119. An in-
crease in pKa is also observed when the negatively
charged phosphate group interacts with the His residues
for complexes with both 20-CMP22 and 30-CMP.23 In the
case of phosphate esters, however, there is a substantial
change (ꢀ28%). This can be attributed to the formation
of the dianion that is capable of perturbing the environ-
ment to a higher degree than compound 5 or 6, where
only a negatively polarized charge center on the car-
bonyl oxygen is resident. Interestingly, the pKa of histi-
dines are unaffected in the cytidine–RNase A complex,22
where such interactions of the phosphate dianion with
the His residues are absent. Our results thus indicate
that the inhibition of RNase A by compound 5 is pri-
marily because of its interaction with His 119. In case
of compound 6, relatively minor changes are observed
for His 12 (ꢀ6%) and His 119 (ꢀ3%). The presence of
the phenyl ring in this case could interact with Phe 120
resulting in a smaller perturbation to the His residues
but nevertheless occupying the active site. The abstrac-
tion of the proton from 20-OH by His 12 (acting as base)
for cytidine phosphates is not also possible for com-
pound 5 or 6, which has no 20-OH that explains the
much lower DpKa observed for the 30-O-carboxy esters.
The preferred docking poses for both the compounds
are consistent with these changes in pKa (discussed
later).
Figure 5. FT-IR difference spectra of RNase A complexes with
compound 5 (–•–) and compound 6 (ÆÆÆ) at RNase A:ligand as 1:1
(left hand axis) with FT-IR spectra of free RNase A (—) (right hand
axis).
tents, respectively. Similarly for compound 6 at the same
protein ligand ratio, difference spectra reflect changes in
a-helix (1662 cmꢁ1), random structure (1643 cmꢁ1), and
b-sheet (1635 cmꢁ1) content. The increase in intensity
observed in the difference spectra between 1549 and
1548 cmꢁ1 in the amide II region for compounds 5
and 6, respectively, can be attributed to the interaction
of the compounds with the backbone of the protein.
The distinctive changes in the content of secondary
structural elements are more prominent in the deconvo-
luted spectra of RNase A and its complexes with com-
pounds 5 and 6 (Fig. 6).
CD spectroscopic studies of free RNase A and its com-
plexes with compounds 5 and 6 also indicate a perturba-
tion in the secondary structure (Fig. 7). The percentage
of secondary structural elements for RNase A and its
complexes with compounds 5 and 6 at 1:0.5 and 1:1 ra-
tios was determined using SELCON3 method in
DICHROWEB.27 We find that there is an increase in
the a-helix content (ꢀ9%) with a concomitant reduction
in random (ꢀ5%) and turn structures (ꢀ4%) for com-
plexation with compound 5 at the higher ratio. An in-
crease in a-helix content of RNase A has also been
observed with the nucleoside 3-azido-3-deoxythymidine
(AZT).28 In case of compound 6 at the higher ratio, min-
or increases in both the a-helix and b-sheet contents
(ꢀ6% and 4%, respectively) are indicative more of con-
formational adjustments on complex formation rather
than a major change in secondary structure. This was
compensated by a decrease in the content of random
structure by ꢀ10%. The stabilization of b-sheet confor-
mation of RNase A has also been observed in the inter-
action of RNase A with (ꢁ)-epigallocatechin.15
The effect of binding of these compounds on the overall
secondary structure of RNase A was also investigated
by FT-IR and CD spectroscopy. In FT-IR spectros-
copy, amide I bands at 1645–1650 cmꢁ1 (mainly C@O
stretching) and the amide II band at 1548–1560 cmꢁ1
(C–N stretching coupled with an N–H bending mode)
of proteins change on interaction with small molecules
and ligands.24 The fitted Gaussian curves in the amide
I region with spectral ranges from 1610 to 1632 cmꢁ1
,
1636 to 1644 cmꢁ1, 1650 to 1662 cmꢁ1, and 1665 to
1680 cmꢁ1 are attributed to the b-sheet, random coil,
a-helix and turn structures, respectively.25,26 Changes
in the peak positions of the amide I and II bands in
the FT-IR spectra of RNase A are observed with both
compounds. This is visible in the difference spectra
(Fig. 5). With compound 5 at a protein ligand ratio of
1:1 the peaks at 1663, 1643, and 1629 cmꢁ1 indicate
changes in a-helix, random structure, and b-sheet con-
Protein–ligand docking studies were performed to ob-
tain some insight into the amino acid residues of RNase
A involved in interactions with compounds 5 and 6.
Though the crystal structure of the complex can repre-
sent specific details of the interactions, general observa-
tions may be obtained from docking studies. Our earlier
studies13 with another synthetic inhibitor of RNase A
revealed that docking poses are similar to those ob-
˚
tained from structural studies (RMSD 0.47 A). Interest-

K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
2823
Figure 6. Deconvoluted spectra of RNase A and its complexes with compound 5 and 6: (A) RNase A, (B) RNase A-compound 5 (1:0.5), (C) RNase
A-compound 5 (1:1), (D) RNase A-compound 6 (1:0.5), and (E) RNase A-compound 6 (1:1). The percentage of a-helix, b-sheet and random coil is
indicated in the legends.
˚
ingly, that inhibitor was also competitive in nature and
was found to dock to the active site, which was further
confirmed by structural studies (PDB entry 2G8R). In
the docking experiments for 5 and 6 with RNase A we
also observe a competitive mode of binding, which cor-
relates well with the experimental studies. The docking
pose shown in Figure 8A (for compound 5) reveals that
the ester carbonyl oxygen is 2.72 A away from the Nd1
atom of His 119 and the distance between the O atom at
the 30 position of compound 5 and Ne2 of His 12 is
2.72 A. In Figure 8B, compound 6 is close to Gln 11,
which is very close to His 12. The presence of the com-
pound 6 in the proximity of His 12 may result in the per-
turbation of the pKa of this residue as opposed to His
119 for compound 5. To compare with the mode of inhi-
bition of 30-TMP and 30-CMP with our compounds, we
have also performed the docking of these two mono-
phosphates with RNase A (Fig. 8C and D, respectively).
In this case we find that the phosphate dianion is within
hydrogen bonding distance of both His 12 and His 119,
˚

2824
K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
Figure 7. CD spectra of (A) RNase A-compounds 5 and (B) RNase A-compound 6 complexes. Free RNase A (—), RNase A:ligand 1:0.5 (ÆÆÆÆÆ), and
RNase A:ligand 1:1 (–•–).
as expected. This is also reflected in the increase of pKa
of both the histidines when complexed with 30-CMP.23
In this context we can compare the docking of com-
pounds 5 and 6 with the previously reported 30-non-
phosphate inhibitor (30-N-piperidine-4-carboxyl-30-
deoxy-ara-uridine)13 where the 4-carboxypiperidine
moiety of the inhibitor molecule binds at the subsite
B1 by hydrogen bonding interactions using its carboxyl
group with Thr 45. The uracyl group binds through
hydrogen bond interactions with the side chains His
119 and Asp 121. For this compound the nucleobase
interacts with P1 subsite residues, whereas for com-
pound 5 the ester carbonyl oxygen and for compound
6 the 50-oxygen of the ribose ring appear to interact with
the active site.
parable with that of 30-phosphate of thymidine,
whereas that of 30-O-benzoyl ester was lower. The
docking poses for these compounds are given in Sup-
plementary Information.
Since both compounds 5 and 6 inhibited the ribonuc-
leolytic activity of RNase A, the next step was to
check whether these carboxylate esters were capable
of inhibiting the ribonucleolytic activity of angiogenin.
In the precipitation assay as described in Section 3.4,
for monitoring the inhibition of the ribonucleolytic
activity of angiogenin (Fig. 2B), we observed that
compounds 5 and 6 inhibit 30% and 18% of the enzy-
matic activity, respectively. The corresponding inhibi-
tion with 30-TMP and 30-CMP is 19% and 35%. The
concentrations of compound 5 and 6 at which 50%
inhibition of ribonucleolytic activity of angiogenin oc-
curs (IC50) are 250 and 417 lM, respectively (IC50 for
30-TMP and 30-CMP are 214 and 395 lM, respec-
tively). According to one of our earlier studies, this
implies that the angiogenic activity should also be af-
fected due to the disruption in the enzymatic activ-
ity.12 A clearer understanding of the mechanism or
mode of binding is underway to assess how these
compounds may be better modified to provide more
potent inhibition.
An extension of the docking study to obtain the struc-
ture–activity relationship between the carboxy esters
of different nucleobases with their inhibition potency
for RNase A was also carried out. In this study, re-
sults of which are given in the Supplementary mate-
rial, we have compared the docking modes of 30-O-
acetyl and 30-O-benzoyl esters of three deoxypyrimi-
dines with estimates of their Ki values (Table S1). In
brief, we found that the 30-O-acetyl ester of thymidine
was found to be a better inhibitor than the corre-
sponding cytidine analogue, whereas the uridine ana-
logue was almost comparable to thymidine.
However, for the 30-O-benzoyl esters the inhibition
follows the order U > T > C. Altering the position of
the carboxyester with the same nucleobase (in this
case thymidine), the 30-O-esters show better inhibition
than the corresponding 50-O-esters. Finally we have
looked at the inhibition potency of 20-deoxy-30-phos-
phate of three pyrimidines and compared this with
the respective 30-O-esters. This also indicated that
the thymidine phosphate was best amongst three.
The inhibition constant of 30-O-acetyl ester was com-
3. Materials and methods
3.1. Materials
RNase A, yeast tRNA, 20,30-cCMP, 30-CMP, 30-TMP,
human angiogenin, and human serum albumin were
from Sigma–Aldrich. All other reagents were from
SRL India. UV measurements were made using a Per-
kin Elmer UV–vis spectrophotometer (Model Lambda
25). Concentrations were determined using the follow-

K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
2825
Figure 8. Stereoview of the docked conformations of (A) compound 5, (B) compound 6, (C) 30-TMP, and (D) 30-CMP with ribonuclease A. The
protein conformations have been kept identical for comparison. His 12, Lys 41, and His 119 have been marked to indicate the location of the catalytic
site. Possible hydrogen bonds are shown as dashed lines.
ing spectral data: RNase A e278.5 = 9800 Mꢁ1 cmꢁ1
,
hol concentration less than 3% in the following studies.
The extinction coefficients of compounds 5 and 6 at
266 nm were determined and their concentrations were
measured using
respectively.
29
20,30-cCMP
e
268 = 8500 Mꢁ1 cmꢁ1
,
and
30-CMP
18
e
260 = 7600 Mꢁ1 cmꢁ130, respectively. The stock solu-
tions of compounds 5 and 6 were prepared in ethanol
and diluted prior to use maintaining a maximum alco-
e ,
266 = 5950 and 6355 Mꢁ1 cmꢁ1

2826
K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
Scheme 1. Synthetic scheme for the synthesis of 30-O-carboxy esters of thymidine. Reagents and conditions: (a) TrCl/py, rt, 48 h; 92%; (b) Ac2O/Py,
rt, 12 h; 95%; (c) BzCl/Py/DCM, rt, 2–3 h; 88%; (d) TFA/DCM (1:4); 75% for 5 and 70% for 6.
3.2. Synthesis of 30-O-carboxy esters of thymidine
phenol blue was added to the mixture. Twenty microli-
ters from each solution was extracted and loaded onto
a 1.1% agarose gel. The undegraded tRNA was visual-
ized by ethidium bromide staining under UV light.
The 30-O-carboxy esters were synthesized starting from
thymidine according to a general synthetic procedure
as described by Zhou et al.31 (Scheme 1). In brief
compound 1 (3.6 mmol) was tritylated by stirring with
trityl chloride (5.4 mmol) in pyridine for two days to ob-
tain compound 2. Compound 3 was obtained by treating
2 (4.3 mmol) with acetic anhydride (4.3 mmol) in dry
pyridine for 12 h at room temperature. Compound 4
was obtained by treating 2 (4.1 mmol) with benzoyl
chloride (4.1 mmol) for 2–3 h at room temperature.
Then detritylation of 3 and 4 was performed with 20%
trifluoroacetic acid in dichloromethane to achieve the
target molecules 5 and 6.
3.4. Precipitation assay
The effect of 5 and 6 on the ribonucleolytic activity of
the RNase A and angiogenin was examined with yeast
tRNA as the substrate as described by Bond.32 Five
microliters of 0.2 lM RNase A and 0.2 lM angiogenin
and their incubated mixtures with 10 ll of 5 (225 lM),
6 (225 lM), 30-TMP (225 lM), and 30-CMP (225 lM)
each were added separately to the assay mixture con-
taining 0.1 mg of yeast tRNA in 0.1 M Tris–HCl buffer
having pH 7.5, 5 mM EDTA, and 10 lg HSA in a final
volume of 50 ll. For RNase A after incubation of the
reaction mixture at 37 ꢁC for 30 min and for angiogenin
after incubation for 4 h, 100 ll of ice-cold 1.14 N per-
chloric acid containing 6 mM uranyl acetate was added
to quench the reaction. The solution was then kept in ice
for another 30 min and centrifuged at 4 ꢁC for 5 min.
Hundred microliters of the supernatant was taken and
diluted to 500 ll. The change in absorbance at 260 nm
was measured and compared to a control set.
3.2.1. Compound 5. Yield: 75%, 1H NMR (CDCl3)
(400 MHz): d 8.39 (br s, 1H); 7.50 (s, 1H); 6.25 (t,
J = 7.2 Hz, 1H); 5.35 (m, 1H); 4.09 (m, 1H); 3.93 (m,
2H); 2.41 (m, 2H); 2.10 (s, 3H); 1.92 (s, 3H); 13C
NMR (CDCl3) (100 MHz): d 170.7; 163.8; 150.5;
136.3; 111.3; 85.8; 85.0; 74.7; 62.5; 37.2; 20.9; 12.5;
HRMS (ES+Na)+: Calculated for C12H16O6N2Na:
307.1341, found: 307.1347.
1
3.2.2. Compound 6. Yield: 70%, H NMR (DMSO-d6)
(400 MHz): d 8.01–7.52 (m, 5H); 7.78 (s, 1H); 6.28 (t,
J = 7.2 Hz, 1H); 5.47 (m, 1H); 4.14 (m, 1H); 3.69 (m,
2H); 2.38 (m, 2H); 1.78 (s, 3H), 13C NMR (DMSO-d6)
(100 MHz): d 165.6, 164.1, 150.9; 136.3, 134.0, 129.7
(2C); 129.2 (3C); 110.2; 84.9, 84.2; 76.1; 61.8; 37.0;
12.7; HRMS (ES+Na)+: Calculated for C17H18O6N2Na:
369.1873, found: 369.1869.
3.5. Inhibition kinetics
The inhibition of RNase A by 5 and 6 was assessed indi-
vidually by a spectrophotometric method as described
by Anderson et al.18 The assay was performed in oligo
vinylsulfonic acid free33 0.1 M Mes–NaOH buffer, pH
6.0, containing 0.1 M NaCl using 20,30-cCMP as the
substrate. For 5, the substrate concentrations ranged
from 0.14 to 0.28 mM and the inhibitor concentrations
ranged from 0 to 6.7 lM. For 6, the substrate concentra-
tions varied from 0.13 to 0.25 mM and inhibitor concen-
trations were from 0 to 62 lM. The RNase A
concentration used was 10 lM. The inhibition constants
were determined from initial velocity data using an Ea-
die–Hofstee plot, where V versus V/[S] is plotted to ob-
tain the kinetic constants according to the following
equation:
3.3. Agarose gel-based assay
Inhibition of RNase A by 5 and 6 was checked qualita-
tively by the degradation of tRNA in an agarose gel. In
this method, 20 ll of RNase A (0.66 lM) was mixed
with 20 ll each of 1.5 mM 5, 6, 30-TMP or 30-CMP solu-
tions to a final volume of 50 ll and incubated for 6 h.
Twenty microliter aliquots of the incubated mixtures
were then mixed with 20 ll of tRNA solution (5.0 mg/
ml tRNA, freshly dissolved in RNase free water) and
incubated for another 30 min. Ten microliters of sample
buffer which contains 10% glycerol and 0.025% bromo-
ꢀ
ꢁ
ꢁKsV
½Sꢂ
½Iꢂ
V ¼
þ V max where Ks ¼ Km 1 þ
Ki

K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
2827
V is the initial velocity, [S] is the substrate concentra-
tion, [I] is the inhibitor concentration, Km is the Michae-
lis constant, Ki is the inhibition constant and Vmax is the
maximum velocity. From the slopes of the lines (Ks), the
inhibition constants were calculated for both the
compounds.
at 25 ꢁC. The spectra were recorded in the range of
190–240 nm with a scan rate of 50 nm/min and a re-
sponse time of 4 s. For baseline correction, CD spectra
of buffer (20 mM phosphate buffer of pH 7.0) containing
variable concentrations of ligands were collected and
were subtracted from each sample spectra. The RNase
A concentration was 12.5 lM and complex mixtures
with RNase A: ligand ratios of 1:0.5 and 1:1 were ana-
lyzed. Results were expressed as H and secondary struc-
tures were determined using SELCON3 method
(Reference set 7) in DICHROWEB,27 an online server
for protein secondary structure analyses from circular
dichroism spectroscopic data.
3.6. Proton NMR study
Exchangeable hydrogen atoms of RNase A were re-
placed by deuterium to facilitate the pH titration by
1H NMR. To exchange hydrogen by deuterium the pro-
tein was dissolved in D2O incubated at 50 ꢁC for 20 min,
and then lyophilized.34 This was repeated thrice. The
lyophilized protein was then dissolved in D2O contain-
ing 0.2 M NaCl. Afterward DCl and NaOD solutions
were used to adjust the pH* for the range 3–8, where
pH* is the direct measure of the pH which was not cor-
rected for a deuterium isotope effect at 30 ꢁC. The [li-
gand]/[enzyme] molar ratio was 5:1, which was
maintained for both compounds 5 and 6. 1H NMR data
were recorded on Bruker 400 MHz spectrometer at
22.5 ꢁC. The acquisition time was 2 s with 100 scans.
The observed d value was recorded with respect to so-
dium 2,2-dimethyl-2-silapentane 5-sulfonate (DSS).
The values of dobs for histidine C-2 protons at different
pH were fitted to a least square sigmoidal curve.
3.9. Docking studies
The crystal structure of RNase A-30–CMP complex
(PDB entry 1RPF) was downloaded from the Protein
Data Bank.35 Protein and ligand (30-CMP) structures
were extracted individually from it. The 3D structures
of the compounds were generated by Sybyl 6.92 (Tripos
Inc., St. Louis, USA) and their energy-minimized
conformations were obtained with the help of the
MMFF94 force field using MMFF94 charges. The
FlexX software as part of the Sybyl suite was used
for docking of all the compounds to RNase A. PyMol36
was used for visualization of the docked confor-
mations.
3.7. FT-IR studies
The coordinates of the docked ligand molecules ob-
tained from FlexX in Sybyl were used in AutoDock
3.0 to get an estimate of the inhibition constants
(Ki).37 Polar hydrogen atoms were added and Kollman
United Atomic (KOLLUA) charges assigned to the
RNase A (16 mg/ml) was dissolved in 20 mM phosphate
buffer of pD 7.2 in 99.9% D2O. RNase A: ligand ratios
of 1:0.5 and 1:1 were prepared for the both compounds 5
and 6. FT-IR measurements were carried out at 25 ꢁC
on a Nexus-870 FT-IR spectrometer (Thermo Nicolet
Corporation) equipped with a germanium-attenuated
total reflection (ATR) accessory, a DTGS KBr detector,
and a KBr beam splitter. All spectra were taken via the
attenuated total reflection (ATR) method with a resolu-
tion of 4 cmꢁ1, with 256 scans each. Control buffer spec-
tra (20 mM phosphate buffer of pD 7.2) were also
recorded under identical conditions. The background
was corrected before every sample.
3
˚
protein. A grid size of 20 · 20 · 20 A with a grid spac-
ing of 0.425 A was used keeping other default parame-
˚
ters. Ligand docking was carried out with the
AutoDock 3.0.5 Lamarckian Genetic Algorithm
(GA).37 For all the ligands the inhibition constant
(Ki) is estimated at 298.15 K. The root mean square
deviation (RMSD) value for the ligand molecule over
all the docked structures, obtained from Sybyl and
AutoDock docking programs, varied between 0.40
and 0.92.
IR spectra obtained were determined following the meth-
od of Byler and Susi.24 The relative amounts of secondary
structural elements of RNase A and the RNase A com-
plexes with compounds 5 and 6 were processed by analyz-
ing the area under the Gaussian curve. Each Gaussian
band is assigned to a secondary structure according to
the frequency of its maximum: a-helix (1650–
1662 cmꢁ1), b-sheet (1610–1632 cmꢁ1), random coil
(1636–1644 cmꢁ1), turn (1665–1680 cmꢁ1), and b-anti-
parallel (1680–1692 cmꢁ1).25,26 The areas of all the com-
ponent bands assigned to a given conformation are then
summed and divided by the total area. The number ob-
tained is taken as the proportion of the polypeptide chain
in that particular conformation.24
Acknowledgments
S.D.G. is grateful to the Department of Science and
Technology (DST), Government of India for financial
support (Grant No. SR/FTP/LS-A-06/2001). J.D.
thanks CSIR, New Delhi, for a senior research fellow-
ship. The authors are thankful to Prof. Tanmaya
Pathak, Department of Chemistry, Indian Institute of
Technology, Kharagpur, for helpful discussion and
comments.
Supplementary data
3.8. Circular dichroism (CD) measurements
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmc.
2008.01.003.
Circular dichroism measurements were made on a Jas-
co-810 CD spectrophotometer, using 2 mm path length

2828
K. S. Ghosh et al. / Bioorg. Med. Chem. 16 (2008) 2819–2828
19. Thompson, J. E.; Raines, R. T. J. Am. Chem. Soc. 1994,
116, 5467.
References and notes
20. Trautwein, K. P.; Holliger, P.; Stackhouse, J.; Benner, S.
A. FEBS Lett. 1991, 281, 275.
21. Messmore, J. M.; Fuchs, D. N.; Raines, R. T. J. Am.
Chem. Soc. 1995, 117, 8057.
22. Meadows, D. H.; Roberts, G. C. K.; Jardetzky, O. J. Mol.
Biol. 1969, 45, 491.
1. Fett, J. W.; Strydom, D. J.; Lobb, R. R.; Alderman, E.
M.; Bethune, J. L.; Riordan, J. F.; Vallee, B. L. Biochem-
istry 1985, 20, 5480.
2. Sorrentino, S.; Glitz, D. G.; Hamann, K. J.; Loegering, D.
A.; Checkel, J. L.; Gleich, G. J. J. Biol. Chem. 1992, 267,
14859.
23. Meadows, D. H.; Jardetzky, O. Proc. Natl. Acad. Sci.
U.S.A. 1968, 61, 406.
3. Matousek, J. Comp. Biochem. Physiol. C 2001, 129,
175.
24. Byler, D. M.; Susi, H. Biopolymers 1986, 25, 469.
25. Susi, H.; Byler, D. M. Methods Enzymol. 1986, 130, 290.
26. Dong, A. C.; Huang, P.; Caughey, W. S. Biochemistry
1990, 29, 3303.
27. Whitmore, L.; Wallace, B. A. Nucleic Acids Res. 2004, 32,
W668.
28. Gaudreau, S.; Novetta-dellen, A.; Neault, J. F.;
Diamantoglou, S.; Tajmir-riahi, H. A. Biopolymers
2003, 72, 435.
29. Sela, M.; Anfinsen, C. B. Biochim. Biophys. Acta 1957, 24,
229.
4. Loverix, S.; Steyaert, J. Curr. Med. Chem. 2003, 10, 779.
5. Shapiro, R.; Riordan, J. F.; Vallee, B. L. Biochemistry
1986, 25, 3527.
6. Iwahashi, K.; Nakamura, K.; Mitsui, Y.; Ohgi, K.; Irie,
M. J. Biochem. 1981, 90, 1685.
7. Nogue’s, M. V.; Vilanova, M.; Cuchillo, C. M. Biochim.
Biophys. Acta 1995, 1253, 16.
8. Russo, N.; Acharya, K. R.; Vallee, B. L.; Shapiro, R.
Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 804.
9. Russo, A.; Acharya, K. R.; Shapiro, R. Methods Enzymol.
2001, 341, 629.
30. Bolen, D. W.; Flogel, M.; Biltonent, R. Biochemistry 1971,
10, 4136.
10. Raines, R. T. Chem. Rev. 1998, 98, 1045.
11. Russo, N.; Shapiro, R. J. Biol. Chem. 1999, 274, 14902.
12. Maiti, T. K.; De, S.; Dasgupta, S.; Pathak, T. Bioorg.
Med. Chem. 2006, 14, 1236.
13. Leonidas, D. D.; Maiti, T. K.; Samanta, A.; Dasgupta, S.;
Pathak, T.; Zographos, S. E.; Oikonomakos, N. G.
Bioorg. Med. Chem. 2006, 14, 6055.
31. Zhou, W.; Upendran, S.; Roland, A.; Jin, Y.; Iyer, R. P.
Bioorg. Med. Chem. Lett. 2000, 10, 1249.
32. Bond, M. D. Anal. Biochem. 1988, 173, 166.
33. Smith, B. D.; Soellner, M. B.; Raines, R. T. J. Biol. Chem.
2003, 278, 20934.
34. Quirk, D. J.; Raines, R. T. Biophys. J. 1999, 76, 1571.
35. Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.;
Bhat, T. N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E.
Nucleic Acids Res. 2000, 28, 235.
14. Ghosh, K. S.; Maiti, T. K.; Dasgupta, S. Biochem.
Biophys. Res. Commun. 2004, 325, 807.
15. Ghosh, K. S.; Maiti, T. K.; Debnath, J.; Dasgupta, S.
Proteins 2007, 69, 566.
36. DeLano, W. L. The PyMOL Molecular Graphics System,
DeLano Scientific, San Carlos, CA, USA, 2006. http://
pymol.sourceforge.net/.
37. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.;
Hart, W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem.
1998, 19, 1639.
16. Ghosh, K. S.; Maiti, T. K.; Mandal, A.; Dasgupta, S.
FEBS Lett. 2006, 580, 4703.
17. Herries, D. G.; Mathias, A. P.; Rabin, B. R. Biochem. J.
1962, 85, 127.
18. Anderson, D. G.; Hammes, G. G.; Walz, F. G. Biochem-
istry 1968, 7, 1637.