Zinc(II)-mediated inhibition of a ribonuclease by an N-hydroxyurea nucleotide

Article abstract of DOI:10.1016/S0960-894X(02)00929-0

The inhibition of ribonuclease Bi by 3′-N-hydroxyurea-3′-deoxythymidine 5′-phosphate is enhanced by 30-fold in the presence of Zn²⁺. Thus, an N-hydroxyurea nucleotide can recruit Zn²⁺ to inhibit the enzymatic activity of a ribonuclease. This result engenders a general strategy for the inhibition of non-metalloenzymes by metal complexes.

Full text of DOI:10.1016/S0960-894X(02)00929-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 409–412
Zinc(II)-Mediated Inhibition of a Ribonuclease by an
N-Hydroxyurea Nucleotide
Joshua J. Higgin,a Gennady I. Yakovlev,b Vladimir A. Mitkevich,b
Alexander A. Makarovb,* and Ronald T. Rainesa,c,*
aDepartment of Chemistry, University of Wisconsin-Madison, Madison, WI 53706, USA
bEngelhardt Institute of Molecular Biology, Russian Academy of Sciences, Vavilov Street 32, 119991 Moscow, Russia
cDepartment of Biochemistry, University of Wisconsin-Madison, Madison, WI 53706, USA
Received 6 September 2002; revised 27 October 2002; accepted 2 November 2002
Abstract—The inhibition of ribonuclease Bi by 30-N-hydroxyurea-30-deoxythymidine 50-phosphate is enhanced by 30-fold in the
presence of Zn2+. Thus, an N-hydroxyurea nucleotide can recruit Zn2+ to inhibit the enzymatic activity of a ribonuclease. This
result engenders a general strategy for the inhibition of non-metalloenzymes by metal complexes.
# 2002 Elsevier Science Ltd. All rights reserved.
Like proteases, ribonucleases are prevalent enzymes
that are worthwhile targets for inhibitor development.1,2
In many laboratory procedures, RNA must be pro-
tected from degradation. Moreover, the neovascular-
ization promoted by angiogenin relies on the
ribonucleolytic activity of that enzyme.3 Indeed, var-
iants of angiogenin with greater ribonucleolytic activity
are more effective at promoting neovascularization,4
and inhibiting the ribonucleolytic activity of angiogenin
could be an effective anti-angiogenesis strategy.5
Zinc is the second most abundant transition metal in
biology and is essential for life.10 In cells, almost all zinc
is bound to proteins as zinc(II).11 The ability of proteins
to bind Zn2+ ions with high affinity portends a new
strategy for ribonuclease inhibition. The efficacy of this
strategy has been demonstrated with serine proteases.
Using X-ray diffraction analysis, Katz and co-workers
discovered that a previously known inhibitor of trypsin,
bis(5 - amidino - 2 - benzimidazolyl)methane (BABIM),
inhibits that enzyme by inadvertently recruiting a single
Zn2+ 12ꢀ14 The Zn2+ coordinates four heteroatoms—
.
The development of ribonuclease inhibitors lags far
behind that of protease inhibitors. The most potent
known small-molecule inhibitor of a ribonuclease is
pdUppAp, which has Ki=0.22 mM for the inhibition of
ribonuclease A in 0.2 M Hepes buffer, pH 7.0, with no
added salt.6 This inhibitor emerged from multiple itera-
tions of inhibitors that closely resemble substrates.7,8 It
is unlikely that this iterative strategy will yield sub-
stantially better ribonuclease inhibitors. UpOC6H4-p-
CH2F is a mechanism-based inactivator of ribonuclease
A.9 Unfortunately, inactivation by UpOC6H4-p-CH2F
is not complete. Hence, new strategies for inhibiting or
inactivating ribonucleases are desirable.
two from BABIM and two from enzymic side chains.
The value of Ki for BABIM alone is 19 mM, and that for
Zn2+ alone is 1 mM. Yet, the Ki for BABIM plus Zn2+
is 5 nM.12
The ability of metal complexes to inhibit non-metal-
loenzymes could extend beyond serine proteases.15ꢀ19
Herein, we report the first zinc(II)-mediated inhibitor of
a ribonuclease. Our ligand is 30-N-hydroxyurea-30-
deoxythymidine 50-phosphate [pdT-30-NHC(O)NHOH;
1]. The logic of this choice is as follows. The use of a
deoxynucleoside creates additional space within the
active site of the enzyme–ligand complex. This space
could be necessary for Zn2+ binding. The use of a thy-
midine facilitates synthesis from a commercially avail-
able starting material (vide infra). The 50-phosphoryl
group provides another interaction with a phosphoryl
group-binding subsite, as in the binding of a polymeric
*Corresponding authors. Tel.:+7-095-135-4095; fax: +7-095-135-
1405; e-mail: aamakarov@genome.eimb.relarn.ru (A.A. Makarov);
tel.:+1-608-262-8588; fax:+1-608-262-3453;
biochem.wisc.edu (R.T. Raines).
e-mail:
raines@-
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00929-0

410
J. J. Higgin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 409–412
RNA substrate. Finally, hydroxamic acids are excep-
tional bidentate chelators of Zn2+ 19
½EŠ ½SŠkcat
T
v ¼
ð1Þ
ð2Þ
.
½SŠ þ KoMbs
As a model ribonuclease, we chose ribonuclease Bi
(binase; EC 3.1.27.3). Binase is a secretory ribonuclease
from Bacillus intermedius that catalyzes the cleavage of
RNA without a need for metal ions or cofactors. The
structure of crystalline binase is known at a resolution
of 1.65 A.20 Analysis of this structure, along with that of
a complex with a nucleoside 30-phosphate,20 suggests
that Glu73 and His102 of binase act as a base and acid,
respectively, during catalysis of RNA cleavage. The
carboxylate and imidazole groups in the side chains of
these residues could also serve as the enzymic ligands
for Zn2+ (Scheme 1).
where
ꢀ
ꢁ
½IŠ ½IꢀZn2þ
Š
KoMbs ¼ KM 1 þ
þ
KiI
Ki
In eqs 1 and 2, [E]T, [S], [I], and [I Zn2+] are the total
.
concentrations of the enzyme, substrate, inhibitor (N-
2+
.
hydroxyurea 1), and I Zn
complex, respectively; KM
is the Michaelis constant for the hydrolysis of poly(I);
KIi is the inhibition constant for the inhibitor alone; and
2+
.
Ki is the inhibition constant for the I Zn
complex.
The relationship between these two inhibition constants
N-Hydroxyurea 1 was synthesized by the route shown
in Scheme 2, which begins with the commercial reagent
30-azido-30-deoxythymidine 50-monophosphate (AZT
monophosphate).21 The ability of N-hydroxyurea 1 to
inhibit the ribonucleolytic activity of binase was asses-
and Kd and KZn (which are the equilibrium dissociation
constants of Zn2+ from the I Zn and E I Zn com-
plexes, respectively) are depicted in Scheme 3. In the
data analysis, the values of [I] and [Zn2+] were assumed
to be equal to the total concentration of inhibitor and
Zn2+, respectively, because the concentration of enzyme
2+
2+
.
. .
sed in the absence and presence of Zn2+ 22
.
The results of measurements of the binase activity inhibi-
tion by N-hydroxyurea 1 at different concentrations of
Zn2+ ions are shown in Figure 1. The intercept of the lines
on the ordinate is indicative of competitive inhibition.
was much lower than that of inhibitor or Zn2+
.
The values of kcat and KM for the hydrolysis of poly(I)
were 162 sꢀ1 and 79 mM, which are similar to those
reported previously.23,24 The inhibition constant for
The data in Figure 1 were used to evaluate the inhibi-
tion by N-hydroxyurea 1 and Zn2+ by using eq 1:
N-hydroxyurea 1 alone (that is, in the absence of Zn2+
)
was KIi=1.3 mM. In contrast, no inhibition of enzy-
matic activity was observed by Zn2+ alone in assays
performed with [Zn2+]ꢁ5 mM (data not shown).
The application of eq 1 to the data in Figure 1 enabled
the calculation of KiKd values for different [I] and
[Zn2+]. As listed in Table 1, these values were approxi-
mately constant at KiKd=3ꢂ10ꢀ7 M2 if [I][Zn2+]ꢁ10ꢀ7
M2. When [I][Zn2+] was increased to 1.7ꢂ10ꢀ6 M2, the
KiKd value increased by 10-fold (Table 1). Most likely,
this increase is indicative of improper usage of the total
concentration of inhibitor and Zn2+ rather than the
actual concentration. For this reason, no assays were
performed with [Zn2+]ꢃ5 mM.
Scheme 1. Basis for the zinc(II)-mediated inhibition of a ribonuclease
by N-hydroxyurea 1.
2+
.
The affinity of the I Zn complex for the enzyme was
discerned from the value of KiKd. The value of Kd
reports on the affinity of I for Zn2+ in the assay mix-
ture. The pKa of a model hydroxamic acid, acetohy-
droxamic acid (CH3C(O)NHOH), is 9.4,25 indicating
Scheme 3. Scheme for zinc(II)-mediated inhibition of enzymatic
activity. Values are for inhibition of ribonuclease Bi by N-hydroxy-
urea 1.
Scheme 2. Route for the synthesis of N-hydroxyurea 1.21

J. J. Higgin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 409–412
411
Table 1. Parameters for inhibition of ribonuclease Bi catalysis by
N-hydroxyurea 1 and Zn2+a
KoMbs
(10ꢀ4 M)
]
KiKd
(10ꢀ7 M2)
2+
.
Ki [I Zn
[I]
[Zn]
(10ꢀ3 M)
(10ꢀ4 M)
(M)2
0.93
1.0
3.5
3.5
0.25
1.0
0.25
5.0
0.91
1.12
1.21
1.43
12.4
2.9
3.8
1.85
2.9
2.9
3.3
32
aData are for those assays depicted in Figure 1 performed in the pre-
sence of Zn2+
.
Acknowledgements
We thank J. A. Hodges for contributive discussions.
This work was supported by Grants GM44783 (NIH to
R.T.R.), TW01058 (NIH to A.A.M), 02-04-48259
(RFBR to A.A.M), and 02-04-49110 (RFBR to G.I.Y).
Figure 1. Lineweaver–Burk plot for the inhibition of binase by N-
hydroxyurea 1 in the absence and presence of Zn2+. Assays were per-
formed at 25 ꢄC in 0.10 M sodium citrate buffer, pH 6.2, containing
NaCl (0.10 M), binase (4.4ꢂ10ꢀ10 M), poly(I), N-hydroxyurea 1, and
Zn2+. &, [I]=0, [Zn2+]=0 (data with [Zn2+]ꢁ5 mM were identical);
References and Notes
1. D’Alessio, G., Riordan, J. F., Eds. Ribonucleases: Struc-
tures and Functions. Academic Press: New York, 1997.
2. Raines, R. T. Chem. Rev. 1998, 98, 1045.
3. Shapiro, R.; Vallee, B. L. Biochemistry 1989, 28, 7401.
4. Harper, J. W.; Vallee, B. L. Proc. Natl. Acad. Sci. U.S.A.
1988, 88, 7139.
?,
[I]=0.93ꢂ10ꢀ4 M, [Zn2+]=0.25 mM; ~, [I]=3.5ꢂ10ꢀ4 M,
[Zn2+]=0; *, [I]=1.0ꢂ10ꢀ4 M, [Zn2+]=1.0 mM; !, [I]=3.5ꢂ10ꢀ4
M, [Zn2+]=0.25 mM; ^, [I]=3.5ꢂ10ꢀ4 M, [Zn2+]=5 mM.22
that 0.063% of acetohydroxamic acid is deprotonated at
pH 6.2. Only the conjugate base of a hydroxamic acid
5. Folkman, J. Nature Med. 1995, 1, 27.
6. Russo, N.; Shapiro, R. J. Biol. Chem. 1999, 274, 14902.
7. Leonidas, D. D.; Shapiro, R.; Irons, L. I.; Russo, N.;
Acharya, K. R. Biochemistry 1997, 36, 5578.
8. Russo, N.; Shapiro, R.; Vallee, B. L. Biochem. Biophys.
Res. Commun. 1997, 231, 671.
9. Stowell, J. K.; Widlanski, T. S.; Kutateladze, T. G.; Raines,
R . T.J. Org. Chem. 1995, 60, 6930.
10. Vallee, B. L. Biofactors 1988, 1, 31.
11. Outten, C. E.; O’Halloran, T. V. Science 2001, 292, 2488.
12. Katz, B. A.; Clark, J. M.; Finer-Moore, J. S.; Jenkins,
T. E.; Johnson, C. R.; Ross, M. J.; Luong, C.; Moore, W. R.;
Stroud, R. M. Nature 1998, 391, 608.
has high affinity for Zn2+ 19
,
and the acetohydrox-
2+
.
amate Zn
complex has an equilibrium dissociation
constant near 10ꢀ5.4 M.26,27 Thus, the value of
Kd=10ꢀ5.4 M/(0.063%)=6.3 mM for acetohydroxamic
acid at pH 6.2. Using this value of Kd as an approx-
imation for that of the N-hydroxyurea 1 Zn2+ complex,
.
the value of Ki=KiKd/Kd=3ꢂ10ꢀ7 M2/6.3 mM=47
mM. Thus, the enzyme has approximately 30-fold more
2+
.
affinity for the I Zn complex (Ki=47 mM) than for I
alone (KIi=1.3 mM).
13. Katz, B. A.; Luong, C. J. Mol. Biol. 1999, 292, 669.
14. Janc, J. W.; Clark, J. M.; Warne, R. L.; Elrod, K. C.;
Katz, B. A.; Moore, W. R. Biochemistry 2000, 39, 4792.
15. Schirmeister, T. Angew. Chem. Int. Ed. 1998, 37, 1830.
16. Thorp, H. H. Chem. Biol. 1998, 5, R125.
17. Louie, A. Y.; Meade, T. J. Chem. Rev. 1999, 99, 2711.
18. Nguyen, R.; Huc, I. Angew. Chem. Int. Ed. 2001, 40, 1774.
19. Farkas, E.; Buglyo, P. J. Chem. Soc., Dalton Trans. 1990,
1549.
The affinity of Zn2+ for the enzyme was discerned like-
wise. From Scheme 3, KZn=KiKd/KIi=3ꢂ10ꢀ7 M2/1.3
mM=0.2 mM. Because no inhibition of enzymatic
activity was observed with [Zn2+]ꢁ5 mM, the enzy-
2+
.
me Zn complex had an equilibrium dissociation con-
stant of >5 mM. Thus, Zn2+ has >25-fold more
.
affinity for the E I complex than for the enzyme alone.
This increase is consistent with the participation of
enzymic ligands in the binding of Zn2+ to the E I com-
.
plex, as is depicted in Scheme 1.
20. Polyakov, K. M.; Lebedev, A. A.; Okorokov, A. L.;
Panov, K. I.; Schulga, A. A.; Pavlovsky, A. G.; Dodson, G. G.
Acta Crystallogr. 2002, D58, 744.
21. 30-Amino-30-deoxythymidine 50-monophosphate. 30-Azido-
30-deoxythymidine 50-monophosphate (50 mg, 135 mmol), tri-
phenylphosphine (50 mg, 192 mmol), and tetra-
butylammonium fluoride (0.10 g, 0.36 mmol) was stirred in
pyridine (20 mL) overnight at 20 ꢄC. Aqueous NH3 (5% v/v;
30 mL) was added, and the resulting solution was stirred for 2
h. The mixture was co-evaporated with ethanol and dried
under vacuum. The mixture was used without purification
directly in the next step.
In conclusion, we have demonstrated the efficacy of a new
strategy for the inhibition of ribonucleases. This strategy
was inspired by the inadvertent recruitment of zinc by a
known protease inhibitor.12ꢀ14 In contrast, ribonuclease
inhibition herein relies on the intentional recruitment of
Zn2+ by an N-hydroxyurea moiety attached covalently
to a nucleotide. The N-hydroxyurea moiety can present
Zn2+ to the active-site residues of the ribonuclease, and
thereby enhance binding beyond that for the inhibitor or
Zn2+ alone. We anticipate that this strategy can be opti-
mized further and used for the inhibition of a variety of
ribonucleases, as well as other types of enzymes.
4-Nitrophenyl N-hydroxycarbamate. 4-Nitrophenyl N-
hydroxycarbamate was synthesized by the route reported for
the synthesis of phenyl N-hydroxycarbamate (Stewart, A. O.;
Brooks, D. W. J. Org. Chem. 1992, 57, 5020).
30-N-Hydroxyurea-30-deoxythymidine 50-monophosphate. 4-

412
J. J. Higgin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 409–412
Nitrophenyl N-hydroxycarbamate (0.10 g, 0.50 mmol) and
tetrabutylammonium fluoride (0.20 g, 0.72 mmol) was added
to the crude 30-amino-30-deoxythymidene 50-monophosphate
and stirred overnight in pyridine (20 mL) at 20 ꢄC. The reac-
tion was quenched with H2O (20 mL). The mixture was co-
evaporated with ethanol and dried under vacuum. The residue
was taken up in H2O (3 mL) and purified by reversed-phase
HPLC using an H2O/acetonitrile gradient and lyophilized to
give 54 mg (99% overall) of fluffy white solid. 1H NMR
(300 MHz, D2O) d 1.82 (s, 3H), 2.28 (m, 2H), 4.03 (m, 2H), 4.08
(m, 1H), 4.36 (m, 1H), 6.19 (t, 1.3H, NH), 7.78 (s, 1H). MS
(ESI) m/z calcd for C11H16N4O9P (M-H) 379.07, found 379.00.
22. Enzyme kinetics. Assays of ribonucleolytic activity were
performed by using UV spectroscopy to measure the cleavage
of poly(inosinic acid) [poly(I)] at 25 ꢄC in 0.10 M sodium
citrate buffer, pH 6.2, containing NaCl (0.10 M), as described
previously.23,24 Concentrations of N-hydroxyurea 1 were
determined by its absorbance at 267 nm using the extinction
coefficient for pdT, which is e=9.6 mMꢀ1 cmꢀ1. Dawson,
R. M. C.; Elliott, W. H.; Elliott, D. C. Data for Biochemical
Research, 3rd Edition; Clarendon Press: New York, 1989.
23. Yakovlev, G. I.; Moiseyev, G. P.; Struminskaya, N. K.;
Borzykh, O. A.; Kipenskaya, L. V.; Znamenskaya, L. V.;
Leschinskaya, I. B.; Chernokalskaya, E. B.; Hartley, R. W.
FEBS Lett. 1994, 354, 305.
24. Yakovlev, G. I.; Struminskaya, N. K.; Znamenskaya,
L. V.; Kipenskaya, L. V.; Leschinskaya, I. B.; Hartley, R. W.
FEBS Lett. 1998, 428, 57.
25. Wise, W. W.; Brandt, W. W. J. Am. Chem. Soc. 1954, 77,
1058.
26. Chang, C. A.; Sekhar, V. C.; Garg, B. S.; Guziec, F. S.,
Jr.; Russo, T. C., Jr. Inorg. Chim. Acta 1987, 135, 11.
27. Farkas, E.; Enyedy, E. A.; Csoka, H. J. Inorg. Biochem.
2000, 79, 205.