The design and synthesis of YC-1 analogues as probes for soluble guanylate cyclase

Article abstract of DOI:10.1016/j.bmcl.2005.10.093

Soluble guanylate cyclase (sGC) is highly activated in the presence of both YC-1 (1-benzyl-3-(5′-hydroxymethyl-2′-furyl)-indazole) and CO. In this report, the design, synthesis, and activity (i.e., sGC activation) of photolabile analogues of 3-(5′-hydroxy

Full text of DOI:10.1016/j.bmcl.2005.10.093

Bioorganic & Medicinal Chemistry Letters 16 (2006) 618–621
The design and synthesis of YC-1 analogues as probes for
soluble guanylate cyclase
Kirk W. Hering,a Jennifer D. Artz,b,c William H. Pearsond and Michael A. Marlettab,c,
*
aDepartment of Chemistry, University of Michigan, Ann Arbor, MI 48109-1055, USA
bDepartment of Chemistry, University of California, Berkeley, CA 94720, USA
cDepartment of Molecular and Cell Biology, University of California, Berkeley, CA 94720, USA
dBerry & Associates, Inc., 2434 Bishop Circle East, Dexter, MI 48130, USA
Received 3 August 2005; revised 14 October 2005; accepted 14 October 2005
Available online 1 December 2005
Abstract—Soluble guanylate cyclase (sGC) is highly activated in the presence of both YC-1 (1-benzyl-3-(50-hydroxymethyl-20-furyl)-
indazole) and CO. In this report, the design, synthesis, and activity (i.e., sGC activation) of photolabile analogues of 3-(50-hydroxy-
methyl-20-furyl)-1-benzylindazole (YC-1) are presented. Initial results with 6-azido-3-(50-hydroxymethyl-20-furyl)-1-benzylindazole
led to the synthesis of a tritium-labeled analogue. When photoactivated, this analogue labeled the a-subunit of sGC.
Ó 2005 Elsevier Ltd. All rights reserved.
Soluble guanylate cyclase (sGC) catalyzes the conver-
sion of GTP to cGMP and pyrophosphate. sGC is a
heterodimeric hemoprotein that is activated several hun-
dred-fold over the basal activity upon the binding of
nitric oxide (NO). In the rat, sGC is comprised of an
80 kDa subunit and a 70 kDa subunit, and contains a
heme protoporphyrin IX cofactor that serves as the
binding site for NO. The cGMP formed in this reaction
serves as a second messenger regulating several cellular
events including smooth muscle relaxation,1 platelet
aggregation,2 and neuronal communication.1 sGC can
also be activated 3- to 5-fold upon binding of carbon
monoxide (CO) to the heme, or 8- to 12-fold in the pres-
ence of the benzylindazole derivative YC-1 [3-(50-hy-
YC-1 has been shown to be efficacious in vivo, hence
interest in this activator has remained high. We report
here the synthesis of YC-1 and the photolabile YC-1
analogues 1–4 shown in Figure 1 that were designed as
photoaffinity labels for sGC. Compound 2 was found
to synergistically activate sGC in the presence of CO
and was found to covalently bind to the a-subunit of
sGC upon irradiation at 254 nm.
Design and synthesis of azido YC-1 derivatives (com-
pounds 1–4). Azides exhibit some undesirable properties
for use in photoaffinity labeling studies in that they typ-
ically require irradiation at 254 nm, a wavelength that is
droxymethyl-20-furyl)-1-benzylindazole],
a
known
activator of sGC.3 However, YC-1 and CO together
act synergistically to activate the enzyme to a level sim-
ilar to that observed with NO.4,5 A variety of experimen-
tal approaches have been utilized in attempts to locate
the YC-1 binding site within sGC including a site-direct-
ed mutagenesis study,6 resonance Raman studies,7,8
a
photoaffinity labeling study,9 and a deletion mutagenesis
study,10 to name a few. However, several of these re-
ports have yielded conflicting results and thus the site
of action of YC-1 has remained an open question.
Keywords: Soluble guanylate cyclase; Nitric oxide; YC-1.
*
Corresponding author. Tel.: +1 510 643 9325; fax: +1 510 643
9388; e-mail: marletta@berkeley.edu
Figure 1. Photolabile YC-1 analogues synthesized.
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.10.093

K. W. Hering et al. / Bioorg. Med. Chem. Lett. 16 (2006) 618–621
619
potentially damaging to hemoproteins, and they fre-
quently exhibit a high degree of nonspecific binding.11
However, azides are one of the smallest known photola-
bile moieties and the analogue design used here allowed
for flexibility in placement around the YC-1 structure.
This strategy should maximize the probability of map-
ping residues that make up the YC-1 binding pocket.
od of time allowed for selective palladium-catalyzed
cross-coupling over arsinimine formation, and thus
good yields of analogues 2, 3, and 4 were obtained by
this route.
Effect of YC-1 and novel azido YC-1 derivatives (com-
pounds 1–4) on sGC activity. Analogues that compete
at the YC-1 binding site would be expected to activate
like YC-1 or act as competitive inhibitors of YC-1. As-
says of analogues 1–4 at 100 lM with sGC in the pres-
ence and absence of CO gave the results shown in
Table 1. Compounds 2 [6-azido-3-(50-hydroxymethyl-
20-furyl)-1-benzylindazole] and 4 slightly activate sGC
(3.4- and 4.7-fold, respectively) and act synergistically
in the presence of CO, activating 29-fold and 20-fold,
respectively. For comparison, the activation of sGC by
YC-1 and the YC-1 analogues 1–4 were synthesized as
described in Scheme 1. The known compounds 5-azido-
indazole (5) and 6-azidoindazole (6)12 were treated with
iodine in the presence of base by the method of Collot13
to furnish the corresponding 3-iodoazidoindazoles 7 and
8 in excellent yields. Treatment of 7 and 8 with benzyl
bromide, tetra-n-butylammonium iodide, and potassium
t-butoxide furnished the Stille-coupling precursors 9 and
10 in 81% and 87% yields, respectively. Azides 13 and 14
were similarly prepared by reaction of 3-iodoindazole13
with the known m- and p-azidobenzyl bromides14 11
and 12, respectively. The furyl-coupling component 16
was easily obtained from furfuryl alcohol (15) by treat-
ment with two equivalents of n-butyllithium, followed
by quenching with tri-n-butyltin chloride. Reaction of
stannane 16 with 1-benzyl-3-iodoindazole13 (17) by the
palladium-catalyzed cross-coupling method of Farina
and Roth15 furnished a 74% yield of YC-1. A similar
reaction between stannane 16 and iodoindazole 9 result-
ed in only a 12% yield of the desired product 1. This is
likely due to a competing reaction of Ph3As with the
azide in a manner analogous to the Staudinger reaction
as reported by Cadogan and Gosney.16 However, run-
ning the reaction at room temperature for a greater peri-
Table 1. Fold activation of sGC by YC-1 and compounds 1–4
(100 lM)a
Compound
No CO
+CO
None
YC-1
1
3.9 1.0
139
9.2 0.1
1.7 0.2
3.4 0.1
2.0 0.2
4.7 0.1
9
1
2
3
4
4.4 1.3
29 2.3
7.1 0.7
20.3 8.0
a Purified sGC was obtained from a baculovirus/SF9 expression system
as described;18 end point assays were performed with concentrations
of DMSO of 4% v/v as described;7 all errors were derived from the
mean + range/2 for triplicate 2 min assays. Basal activity was
33 3 nmol/min/mg (n = 3).
Scheme 1. Synthesis of YC-1 and azide-containing YC-1 analogues. Reagents and conditions: (a) I2, KOH, DMF; (b) BnBr, KOt-Bu, Bu4NI, THF;
(c) 3-iodoindazole, KOt-Bu, Bu4NI, THF; (d) (i) 2 equiv BuLi, THF, ꢀ78 °C then ꢀ20 °C, 2 h; (ii) Bu3SnCl, THF, ꢀ78 °C then rt, 16 h; (e)
Pd2(dba)3, AsPh3, DMF, 1 h, 100 °C; (f) Pd2(dba)3, AsPh3, DMF, 3 days, rt.

620
K. W. Hering et al. / Bioorg. Med. Chem. Lett. 16 (2006) 618–621
YC-1 was 9.2-fold in the absence of CO and 139-fold in
the presence of CO. Compound 2 was chosen for further
studies based on the observed activity and the novel
placement of the photolabile moiety on the YC-1
skeleton.
In order to determine the potency of 2, the effect of
50 lM 2 on YC-1 activation of sGC–CO was deter-
mined (Fig. 2). These data were fit to a standard four-
parameter logistic equation (sigmoid) with variable
slope and the calculated EC50 values of sGC–CO activa-
tion by YC-1 in the absence and presence of 2 were
determined to be 14.5 4.4 lM and 21.9 8.7 lM,
respectively. The results show that 2 is either a very
weak inhibitor of YC-1 action or consistent with it
acting at a distinct site (see below).
Scheme 2. Synthesis of tritium-labeled 2. Reagents and conditions: (a)
MnO2, CH2Cl2, rt, 20 h, 89%; (b) [3H]NaBH4, CeCl3, MeOH/CH2Cl2,
rt, 5 min, 89%.
and NaBH4 allowed for selective reduction of the
aldehyde in the presence of the azide. Thus, reaction
of aldehyde 18 with [3H]NaBH4 and CeCl3 gave an
excellent yield of [3H]2 with a specific activity of
533.3 Ci/mol.
Koesling and coworkers report that the addition of
YC-1 to a solution of sGC–CO in the presence of
GTP causes the Soret absorption peak to shift from
424 nm to 420 nm.17 Examination of the sGC–CO
heme Soret with 2 showed that it did not act like
YC-1, in that 2 did not shift the heme Soret from
424 nm for sGC–CO in the presence or absence of
GTP (data not shown).
sGC and [3H]2 were irradiated at 254 nm and 20 °C in
the presence and absence of either a 10-fold excess of
unlabeled 2 or a 10-fold excess of YC-1, under argon
with the lamp 1 cm above the solution for 10 min.
Samples were then separated by SDS–PAGE, fixed
in EN3HANCE Autoradiography Enhancer, dried,
and exposed to X-ray film for 7 days. The results of
this study, shown in Figure 3, indicate that [3H]2
bound primarily to the a-subunit of sGC, similar to
the result reported by Stasch et al. in a related photo-
affinity study of sGC with the YC-1-like compound
BAY 51-9491.9 In addition, these data indicated that
unlabeled 2 almost completely competes out the bind-
ing of [3H]2, suggesting a specific site of interaction.
However, these results also show that YC-1 has little
or no competition with [3H]2, giving further evidence
that 2 could bind in an alternate site on sGC than
Photoaffinity labeling studies using compound [3H]2.
The synergistic activation of sGC in the presence of
CO and 2 provided a compelling case for further
study, therefore, photoaffinity labeling studies were
pursued. Tritium labeled 2 was prepared as shown
in Scheme 2. The alcohol 2 was oxidized to the alde-
hyde 18 by treatment with MnO2. Preliminary studies
showed that reaction of 18 with NaBH4 resulted in
reduction of both the aldehyde and the azide moie-
ties. However, it was found that treatment with CeCl3
Figure 2. Activation of sGC by a range of YC-1 concentrations in the
presence and absence of 50 lM 2. Purified sGC was obtained from a
baculovirus/SF9 expression system as described;18 end point assays
were performed with concentrations of DMSO of 4% v/v as described;7
all errors were derived from the mean + range/2 for triplicate 2 min.
assays. The data were fit with a standard four-parameter logistic
equation (sigmoid), and the calculated EC50 values of sGC–CO
activation by YC-1 in the absence and presence of 2 were determined
to be 14.5 4.4 and 21.9 8.7 lM, respectively.
Figure 3. Autoradiograph of [3H]2 (50 lM, 26.7 mCi, lane 1; 10 lM,
5.3 mCi, lanes 2–4) labeled sGC (25 lg) in the presence and absence of
100 lM 2 (lane 3) and 100 lM YC-1 (lane 4) and Coomassie stained
gel of Novex Mark 12 M.W. ladder (lane 5) and purified sGC (lane 6)
after separation by SDS–PAGE.

K. W. Hering et al. / Bioorg. Med. Chem. Lett. 16 (2006) 618–621
621
YC-1. The latter result is particularly important since
References and notes
it shows that there are distinct sites on sGC for small
molecule activators that could be exploited in drug
design.
1. Warner, T. D.; Mitchell, J. A.; Sheng, H.; Murad, F. Adv.
Pharmacol. 1994, 26, 171.
2. Buechler, W. A.; Ivanova, K.; Wolfram, G.; Drummer, C.;
Heim, J. M.; Gerzer, R. Ann. N.Y. Acad. Sci. 1994, 714,
151.
3. Ko, F. N.; Wu, C. C.; Kuo, S. C.; Lee, F. Y.; Teng, C. M.
Blood 1994, 84, 4226.
4. Friebe, A.; Schultz, G.; Koesling, D. EMBO J. 1996, 15,
6863.
5. Stone, J. R.; Marletta, M. A. Chem. Biol. 1998, 5, 255.
6. Friebe, A.; Russwurm, M.; Mergia, E.; Koesling, D.
Biochemistry 1999, 38, 15253.
7. Denninger, J. W.; Shelvis, J. P. M.; Brandish, P. E.; Zhao,
Y.; Babcock, G. T.; Marletta, M. A. Biochemistry 2000,
39, 4191.
In conclusion, a number of photolabile compounds
structurally related to YC-1 were synthesized and their
activation of sGC was determined. Although, none of
the synthesized analogues were as effective as YC-1 in
terms of activity toward sGC–CO, compound 2 was
identified as the best activator of sGC–CO (29-fold)
among those tested. Competition binding experiments
with 2 and YC-1 indicated that 2 was a relatively
weak binder or was binding to a different site than
YC-1. In order to further examine the interaction
between 2 and sGC, the tritium-labeled analogue
[3H]2 was prepared and utilized in a photoaffinity
labeling experiment. It was found that this analogue
was covalently bound primarily to the a-subunit of
sGC. However, the photoaffinity labeling experiment
failed to show that this compound was competitive
with YC-1. These data support a conclusion that the
sGC-activating YC-1 derivative 2 may be binding in
a distinct pocket and further suggest the existence of
novel activators of sGC distinct from YC-1. Current-
ly, further studies are under way to specify the [3H]2
binding site in the a-subunit of sGC and to further
characterize its mechanism of action.
8. Makino, R.; Obayashi, E.; Homma, N.; Shiro, Y.; Hori,
H. J. Biol. Chem. 2003, 278, 11130.
9. Stasch, J. P.; Becker, E. M.; Alonso-Alija, C.; Apeler, H.;
Dembowsky, K.; Feurer, A.; Gerzer, R.; Minuth, T.;
Perzborn, E.; Pleiß, U.; Schroder, H.; Schroeder, W.;
Stahl, E.; Steinke, W.; Straub, A.; Schramm, M. Nature
2001, 410, 212.
10. Koglin, M.; Behrends, S. J. Biol. Chem. 2003, 278, 12590.
11. Bayley, H.; Knowles, J. R. Methods Enzymol. 1977, 46, 69.
12. Mullock, E. B.; Suschitzky, H. J. Chem. Soc. (C) 1968,
1937.
13. Collot, V.; Dallemagne, P.; Bovy, P. R.; Rault, S.
Tetrahedron 1999, 55, 6917.
14. Mornet, R. J. Chem. Soc., Perkin Trans. 1984, 1, 879.
15. Farina, V.; Roth, G. P. Advances in Metal-Organic
Chemistry 1996, 5, 1.
16. Cadogan, J. I. G.; Gosney, I. J. Chem. Soc. Chem.
Commun. 1973, 586.
Supplementary data
17. Kharitonov, V. G.; Sharma, V. S.; Magde, D.; Koesling,
D. Biochemistry 1999, 38, 10699.
18. Brandish, P. E.; Buechler, W. A.; Marletta, M. A.
Biochemistry 1998, 37, 16898.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2005.10.093.