Development of a selective photoactivatable antagonist for
corticotropin-releasing factor receptor, type 2 (CRF
2
)
Ines Bonk
1
and Andreas Ru¨ hmann
2
1
WITA Proteomics AG, Teltow/Berlin, Germany;
2
Institute for Molecular Biosciences, The University of Queensland, St Lucia,
Australia
A novel photoactivatable analog of antisauvagine-30 (aSvg-
30), a specific antagonist for corticotropin-releasing factor
(CRF) receptor, type 2 (CRF
2
), has been synthesized and
characterized. The N-terminal amino-acid
D
-Phe in aSvg-30
[
D
-Phe11,His12]Svg
(11)40)
was replaced by a phenyldiazirine,
the 4-(1-azi-2,2,2-trifluoroethyl)benzoyl (ATB) residue. The
photoactivatable aSvg-30 analog ATB-[His12]Svg was
tested for its ability to displace [
125
I-Tyr0]oCRF or

¼ 3.1 ± 0.2 n
M
) but not the
rCRF
1
receptor (K
i
¼ 142.5 ± 22.3 n
M
) and decreased
Svg-stimulated cAMP activity in mCRF
2b
-expressing cells in
a similar fashion as aSvg-30. A 66-kDa protein was identified
by SDS/PAGE, when the radioactively iodinated analog of
ATB-[His12]Svg
(12)40)
was covalently linked to mCRF
2b
receptor. The specificity of the photoactivatable
125
I-labeled
CRF
2b
antagonist was demonstrated with SDS/PAGE by
the finding that this analog could be displaced from the
receptor by antisauvagine-30, but not other unrelated pep-
tides such as vasoactive intestinal peptide (VIP).
Keywords: photoaffinity labeling; antisauvagine-30; corti-
cotropin-releasing factor (CRF) receptor; CRF antagonist;

and CRF
2
has
been found in the brain and pituitary of rodents when
compared with humans, many of the anxiety-related
behavioral effects have been suggested to be governed by
CRF
1
receptor [5]. In this view, several CRF
1
-specific
nonpeptide antagonists are currently being investigated in
clinical phase II studies as potential drugs for anxiety-
related diseases [6].
The 40-amino-acid peptide urocortin (Ucn) a naturally
occurring CRF analog was proposed to be the endogenous
ligand for the CRF
2
receptor [7]. However, Ucn also
exhibited high affinity binding to CRF
1
[7], and fibers that
express Ucn did not correlate with targets in the brain
bearing CRF
2
receptors [8]. With the recent discovery of
Ucn II [9] and Ucn III [10] also known as stresscopin-related
peptide or stresscopin [11], respectively, novel peptide
agonists specifically binding to CRF
2

hmann, Institute for Molecular Biosciences,
The University of Queensland, St. Lucia QLD 4072, Australia.
Fax: + 61 7 3365 1990, Tel.: + 61 7 3365 1271,
E-mail: [email protected]
Abbreviations: Svg, sauvagine; aSvg-30, antisauvagine-30; CRF,
corticotropin-releasing factor; h/r/oCRF, human/rat/ovine CRF;
astressin, {cyclo(30-33)[
D
-Phe12,Nle21,38, Glu30,Lys33]h/rCRF(12-
41)}; ATB, 4-(1-azi-2,2,2-trifluoroethyl)benzoyl; HEK, human
embryonic kidney.
(Received 2 July 2002, revised 3 September 2002,
accepted 10 September 2002)
Eur. J. Biochem. 269, 5288–5294 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03246.x
[21] and astressin [22], a conformationally constrained
nonselective CRF peptide antagonist [12,23], we were now
interested in the development of a potent and selective
photoactivatable CRF antagonist, based on the amino-acid
sequence of aSvg-30 to further investigate the different
structural requirements for agonist and antagonist binding
to CRF
1
and CRF
2
.
Several CRF receptor cross-links with molecular
masses in the range of 58 000–75 000 have been char-
acterized applying bifunctional reagents to membranes of
bovine anterior pituitary membranes [24], AtT-20 mouse
pituitary tumor cells [25] rat brain, and anterior pituitary

aSvg-30 were synthesized and tested for selective binding to
CRF
1
or CRF
2
receptor.
MATERIALS AND METHODS
Synthesis of 4-(1-Azi-2,2,2-trifluoroethyl)benzoic acid
4-(1-Azi-2,2,2-trifluoroethyl)benzoic acid was synthesized in
an eight step synthesis as described [21,31].
Synthesis and purification of peptides
The CRF peptides (Fig. 1) were synthesized, purified, and
characterized as described [12,21,22].
For the synthesis of the cyclized CRF analogs, amino-acid
derivatives Fmoc-Glu(OAl)-OH and Fmoc-Lys(Aloc)-OH
(PerSeptive Biosystems GmbH, Hamburg, Germany) were
used. The side-chain protected peptides were reacted with
Pd°[PPh
3
]
4
in HOAc/N-methylaniline/dichloromethane
(2 : 1 : 40, v/v/v) for three hours and then cyclized with
1-hydroxybenzotriazole/O-(benzotriazol-1-yl)-N,N,N¢,N¢/
tetramethyluronium hexafluorophosphate in dimethylform-
amide and N,N-diisopropylethylamine in N-methylpyrroli-
dine for 8 h. After removal of the N-terminal Fmoc group
with piperidine in N-methylpyrrolidine, 4-(1-azi-2,2,2-tri-
fluoroethyl)benzoic acid was linked to the N-terminus of the
peptide resin with 1-hydroxybenzotriazole/O-(benzotriazol-

[His12]Svg
(12)40)
: ESI MS calculated 3716.3, observed
3715.4, R
t
¼ 26.6 min).
Fig. 1. Comparison of the amino-acid sequence
of (A) oCRF, (B) Svg, (C) astressin, (D) ATB-
[
125
I-labeled His13,Ala32]astressin, and (E)
ATB-[
125
I-labeled His12] Svg
(12-40)
.
Ó FEBS 2002 Photoactivatable CRF
2
receptor antagonist (Eur. J. Biochem. 269) 5289
Iodination to the photoactivatable aSvg-30 analog
ATB-[His12]Svg
(12)40)
was iodinated as described [33,34]
and subsequently purified with RP-HPLC and solvents A
and B as described above. The sample was eluted with 45%
B for 5 min and then with a linear gradient of 45–95% B in
25 min (
125
I-ATB-[His12]Svg
(12)40)

preparations as described [12,35].
Binding assays with CRF peptides
Binding of the CRF analogs to the rCRF
1
and mCRF
2b
receptor was performed essentially as described previously
[12].Briefly,5lL of membrane suspension (25 lgof
protein from HEK-rCRF
1
cells; 50 lg of protein from
HEK-mCRF
2b
cells) was added to a plate containing CRF
peptides (0–1 l
M
) and 50 000 c.p.m. of either [
125
I-
Tyr0]oCRF (specific activity 81.4 TBqÆmmol
)1
, 68.25 p
M
,
DuPont NEN, Boston) for the analysis of rCRF
1
or [
125
I-
Tyr0]Svg (specific activity 81.4 TBqÆmmol

[
125
I-Tyr0]Svg to membranes of transfected cells was
calculated by subtraction of unspecific binding found in
the presence of 1 l
M
of oCRF or Svg from total binding,
respectively. Data analysis was achieved with the nonlinear
curve fitting program
LIGAND
. Statistical analysis was
performed with
ANOVA
, and significant differences between
groups were determined by post hoc comparison using the
Dunn test. Values of P < 0.01 were considered statistically
different.
cAMP stimulation
HEK-rCRF
1
cells or HEK-mCRF
2b
cells were incubated
with different CRF agonists in the presence or absence of
1 l
M
or 10 n
M
antagonist, or CRF antagonist (1 l
M

Synthesis of ATB-[His12]Svg
(12)40)
and its radioactively
labeled analog
4-(1-Azi-2,2,2-trifluoroethyl)benzoic acid was successfully
linked to [His12]Svg
(12)40)
(Fig. 1). Subsequent radiolabe-
ling with
125
I gave ATB-[
125
I-His12]Svg
(12)40)
with a specific
activity of 74 TBqÆmmol
)1
.
Binding and cAMP assay
For the determination of the binding affinity and the
biological potency of the photoactivatable CRF antagonists,
HEK 293 cell lines, stably transfected with cDNA coding for
rCRF
1
or mCRF
2b
were used [21,36]. Scatchard analysis
indicated high-affinity binding of oCRF (K
i
¼ 0.6 ±

¼ 5.7 ± 1.6 n
M
;CRF
2
:
K
i
¼ 4.0 ± 2.3 n
M
). In contrast the photoactivatable aSvg-
30 analog (compound 1) exhibited high-affinity binding to
CRF
2
(K
i
¼ 3.1 ± 0.2 n
M
) but low affinity to CRF
1
(K
i
¼ 142.5 ± 22.3 n
M
) similar when compared with
aSvg-30 (CRF
1
: K
i
¼ 153.6 ± 33.5 n
M

M
, respectively. The
potency of the two peptide agonists to enhance cAMP in
5290 I. Bonk and A. Ru
¨
hmann (Eur. J. Biochem. 269) Ó FEBS 2002
HEK-mCRF
2b
cells was significantly different (oCRF:
EC
50
¼ 11.79 ± 1.96 n
M
; Svg: EC
50
¼ 0.23 ± 0.05 n
M
)
(not shown). Ovine CRF-stimulated cAMP production in
HEK-rCRF
1
cells could be inhibited by CRF antagonists in
the following rank order: astressin (compound 4), ATB-
[Ala
32
]astressin (compound 3) >> ATB-[His12]Svg
(12)40)
(compound 1) > [
D
-Phe11,His12]Svg

125
I-labeled compound 1 and membranes of
HEK-mCRF
2b
cells (Fig. 3A). No photo cross-link was
observed in analogous experiments with membranes of
HEK-rCRF
1
cells (Fig. 3B). Binding of
125
I-labeled com-
pound 1 to the CRF
2
receptor could be efficiently inhibited
by addition of increasing concentrations of antisauvagine-
30 but not 10 l
M
vasoactive intestinal peptide (VIP) in
agreement with the assumed specificity of this photoprobe
(Fig. 3A). Furthermore no cross-link could be identified
without light activation at 360 nm (not shown).
DISCUSSION
After the successful characterization of a photoactivatable
radioactively labeled CRF agonist and antagonists at CRF
1
receptors, we were now interested in the development of a
photoactivatable CRF
2
-selective antagonist based on the
amino-acid sequence of antisauvagine-30 in order to further

(n
M
) K
i(rCRF1)
/ K
i(mCRF
2b
)
1 ATB-[His12] Svg
(12)40)
3.1 ± 0.2 142.5 ± 22.3 45.97
2[
D
-Phe11, His12]Svg
(12)40)
f
1.4 ± 0.4 153.6 ± 33.5 109.71
3 ATB-[Ala32]astressin
b,c,g
2.6 ± 1.1 5.3 ± 1.3 2.04
4 Astressin
a
4.0 ± 2.3 5.7 ± 1.6 1.42
5 Svg 4.5 ± 0.4 0.7 ± 0.1 0.15
6[
D
-Phe11]Svg
(11–40)
3.5 ± 0.2 237.3 ± 27.7
c

, P < 0.01 vs. 4;
d
, P < 0.01 vs. 5;
e
, P < 0.001 vs. 3 and 4.
f
Antisauvagine-30 [12];
g
photo astressin [22].
Fig. 2. Displacement of [
125
I-Tyr0]oCRF (A) or [
125
I-Tyr0]Svg (B)
bound to membrane homogenates of HEK 293 cells stably transfected
with cDNA coding for rat CRF receptor, type 1 (rCRF
1
)(A),ormouse
CRF receptor, type 2b (mCRF
2b
)(B).Displacement was by ATB-
[His12]Svg
(12)40)
(compound 1, d), aSvg-30 (compound 2, j), ATB-
[Ala32]astressin (compound 3, m), astressin (compound 4, n), Svg
(compound 5, h), [
D
-Phe11]Svg
(11)40)
(compound 6, ·), [Tyr11,

when compared with
antisauvagine-30. However, this difference was not statisti-
cally significant.
The photoactivatable antisauvagine-30 analog was
shown to be as potent as its parent peptide when
stimulating cAMP accumulation alone or suppressing
agonist-induced second messenger production in CRF
2
expressing cells. Both compounds exhibited significantly
lower potency to suppress agonist-stimulated cAMP pro-
duction in CRF
1
cells when compared with astressin or its
photoactivatable analog thus indicating the specificity of
the novel photo ligand. It is noteworthy that the
N-terminal amino acid
D
-Phe in antisauvagine-30 can be
replaced by a phenyldiazirine,
L
-Tyr or
D
-Tyr [37] residue
without diminishing the binding affinity of the ligands to
CRF
2
receptors. A combination of an aromatic with a
heteroaromatic ring at the N-terminus increases the
binding affinity of antisauvagine-30 analogs to CRF
2

histidine motif of the peptide ligand is then presented to a
core within the transmembrane domain of the receptor.
Subsequent change of the receptor-ligand complex may
trigger G
s
-protein coupling and intracellular second mes-
senger production. This effect has been found to be more
pronounced in receptor–ligand complexes with photoacti-
vatable antisauvagine-30 or the previously described
Svg
(11)40)
when compared with antisauvagine-30 thus
indicating that the transmembrane binding pocket of the
CRF
1
receptor may discriminate between the size and
charge of the N terminal dipeptide fragment of antisauv-
agine-30 analogs.
However, the photoactivatable antisauvagine-30 analog
did not cross-link to membrane homogenates of cells
permanently transfected with the gene coding for CRF
1
receptor. In contrast, a photochemical cross-link to a
protein with a molecular weight of 66 kDa was formed in
photoaffinity labeling studies with the CRF
2
receptor. The
size of the cross-link was in agreement with the careful
analysis of a chemical cross-link obtained from earlier
studies with radioactively labeled sauvagine [12]. Formation

cells HEK-rCRF
1
cells
cAMP prod.
Antag./Svg
k
Rel potency
Antag
l
cAMP prod.
Antag./oCRF
k
Rel potency
Antag
l
1 ATB-[His12]Svg
(12)40)
0.007 ± 0.002 0.48 ± 0.04 0.15 ± 0.01
j
0.58 ± 0.04
2[
D
-Phe11,His12]Svg
(11)40)
0.004 ± 0.001 0.42 ± 0.02 0.04 ± 0.01 0.71 ± 0.02
3 ATB-[Ala32]astressin
b,c
0.010 ± 0.001 0.30 ± 0.05 0.10 ± 0.01
j
0.11 ± 0.03

0.006 ± 0.002 0.80 ± 0.04
f
0.06 ± 0.03 0.90 ± 0.02
Control without peptide – 0.004 ± 0.001 – 0.01 ± 0.003 –
Statistically significant differences between the relative potencies of the peptides:
a
, P < 0.001 vs. 2;
b
, P < 0.0001 vs. 3 and 7;
c
, P < 0.001 vs.
3 and 10;
d
, P < 0.001 vs. 2;
e
, P < 0.0001 vs. 3 and 7;
f
, P < 0.0001 vs. 1, 2, 3, and 7;
g
, P <0.001 vs. 1;
h
, P < 0.0001 vs. 2, 6, 7, 8 and 10;
i
, P < 0.0001 vs. 1, 2, 6, 7, 8, and 10. Statistically significant differences between the relative agonist activities of the peptides:
j
, P > 0.001 vs.
control without peptide.
k
The ratio of cAMP production of transfected HEK cells stimulated by antagonist (Antag.) or Svg or oCRF served
as a measure of the intrinsic activity.

2
binding sites and elucidate its functional role in the brain
and peripheral organs.
ACKNOWLEDGEMENTS
We are grateful to Dr Frank M. Dautzenberg and Dr Andreas K. E.
Ko
¨
pke for providing the HEK-rCRF
1
cellsandDrChijenR.Linand
Dr Michael G. Rosenfeld for providing the HEK-mCRF
2b
cells.
Thomas Liepold is acknowledged for the performance of the amino-
acid analysis. Dr Klaus Eckart is acknowledged for the performance of
the mass spectrometric experiments. This work was supported by the
Max-Planck Society.
REFERENCES
1. Vale, W., Spiess, J., Rivier, C. & Rivier, J. (1981) Characterization
of a 41 residue ovine hypothalamic peptide that stimulates secre-
tion of corticotropin and b-endorphin. Science 213, 1394–1397.
2. De Souza, E.B. & Grigoriadis, D.E. (1995) Corticotropin-releas-
ing factor. Physiology, pharmacology, and role in central nervous
system and immune disorders. In Psychopharmacology: the Fourth
Generation of Progress (Bloom, F.E. & Kupfer, D.J., eds),
pp. 505–517. Raven Press NY.
3. Dautzenberg, F.M. & Hauger, R.L. (2002) The CRF peptide
family and their receptors: yet more partners discovered. Trends
Pharmacol. Sci. 23, 71–77.
4. Reul, J.M. & Holsboer, F. (2002) Corticotropin-releasing factor

Bilezikjian, L., Rivier, J., Sawchenko, P.E. & Vale, W.W. (2001)
Identification of urocortin III, an additional member of the cor-
ticotropin-releasing factor (CRF) family with high affinity for the
CRF2 receptor. Proc. Natl. Acad. Sci. USA 98, 7570–7575.
11. Hsu, S.Y. & Hsueh, A.J. (2001) Human stresscopin and stress-
copin-related peptide are selective ligands for the type 2 cortico-
tropin-releasing hormone receptor. Nat. Med. 7, 605–611.
12. Ru
¨
hmann,A.,Bonk,I.,Lin,C.R.,Rosenfeld,M.G.&Spiess,J.
(1998) Structural requirements for peptidic antagonists of the
corticotropin-releasing factor receptor (CRF): Development of
CRFR2b–selective antisauvagine-30. Proc. Natl. Acad. Sci. USA
95, 15264–15269.
13. Radulovic, J., Ru
¨
hmann, A., Liepold, T. & Spiess, J. (1999)
Modulation of learning and anxiety by corticotropin-releasing
factor (CRF) and stress: differential roles of CRF receptors 1 and
2. J. Neurosci. 19, 5016–5025.
14. Higelin,J.,Py-Lang,G.,Paternoster,C.,Ellis,G.J.,Patel,A.&
Dautzenberg, F.M. (2001)
125
I-Antisauvagine-30: a novel and
specific high-affinity radioligand for the characterization of corti-
cotropin-releasing factor type 2 receptors. Neuropharmacology 40,
114–122.
15. Brauns, O., Liepold, T., Radulovic, J. & Spiess, J. (2001) Phar-
macological and chemical properties of astressin, antisauvagine-30
Fig. 3. Photoaffinity cross-linking of radioactively labeled compound 3 to

or in the presence of 10 l
M
of antisauvagine-30 (aSvg-30) (lane 2).
Molecular masses of prestained markers are indicated on the right and
left.
Ó FEBS 2002 Photoactivatable CRF
2
receptor antagonist (Eur. J. Biochem. 269) 5293
and a-helCRF: significance for behavioral experiments. Neuro-
pharmacology 41, 507–516.
16. Takahashi, L.K., Ho, S.P., Livanov, V., Graciani, N. & Arneric,
S.P. (2001) Antagonism of CRF (2) receptors produces anxiolytic
behavior in animal models of anxiety. Brain Res. 902, 135–142.
17. Pelleymounter, M.A., Joppa, M., Ling, N. & Foster, A.C.
(2002) Pharmacological evidence supporting a role for central
corticotropin-releasing factor (2) receptors in behavioral, but not
endocrine, response to environmental stress. J. Pharmacol. Exp.
Ther. 302, 145–152.
18. Cullen, M.J., Ling, N., Foster, A.C. & Pelleymounter, M.A.
(2001) Urocortin, corticotropin releasing factor-2 receptors and
energy balance. Endocrinology 142, 992–999.
19. Pelleymounter, M.A., Joppa, M., Carmouche, M., Cullen, M.J.,
Brown, B., Murphy, B., Grigoriadis, D.E., Ling, N. & Foster,
A.C. (2000) Role of corticotropin-releasing factor (CRF) receptors
intheanorexicsyndromeinducedbyCRF.J. Pharmacol. Exp.
Ther. 293, 799–806.
20. Lu, L., Liu, D., Ceng, X. & Ma, L. (2000) Differential roles of
corticotropin-releasing factor receptor subtypes 1 and 2 in opiate
withdrawal and in relapse to opiate dependence. Eur. J. Neurosci.
12, 4398–4404.

J. Biol. Chem. 263, 10927–10931.
27. Grigoriadis, D.E. & De Souza, E.B. (1989) Heterogeneity between
brain and pituitary corticotropin-releasing factor receptors is due
todifferential glycosylation. Endocrinology 125, 1877–1888.
28. Schuster, D.I., Probst, W.C., Ehrlich, G.K. & Singh, G. (1989)
Photoaffinity labeling. Photochem. Photobiol. 49, 785–804.
29. Guillory, R.J. (1989) Design, implementation and pitfalls of
photoaffinity labelling experiments in in vitro preparations. Gen-
eral principles. Pharmac. Ther. 41, 1–25.
30. Bayley, H. (1987) Diazirines as Photoactivatable Reagents in
Biochemistry. In: Chemistry of Diazirines, Vol. 2 (Liu, M.T.H.,
ed.), pp. 75–99. CRC Press, Boca Raton, FL.
31. Nassal, M. (1983) 4-(1-Azi-2,2,2-trifluoroethyl) benzoic acid, a
highly photolabile carbene generationg label readily fixable to
biochemical agents. Liebigs Ann. Chem. 1510–1523.
32. Ru
¨
hmann, A. & Wentrup, C. (1994) Synthesis of a photo-
activatable 9-IX-oleic acid for protein kinase C labeling. Tetra-
hedron 50, 3785–3796.
33. Ru
¨
ckert, Y., Rohde, W. & Furkert, J. (1990) Radioimmunoassay
of corticotropin-releasing hormone. Exp. Clin. Endocrinol. 96,
129–137.
34. Vale, W., Vaughan, J., Yamamoto, G., Bruhn, T., Douglas, C.,
Dalton,D.,Rivier,C.&Rivier,J.(1983)Assayofcorticotropin
releasing factor. Meth. In Enzymol. 103, 565–577.
35. Graham, F.L., Smiley, J., Russell, W.C. & Nairn, R. (1977)
Characteristics of a human cell line transformed by DNA from

¨
hmann, A., Bonk, I. & Ko
¨
pke, A.K.E. (1999) High-affinity
binding of urocortin and astressin but not CRF to G Protein-
Uncoupled CRFR1. Peptides 20, 1311–1319.
5294 I. Bonk and A. Ru
¨
hmann (Eur. J. Biochem. 269) Ó FEBS 2002