P2-P4 Macrocyclic inhibitors of hepatitis C virus NS3-4A serine protease

Article abstract of DOI:10.1016/j.bmcl.2006.05.022

Synthesis and HCV NS3 serine protease inhibitory activity of 4-hydroxyproline derived macrocyclic inhibitors and SAR around this macrocyclic core is described in this communication. X-ray structure of inhibitor 38 bound to the protease is discussed.

Full text of DOI:10.1016/j.bmcl.2006.05.022

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3960–3965
P2–P4 Macrocyclic inhibitors of hepatitis C
virus NS3-4A serine protease
Ashok Arasappan,* F. George Njoroge, Kevin X. Chen, Srikanth Venkatraman,
Tejal N. Parekh, Haining Gu, John Pichardo, Nancy Butkiewicz,
Andrew Prongay, Vincent Madison and Viyyoor Girijavallabhan
Schering Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
Received 29 March 2006; revised 5 May 2006; accepted 5 May 2006
Available online 30 May 2006
Abstract—Synthesis and HCV NS3 serine protease inhibitory activity of 4-hydroxyproline derived macrocyclic inhibitors and SAR
around this macrocyclic core is described in this communication. X-ray structure of inhibitor 38 bound to the protease is discussed.
Ó 2006 Elsevier Ltd. All rights reserved.
Hepatitis C virus (HCV) infection is a global health cri-
sis leading to liver cirrhosis, hepatocellular carcinoma,
and liver failure in humans.1 It has been estimated that
3% of the human population worldwide is chronically
infected with HCV.2 a-Interferon monotherapy and
a-interferon–ribavirin combination therapy are the stan-
dard treatment for this infection.3 While therapeutic
effectiveness has been improved with introduction of
pegylated version of a-interferon, PEG-INTRON and
PEGASYS, it is still far from ideal.4 Hence there is an
unmet medical need to discover and develop new and
more effective protocol for the treatment of HCV
infection.
also reported their results using a similar strategy.9
More recently we disclosed that introduction of an
appropriate P02 residue allowed us to truncate these
inhibitors at P3 without adversely affecting the potency
(compd 2, Fig. 1).10 X-ray crystal structure of inhibitor
bound complex revealed that the P1-P02 moiety formed a
C-clamp over Lys 136 of the protease, resulting in exten-
P2
O
O
H
N
H
N
P'
P6
O
P4
HN
H
N
H
N
O
P1
O
CapHN
N
H
1
O
P5
O
P3
Since identification of this virus, the NS3 serine protease
contained within the N-terminal region of the NS3 pro-
tein has been studied extensively.5 This chymotrypsin-
like serine protease is implicated in viral replication
and therefore, an attractive target for HCV antiviral
therapeutics.6 Proof of concept studies in humans have
validated this hypothesis.7
Ala
156
O
O
H
H
N
N
NH2
N
N
H
H
O
N
O
O
O
O
2, Ki* = 340 nM
O
P3
Our earlier exploration toward developing inhibitors of
HCV NS3 serine protease resulted in P6–P0 hexapeptide
derivatives of type 1, containing a-ketoamide serine
trap, with good binding potency.8 Other groups have
n
O
Y
O
O
H
H
N
N
N
H
N
N
H
O
O
O
N
X
Keywords: HCV NS3-4A serine protease inhibitor; HCV protease in-
hibitor; Macrocycle; Ketoamide.
O
3
O
P3
*
Corresponding author. Tel.: +1 908 740 2607; fax: +1 908 740
7152; e-mail: ashok.arasappan@spcorp.com
Figure 1. Design of macrocyclic inhibitors.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.022

A. Arasappan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3960–3965
3961
sive hydrophobic interaction which translated into im-
proved binding potency. Further analysis of the X-ray
structure revealed the presence of Ala 156 in close prox-
imity to the P2/P3 capping region, and thus could be
exploited advantageously to improve the potency. We
therefore decided to cyclize the P2 residue to the P3 cap-
ping moiety, thus forming P2–P4 macrocyclic inhibitors.
the stage for macrocyclization, which was realized via
Mitsunobu protocol using ADDP and triphenylphos-
phine.14 Cleavage of the tert-butyl ester of 7 under acidic
conditions provided the 16-membered carbamate macro-
cycle 8. The choice of tert-butyl ester protecting groupwas
essential so that the deprotection could be performed un-
der acidic conditions. Attempted removal of the corre-
sponding methyl ester (structure not shown) using
LiOH conditions afforded the acid with concomitant ring
opening of the macrocyclic phenyl carbamate. Urea mac-
rocycle 13, shown in Scheme 1, was prepared in an analo-
gous manner employing the activated carbamate 10.
Since HCV NS3 serine protease inhibitors containing P2
4-hydroxy proline derivatives have been reported,11 we
envisioned using the hydroxyl group as a handle to
introduce linkers. The choice of P3 capping moiety to
be cyclized was more challenging, as most of our acyclic
derivatives contained t-Boc or i-Boc groups as P3 caps.
Our early investigations in the P3 capping area had iden-
tified some substituted benzamido functionality as a
good replacement for i-Boc residue.12 Based on these
observations we planned on incorporating an aromatic
group as the P3 capping moiety, with suitable substitu-
tion pattern that will allow for cyclization onto the P2
proline via an appropriate linker.
For the synthesis of N-methyl urea macrocycles, 24 and
25, the isocyanate surrogate (4-nitrophenyl carbamoyl
moiety) was introduced at the P3 nitrogen terminus.
Reaction of the N-methylaniline derivative 15 with the
activated carbamoyl derivative 14, resulted in the forma-
tion of N-methyl urea 16. Hydrolysis of the ethyl ester of
16, and subsequent coupling with 4-hydroxyproline de-
rived amino ester 17 resulted in the macrocyclic precur-
sor 18. Mitsunobu protocol described above was then
utilized to synthesize the N-methyl urea macrocycle 22.
Deprotection of the tert-butyl ester using TFA provided
acid 24. Similar strategy employing the dipeptide de-
rived 4-nitrophenyl carbamate 19 and N-methylbenzyl-
amine derivative 20 afforded the 17-membered N-
methylurea macrocycle 25 (Scheme 2).
Furthermore, we opted to cap the P02 carboxyl terminus
with N,N-dimethylamide residue, since earlier SAR
studies in the P02 area indicated this functionality provid-
ed superior C-terminus cap compared to parent –NH2.12
Modeling results suggested that 16- to 18-membered
ring size would provide maximum contact with the en-
zyme surface. Recently, we disclosed our efforts directed
toward effective implementation of this macrocycliza-
tion strategy.13 Herein, we report further modifications
around the macrocyclic core leading to potent HCV
NS3 serine protease inhibitors.
Having constructed the macrocyclic acids, installation
of the P1–P0 residue 2613 was done under coupling con-
dition using EDCI, HOOBt, and NMM. The resulting
mixture of hydroxyl amide diastereomers were oxidized
using Dess–Martin’s periodinane16 to provide the P2–P4
macrocyclic a-ketoamide targets 27–30 (Scheme 3).
Synthetic approach for preparation of the designed mac-
rocycles started from previously described dipeptide
derivative 414 (Scheme 1). Treatment of 4 with activated
carbonate 515 resulted in formation of carbamate 6.
Removal of benzyl protecting group on the linker set
The construction of amide derived macrocycles is shown
in Scheme 4. The dipeptide derivative 31 was coupled
with commercially available acids 32 and further elabo-
O
O
O
O
OBn
OBn
O
HO
O
O
CO2tBu
CO2tBu
N
N
CO2R1
H
N
N
HCl.H2N
O
a
H
O
NO2
O
5
N
O
O
7, R1 = tBu
8, R1 = H
O
b
O
c
OH
6
4
O
O
O
O
H
N
OBn
BnO
O
OBn
HCl.H2N
CO2R2
CO2Me
N
N
CO2Me
N
H
N
H
N
H
N
O
HN
O
NO2
O
O
10
b
12, R2 = Me
13, R2 = H
O
O
a
d
OBn
11
9
Scheme 1. Reagents and conditions: (a) Et3N, imidazole (cat.), CH2Cl2/DMF (20–38%); (b) see Ref. 14; (c) TFA/CH2Cl2 (10%, three steps); (d) aq
1 M LiOH/THF/MeOH (15%, three steps).

3962
A. Arasappan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3960–3965
BnO
O
H
N
Me
H
O
CO2Et
CO2tBu
N
N
CO2Et
Me
N
N
b
H
a
O
N
O
O2N
O
O
BnO
NHMe
O
18
OBn
14
15
OBn
CO2tBu
.HCl
16
N
H
OBn
17
d
NHMe
BnO
BnO
O
BnO
O
O
O
20
CO2Me
CO2Me
CO2R3
N
Me
N
N
d
H
N
H
H
c
R4
O
N
N
N
O
BnO
O
O
O
19
O
O
O2N
21
22, R3 = tBu, R4 = NMe
23, R3 = Me, R4 = CH2NMe
e
24, R3 = H, R4 = NMe
25, R3 = H, R4 = CH2NMe
Scheme 2. Reagents and conditions: (a) Et3N, imidazole (cat.), CHCl3, 50 °C (56%); (b) i—aq LiOH, EtOH; ii—17, EDCI, HOOBt, NMM, CH2Cl2/
DMF (34%, two steps); (c) Et3N, imidazole (cat.), CH2Cl2 (63%); (d) see Ref. 14; (e) for 24, TFA/CH2Cl2 (21%, three steps); for 25, aq 1 M LiOH/
MeOH (35%, three steps).
rated to provide the macrocyclic acids 34.14 Conversion
of these acids 34 to the target compounds of type 35
essentially followed procedures described in Scheme 3.
HCV NS3 serine protease inhibitory data17 for the mac-
rocyclic targets were obtained using the continuous
spectrophotometric assay described earlier18 and the re-
sults are summarized in Table 1. Cyclization of the P2
moiety to the P3 capping group with an appropriate
O
O
a
CO2H
N
H
n
O
R5
N
O
O
O
OH
O
Ph
n
H
O
OBn
HCl.H2N
N
N
CO2Me
N
H
N
X
CO2R6
a
N
O
O
HCl.H2N
H
26
O
N
O
m
8, R5 = O (P3 = tBu, not Chx)
13, R5 = NH
24, R5 = NMe
25, R5 = CH2NMe
P3
O
P3
31
OH or OBn
33, R6 = Me
X
CO2H
b
m
34, R6 = H
32
O
O
O
O
H
N
H
N
n
O
N
N
N
O
H
H
O
O
O
O
O
R5
N
H
H
N
O
c
N
N
X
N
H
N
O
H
N
O
O
O
O
m
27, R5 = O (P3 = tBu, not Chx)
28, R5 = NH
O
P3
35
29, R5 = NMe
n = 0-2; m = 0-2; X = H, Me, OMe
30, R5 = CH2NMe
Scheme 4. Reagents and conditions: (a) i—32, EDCI, HOOBt, NMM,
CH2Cl2/DMF; ii—see Ref. 14; (b) aq 1 M LiOH/THF/MeOH; (c) i—
26, EDCI, HOOBt, NMM, CH2Cl2/DMF; ii—Dess–Martin’s period-
inane, CH2Cl2 (50–60%, two steps).
Scheme 3. Reagents and conditions: (a) i—26, EDCI, HOOBt, NMM,
CH2Cl2/DMF; ii—Dess–Martin’s periodinane, CH2Cl2 (50–60%, two
steps).

A. Arasappan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3960–3965
3963
Table 1.
Table 1 (continued)
Compound Macrocycle
Kꢀi (nM)
O
O
Ph
H
H
macrocycle
N
N
NMe2
N
H
O
O
O
O
O
Compound Macrocycle
Kꢀi (nM)
39a
32
N
H
O
N
O
O
O
O
27
28
29
30a
200
N
H
O
N
O
O
O
O
O
40a
30
N
H
O
N
O
O
O
O
N
1600
H
N
O
HN
O
O
O
O
41a
42
43
44
12
19
30
6
N
H
O
N
O
O
O
O
N
90
H
O
N
N
O
Me
O
O
O
P3 =
N
H
O
O
O
N
O
CO2H
O
P3
N
26
H
O
N
N
O
Me
O
O
O
P3 =
N
H
O
O
N
O
CO2H
O
O
P3
36a
37a
38a
210
N
H
O
N
O
O
O
O
N
H
O
N
O
O
O
O
N
8
H
O
N
CO2H
O
a (S)-Isomer at P1 center.
O
O
O
linker resulted in the carbamate macrocycle, 27, with a
binding potency of 200 nM. While this result was very
encouraging in that the potency was retained, further
improvement was needed.
N
H
25
O
N
O
O
We first decided to concentrate on modifying the P3 cap-
ping linkage in order to improve the potency and also

3964
A. Arasappan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3960–3965
the metabolic stability of these inhibitors. Hence the
urea macrocycle derived target 28 was next synthesized.
Unfortunately inhibitor 28 had lost most of the potency
which maybe attributed to high degree of conformation-
al restriction around the macrocyclic region. Interesting-
ly, alkylation (i.e., methylation) of the distal urea
nitrogen resulted in restoring, if not improving, the
potency (compd 29, Kꢀi ¼ 90nM). Increasing the ring
size provided macrocyclic inhibitor 30, with further
improvement in potency. Since our early work in the
P3 capping region with acyclic inhibitors12 indicated that
substituted benzamido functionality was well tolerated,
we then investigated P3 amide linked macrocyclic cores.
In order to maintain the ring size as 16-membered, we
prepared the benzamido P3 capped inhibitor 36, con-
taining four-carbon linker. Encouragingly, this inhibitor
36 was equipotent to the parent carbamate macrocyclic
inhibitor, 27. Moving the aromatic group one-carbon
away from the P3 amide linkage, while retaining the ring
size, resulted in target 37 with excellent binding potency,
Kꢀi ¼ 8 nM.13 Further adjustment of the aromatic group
resulted in 38, another potent inhibitor with
Kꢀi ¼ 25 nM. Having established the requirements for
the macrocyclic core, which resulted in an excellent
HCV NS3 serine protease inhibitor (37), we then set
out to study the effect of substitution on the aromatic
ring. While introduction of 3-methoxy (compd 39) or
3-methyl (compd 40) functionality seemed to affect the
binding in a negative way, the 2-methyl group (compd
41) essentially retained the potency. The effect of substi-
tution on the P3 cyclohexyl ring was investigated next.
Preparation of the 4-carboxyl substituted cyclohexylgly-
cine derivatives has been reported previously.19
Introduction of trans or cis-carboxyl group at the
4-position of cyclohexyl P3 residue resulted in minimal
loss in potency (compounds 42 and 43) compared to
the parent amide linked inhibitor, 37. However,
3-carboxyl cyclohexylglycine at P3 afforded inhibitor
44 which was equipotent to 37 (Kꢀi ¼ 6 nM). Thus, the
hypothesis of appropriately positioned carboxyl groups
for interaction with the amino sidechain of Lys 136
did not prove to be the case.
Figure 2. X-ray structure of 38 bound to the protease.
In summary we have established that cyclization of P2
moiety with P3 capping residue via a suitable linker pro-
vided additional binding contact with the protease sur-
face. Targets derived from these designed macrocycles
exhibited potent HCV NS3 serine protease inhibitory
activity. While the carbamate and N-methylurea linked
macrocycles provided a few fold improvement in bind-
ing, the 16-membered amide linked macrocyclic targets
resulted in inhibitors with single-digit nanomolar poten-
cy. X-ray crystal structure of one of the inhibitors (38)
bound to the protease revealed that the P region macro-
cyclic core formed a ring around Ala 156, and the P3
capped aromatic group was positioned well in the S4
cavity. Together, these hydrophobic interactions result-
ed in tremendous improvement in potency.
Acknowledgment
We thank the Structural Chemistry group for providing
NMR and Mass Spectral assistance.
References and notes
X-ray crystal structure of the inhibitor 38 bound to the
protease is shown in Figure 2.20 The peptidic core is
bound to the protease through a series of hydrogen
bonding interactions. The catalytic Ser 139 attacked
the electrophilic keto functionality in a covalent and
reversible manner. Interestingly, the amide carbonyl
group adjacent to the keto moiety occupied the oxyan-
ion hole, instead of the tetrahedral oxygen anion result-
ing from the attack of Ser 139, as expected. The propyl
residue at P1 is buried into the shallow hydrophobic S1
pocket. The cyclohexyl moiety at P3 occupied the S3
pocket, providing hydrophobic contact with the enzyme.
As observed in our previous inhibitors, the P0 residue
wrapped over the sidechain of lysine 136, resulting in
P1-P02 C-clamp, thus locking Lys 136 in place. Most
interestingly, the macrocyclic core formed a ring around
Ala 156, while the aromatic group at the P3 cap region
buried deep into the S4 pocket. These resultant
hydrophobic interactions with the enzyme translated
into improved potency.
1. (a) Cohen, J. Science 1999, 285, 26; (b) Houghten, M. In
Virology; Fields, B. N., Knipe, D. M., Howley, P. M.,
Eds.; Raven: New York, 1996; pp 1035–1058; (c) Cuth-
bert, J. A. Clin. Microbiol. Rev. 1994, 7, 505.
2. Consensus Panel. EASL International Consensus Confer-
ence on Hepatitis C, Paris, 26–28 February 1999, Con-
sensus Statement J. Hepatol. 1999, 30, 956.
3. Dymock, B. W. Emerging Drugs 2001, 6, 13, and refer-
ences cited therein.
4. Feld, J. J.; Hoofnagle, J. H. Nature 2005, 436, 967.
5. (a) De Francesco, R.; Migliaccio, G. Nature 2005, 436,
953; (b) De Francesco, R.; Tomei, L.; Altamura, S.;
Summa, V.; Migliaccio, G. Antiviral Res. 2003, 58, 1; (c)
Steinkuhler, C.; Koch, U.; Narjes, F.; Matassa, V. G.
Curr. Med. Chem. 2001, 8, 919; (d) Kwong, A. D.; Kim, J.
L.; Rao, G.; Lipovsek, D.; Raybuck, S. A. Antiviral Res.
1998, 40, 1.
6. (a) Kolykhalov, A. A.; Mihalik, K.; Feinstone, S. M.;
Rice, C. M. J. Virol. 2000, 74, 2046; (b) Bartenschlager,
R.; Lohmann, V. J. Gen. Virol. 2000, 81, 1631.

A. Arasappan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3960–3965
3965
7. (a) Llinas-Brunet, M.; Bailey, M. D.; Bolger, G.; Brochu,
C.; Faucher, A.-M.; Ferland, J. M.; Garneau, M.; Ghiro,
E.; Gorys, V.; Grand-Maitre, C.; Halmos, T.; Lapeyre-
Paquette, N.; Liard, F.; Poirier, M.; Rheaume, M.;
Tsantrizos, Y. S.; Lamarre, D. J. Med. Chem. 2004, 47,
1605; (b) Perni, R. B.; Almquist, S. J.; Byrn, R. A.;
Chandorkar, G.; Chaturvedi, P. R.; Courtney, L. F.;
Decker, C. J.; Dinehart, K.; Gates, C. A.; Harbeson, S. L.;
Heiser, A.; Kalkeri, G.; Kolaczkowski, E.; Lin, K.; Luong,
Y.-P.; Rao, B. G.; Taylor, W. P.; Thomson, J. A.; Tung,
R. D.; Wei, Y.; Kwong, A. D.; Lin, C. Antimicrob. Agents
Chemother. 2006, 50, 899; (c) Zeuzem, S.; Sarrazin, C.;
Rouzier, R.; Tarral, A.; Brion, N.; Forestier, N.; Gupta,
S.; Deckman, D.; Fellows, K.; Hussain, M.; Cutler, D.;
Zhang, J. In 56th Annual Meeting of AASLD, 2005, San
Francisco, CA, USA.
8. (a) Saksena, A. K.; Girijavallabhan, V. M.; Lovey, R. G.;
Jao, E.; Bennett, F.; McCormick, J. L.; Wang, H.; Pike, R.
E.; Bogen, S. L.; Liu, Y.-T.; Arasappan, A.; Parekh, T.;
Pinto, P. A.; Njoroge, G.; Ganguly, A. K.; Brunck, T. K.;
Kemp, S. J.; Levy, O. E.; Lim-Wilby, M. U.S. Patent
6800434 B2, 2004; Chem. Abstr. 2002, 136, 151440; (b)
Arasappan, A.; Njoroge, F. G.; Parekh, T. N.; Yang, X.;
Pichardo, J.; Butkiewicz, N.; Prongay, A.; Yao, N.;
Girijavallabhan, V. Bioorg. Med. Chem. Lett. 2004, 14,
5751.
11. (a) Poupart, M.-A.; Cameron, D. R.; Chabot, C.; Ghiro,
E.; Goudreau, N.; Goulet, S.; Poirier, M.; Tsantrizos, Y.
S. J. Org. Chem. 2001, 66, 4743; (b) Llinas-Brunet, M.;
Bailey, M.; Fazal, G.; Ghiro, E.; Gorys, V.; Goulet, S.;
Halmos, T.; Maurice, R.; Poirier, M.; Poupart, M.-A.;
Rancourt, J.; Thibeault, D.; Wernic, D.; Lamarre, D.
Bioorg. Med. Chem. Lett. 2000, 10, 2267.
12. Unpublished results from our group.
13. Chen, K. X.; Njoroge, F. G.; Arasappan, A.; Venkatr-
aman, S.; Vibulbhan, B.; Yang, W.; Parekh, T. N.;
Pichardo, J.; Prongay, A.; Cheng, K.-C.; Butkiewicz, N.;
Yao, N.; Madison, V.; Girijavallabhan, V. J. Med. Chem.
2006, 49, 995.
14. Arasappan, A.; Chen, K. X.; Njoroge, F. G.; Parekh, T.
N.; Girijavallabhan, V. J. Org. Chem. 2002, 67, 3923.
15. The activated carbonate 5 and the activated carbamates
10, 14, and 19 were prepared by treatment of the
corresponding phenol or amino derivative with commer-
cially available 4-nitrophenylchloroformate in the pres-
ence of pyridine and THF/MeCN as the solvent.
16. (a) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155;
(b) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113,
7277.
17. The HCV NS3 serine protease inhibitory value is provided
as Kꢀi . For a definition of Kꢀi and discussion, see:
Morrison, J. F.; Walsh, C. T. In Adv. Enzymol.; Meister,
A., Ed., 1988; Vol. 61, pp 201–301.
18. Zhang, R.; Beyer, B. M.; Durkin, J.; Ingram, R.; Njoroge,
F. G.; Windsor, W. T.; Malcolm, B. A. Anal. Biochem.
1999, 270, 268, For the present study, the substrate Ac-
DTEDVVP(Nva)-O-PAP was employed.
9. (a) Chen, S.-H.; Tan, S.-L. Curr. Med. Chem 2005, 12,
2317, and references cited therein; (b) Goudreau, N.;
Llinas-Brunet, M. Expert Opin. Investig. Drugs 2005, 14,
1129, and references cited therein.
10. Arasappan, A.; Njoroge, F. G.; Chan, T.-Y.; Bennett, F.;
Bogen, S. L.; Chen, K.; Gu, H.; Hong, L.; Jao, E.; Liu, Y.-
T.; Lovey, R. G.; Parekh, T.; Pike, R. E.; Pinto, P.;
Santhanam, B.; Venkatraman, S.; Vaccaro, H.; Wang, H.;
Yang, X.; Zhu, Z.; Mckittrick, B.; Saksena, A. K.;
Girijavallabhan, V.; Pichardo, J.; Butkiewicz, N.; Ingram,
R.; Malcolm, B.; Prongay, A.; Yao, N.; Marten, B.;
Madison, V.; Kemp, S.; Levy, O.; Lim-Wilby, M.;
Tamura, S.; Ganguly, A. K. Bioorg. Med. Chem. Lett.
2005, 15, 4180.
19. Venkatraman, S.; Njoroge, F. G.; Girijavallabhan, V.;
McPhail, A. T. J. Org. Chem. 2002, 67, 2686. The 3-
carboxyl substituted cyclohexylglycine was prepared in a
similar manner.
20. Crystallographic data for the inhibitor–protease complex
included in this paper has been deposited with the RCSB
Protein Data Bank as PDB ID 2GVF. The structural
details can be viewed at www.rscb.org using the ID
number above.