Structure-activity relationship studies of permeability modulating peptide AT-1002

Article abstract of DOI:10.1016/j.bmcl.2008.07.028

AT-1002 a 6-mer synthetic peptide belongs to an emerging novel class of compounds that reversibly increase paracellular transport of molecules across the epithelial barrier. The aim of this project was to elaborate on the structure-activity relationship of this peptide with the specific goal to replace the P2 cysteine amino acid. Herein, we report the discovery of peptides that exhibit reversible permeability enhancement properties with an increased stability profile.

Full text of DOI:10.1016/j.bmcl.2008.07.028

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4584–4586
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure–activity relationship studies of permeability modulating
peptide AT-1002
*
Min Li, Ed Oliver, Kelly M. Kitchens, John Vere, Sefik S. Alkan, Amir P. Tamiz
Alba Therapeutics Corporation, 800 West Baltimore Street, Suite 400, Baltimore, MD 21201, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
AT-1002 a 6-mer synthetic peptide belongs to an emerging novel class of compounds that reversibly
increase paracellular transport of molecules across the epithelial barrier. The aim of this project was to
elaborate on the structure–activity relationship of this peptide with the specific goal to replace the P2
cysteine amino acid. Herein, we report the discovery of peptides that exhibit reversible permeability
enhancement properties with an increased stability profile.
Received 12 May 2008
Revised 8 July 2008
Accepted 10 July 2008
Available online 15 July 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Peptide
Drug delivery
Structure–activity relationship studies
Paracellular permeability
Innovative drug delivery approaches have been developed to
administer molecules with suboptimal absorption profile.1 How-
ever, the utility of the paracellular transport mechanism for drug
delivery has remained unsatisfactory. This is mainly due to limited
understanding of tight junction (TJ) physiology and more specifi-
cally to lack of access to reagents capable of increasing TJ perme-
ability without irreversibly compromising barrier integrity and
function. Recent work by Fasano and coworkers on Zonula occlu-
dens toxin (ZOT) has shed light on the intricate mechanism in-
volved in the regulation of TJ permeability.2 ZOT, a 45-kDa
protein derived from Vibrio cholerae, has been shown to reversibly
increase TJ permeability in a dose-dependent manner.3 Abundant
evidence suggests that ZOT is a modulator of TJ permeability.4 Re-
cent structure activity relationship studies using ZOT resulted in a
In further investigations we discovered a dose- and time-
dependent dimerization in a variety of biologically relevant condi-
tions (Fig. 1). Hence, data obtained from biological experiments
may not accurately reflect the activity and the mechanism of action
of AT-1002. Therefore, it was imperative to identify more stable
non-dimerizing analogs of this peptide to allow additional proof
HN
NH2
NH
O
O
O
H
N
H
N
H
N
H2N
N
H
N
H
OH
small 12-kDa fragment (
D
G) that exhibits similar activity in vitro
G has proven to be a versa-
O
O
O
SH
and in vivo when compared to ZOT.5
D
AT-1002
tile reagent useful for drug delivery in a variety of in vivo models
including intranasal, intra-duodenal and oral delivery.6 Additional
advances in structure–activity relationship studies have led to
identification of AT-1002 as the active fragment of
DG with mini-
mum structure requirement for in vitro and in vivo permeability
modulation activity.7
During our recent studies with AT-1002, we discovered that this
peptide can undergo Cys–Cys dimerization in a variety of in vitro
and in vivo conditions, which may hamper the utility of this
compound.
* Corresponding author. Tel.: +1 410 319 0780; fax: +1 410 244 8616.
Figure 1. Structure of AT-1002 and its overlay HPLC profile at different time
E-mail address: atamiz@albatherapeutics.com (A.P. Tamiz).
intervals: 1.0 h (green), 2.0 h (purple), 3.0 h (pink), and 8.0 h (dark).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.07.028

M. Li et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4584–4586
4585
Table 1
of concept studies. Herein, we report a focused structure–activity
Lucifer yellow (LY) permeability and TEER assay of AT-1002 and its analogs 1–33
relationship study that led to discovery of new generation non-
dimerizable permeability enhancers. Compounds were prepared
using solid phase peptide chemistry. Individual steps such as cou-
pling, deprotection, and cleavage were performed following a re-
ported protocol8 as depicted in Scheme 1.
Little is known about the role of the P2 cysteine in AT-1002 per-
meability modulating function. Therefore, a variety of amino acids
were incorporated into the P2 position to search for the best cys-
teine surrogate(s) (Table 1).
Examples include not only naturally occurring amino acids, but
also examples of unnatural amino acids. For comparison and con-
trol purposes, AT-1002 disulfide dimer and its ‘all-carbon’ analog
were synthesized as depicted in Scheme 2.
In order to screen compounds for their functional permeability
activity, a versatile in vitro permeability assay was implemented
using Caco-2 cell monolayers as previously described.9 Table 1 de-
picts permeability data as the enhancement fold of LY permeability
from the apical to basolateral direction in the presence of the pep-
tide compared to untreated control cells. TEER data are presented
as the percent TEER of control compared to untreated control cells.
Briefly, Lucifer yellow (LY) flux was measured across Caco-2 cell
monolayers in the presence of peptides (7.0 mM) introduced from
the apical compartment in triplicates. Permeability experiments
were conducted in a humidified atmosphere of 37 °C while main-
taining sink conditions. Samples were collected from the receiver
chamber at t = 60, 120, and 180 min. Cell monolayer integrity
was monitored using an epithelial voltohmmeter to measure trans-
epithelial resistance (TEER) at t = 0, 60, 120, and 180 min. TEER
measurements were used as an indicator of cell monolayer integ-
rity during compound treatment, and TEER values typically reduce
as paracellular permeability increases. The following discussion on
structure-activity relationship is based on the enhancement of LY
permeability. Compounds that enhanced LY permeability more
than 3-fold are considered active.
Peptide
Sequence or P2
modification
Enhancement fold
LY Permeability
% of control
TEER
AT-1002
1
2
3
4
5
6
7
8
FCIGRL
FCIGR
FCIG
CIGRL
IGRL
3-Me-Phe
4-CN-Phe
t-Bu-Gly
Tyr
Arg
Lys
Asp
Glu
His
Dab
Thi
Ser
Thr
Abu
Allyl-Gly
Asn
Gln
52 12
31 8.7
2.2 1.3
NT
11 0.6
21 0.9
103 5.2
NT
0.9 0.2
0.6 0.1
0.3 0.004
6.7 1.0
0.6 0.2
2.2 0.2
0.4 0.2
0.7 0.4
1.3 0.7
1.0 0.2
2.1 1.5
1.4 0.6
2.1 1.5
3.9 2.8
1.1 0.1
103 26
1.5 0.5
1.5 0.4
5.7 4.5
1.0 0.05
2.1 1.4
1.1 0.1
0.7 0.1
1.0 0.2
0.6 0.05
1.6 0.7
0.5 0.01
1.9 0.5
0.8 0.01
0.7 0.1
70 2.0
101 7.1
118 0.2
35 1.6
118 0.2
102 21
110 3.6
148 25
104 26
102 2.8
94 16
119 22
94 16
66 27
96 3.0
20 5.2
83 11
112 6.0
50 19
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Val
Leu
Cha
121 4.5
68 40
Met(O)
Met(O)2
Styryl-Gly
Cyclopropyl-Ala
Nva
106 0.3
110 2.6
113 1.3
111 0.4
98 22
148 39
115 4.8
130 10
137 4.5
2-Pyridyl-Ala
4,5-Dehyro-Leu
(
D)-Allyl-Gly
Disulfide
Dab, 2,4-diaminobutyric acid; Thi, 2-thienylalanine; Abu, 2-aminobutyric acid;
Cha, 3-cyclohexylalanine; Met(O), methionine sulfoxide; Met(O)2, methionine
sulfone; Nva, 2-aminopentanoic acid. NT indicates the compound was not tested.
As predicted, synthetically prepared disulfide dimer (33) and
all-carbon version (34) were inactive in our assay. Table 1 depicts
truncated peptide analogs of the parent structure (AT-1002).
Removal of Leu (1) did not affect the peptide activity when com-
pared to the parent compound. However, further deletion of amino
acids from the C-terminus (2) or N-terminus (4) rendered these
peptides inactive. For the purpose of this study, we sought to ex-
plore cysteine modifications with the 6-mer parent structure.
Phe
O
Ile
O
Arg
O
H
N
H
N
H
N
FmocHN
N
H
N
O
H
O
O
O
Leu
a, b, c
O
Phe
O
Ile
Arg
O
Ile
O
O
Arg
O
H
N
H
N
H
N
H
N
H
N
P6-P3
H2N
H2N
N
H
N
OH
OH
HO
H2N
N
H
O
O
H
O
O
O
Leu
Leu
O
O
Leu
X
X
O
O
O
H
N
H
N
Phe
O
Ile
Arg
O
H
N
H
N
H
N
P2-P1
N
H
N
H
N
H
H2N
N
H
N
H
Phe
O
Ile
O
Arg
O
O
P2
O
O
Leu
33 X = S
AT1002 P2 = CH2SH
34 X = CH2
d
Phe
O
Ile
O
Arg
O
H
H
H
Cleavage
N
N
N
Scheme 2. Synthesis of AT-1002 dimers. Reagents and conditions: (a) second
generation Grubb’s reagent, CH2Cl2, reflux, 24 h; (b) H2, Pd(OH)2, MeOH, rt,
overnight; (c) deprotection and cleavage, 12% (four steps overall); (d) 1% H2O2,
CH3CN/H2O (50:50, v/v), rt, 2 h, 100%.
H2N
N
N
OH
H
H
O
P2
O
O
Leu
Compounds 5-32
Scheme 1. Synthesis of AT-1002 analogs. Reagents and conditions: coupling: resin
(1.0 equiv), amino acids (3.0 equiv), HBTU (3.0 equiv), HOBt (3.0 equiv), DIEA
(6.0 equiv), DMF, rt, 3 h; Deprotection: piperidine:DMF (20:80, v/v), rt, 1 h;
cleavage: TFA/TES/H2O (95:2.5:2.5, v/v/v), rt, 2 h.
Hydrophobic amino acids at the P2 position (5–8) with excep-
tion of t-Butyl-Gly (7) were inactive when compared to

4586
M. Li et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4584–4586
References and notes
1. (a) Fasano, A. Trends Biotechnol. 1998, 16, 152; (b) Fasano, A. J. Pharm. Sci. 1998,
87, 1351; (c) Kreuter, J. J. Anat. 1996, 189, 503; (d) Michel, C.; Aprahamian, M.;
Defontaine, R.; Couvreur, P.; Damge, C. J. Pharm. Pharmacol. 1991, 43, 1; (e) Ebel,
J. P. Pharm. Res. 1990, 7, 848; (f) Le Fevre, M. E.; Joel, D. D. In Intestinal
toxicologyToxicology; Schiller, C. M., Ed.; Raven Press: New York, 1984; pp 45–
56.
2. (a) Fasano, A.; Baudry, B.; Pumplin, D. W.; Wasserman, S. S.; Tall, B. D.; Ketley,
J.; Kaper, J. B. Proc. Natl. Acad. Sci U.S.A. 1991, 5242; (b) Baudry, B.; Fasano, A.;
Ketley, J.; Kaper, J. B. Infect. Immun. 1992, 60, 428; (c) Fasano, A.; Uzzau, S. J. Clin.
Invest. 1997, 99, 1158.
3. (a) Fasano, A.; Fiorentini, C.; Donelli, G.; Uzzau, S.; Kaper, J. B.; Margaretten, K.;
Ding, X.; Guandalini, S.; Comstock, L.; Goldblum, S. J. Clin. Invest. 1995, 97, 710;
(b) Johnson, J. A.; Morris, J. G., Jr.; Kaper, J. B. J. Clin. Microbiol. 1993, 31, 732; (c)
Izumi, Y.; Hirose, T.; Tamai, Y.; Hirai, S.; Nagashima, Y.; Fujimoto, T.; Tabuse, Y.;
Kemphues, K. J.; Ohno, S. J. Cell. Biol. 1998, 143, 95.
Figure 2. Cell viability in using Caco-2 cells. *Significant reduction in cell viability
(p < 0.05). Results are reported as means standard deviation (SD) (n = 3).
4. (a) Wang, W.; Uzzau, S.; Goldblum, S. E.; Fasano, A. J. Cell. Sci. 2000, 113, 4435;
(b) Fasano, A.; Not, T.; Wang, W.; Uzzau, S.; Berti, I.; Tommasini, A.; Goldbulm,
S. E. Lancet 2000, 355, 1518; (c) Fasano, A.; Uzzau, S.; Fiore, C.; Margaretten, K.
Gastroenterology 1997, 112, 839; (d) Karasawa, T.; Tatsuya, M.; Kurazona, H.;
Nair, G. B. Microbiol. Lett. 1993, 106, 143.
5. (a) Allen, J. C.; Walker, R.; Rosen, G. J. Clin. Oncol. 1998, 6, 649; (b) Borst, P.;
Evers, R.; Kool, M.; Wijnholds, J. J. Natl. Cancer Inst. 2000, 92, 1295; (c)
Breedveld, P.; Zelcer, N.; Pluim, D.; Sonmezer, O.; Tibben, M. M.; Beijnen, J. H.;
Schinkel, A. H.; van Tellingen, O.; Borst, P.; Schellens, J. H. Cancer Res. 2004, 64,
5804.
6. (a) Menon, D.; Karyekar, C. S.; Fasano, A.; Lu, R.; Eddington, N. D. Int. J. Pharm.
2005, 306, 122; (b) Salama, N. N.; Fasano, A.; Thakar, M.; Eddington, N. D. J.
Pharm. Sci. 2004, 93, 1310; (c) Salama, N. N.; Fasano, A.; Lu, R.; Eddington, N. D.
Int. J. Pharm. 2003, 251, 113; (d) Song, K. H.; Fasano, A.; Eddington, N. D. Eur. J.
Pharm. Biopharm. 2008, 691, 231; (e) Song, K. H.; Fasano, A.; Eddington, N. D.
Int. J. Pharm. 2008, 351(1-2), 8.
7. (a) Li, M.; Oliver, E.; Vere, J.; Kitchens, K. M.; Ginski, M.; Gopalakrishnan, S.;
Pandey, N.; Alkan, S. S.; Paterson B.; Tamiz, A. P. Abstract of Papers, 235th
National Meeting of the American Chemical Society, New Orleans, LA; America
Chemical Society: Washington, DC, 2008.; (b) Kitchens, K. M.; Gopalakrishnan,
S.; Vere, J.; Carrasco, R.; Li, M.; Oliver, E.; Ginski, M.; Pandey, N.; Paterson, B.;
Alkan, S. S.; Tamiz, A. P. Abstract of Papers, 235th National Meeting of the
American Chemical Society, New Orleans, LA; America Chemical Society:
Washington, DC, 2008.
AT-1002. Neither the basic (9, 10, and 13) nor the acidic amino acid
(11 and 12) surrogates exhibited activity in our assay. Surprisingly,
compounds 14 and 16, which bear cysteine isosteres Dab and Ser,
respectively, were also inactive. However, Thr (17) and Val (22)
analogs did induce an increase in LY flux. Compound 19 was the
most potent peptide among all the synthesized analogs (Table 1).
Encouraged by this finding, multiple P2 derivatives of 19 were pre-
pared and evaluated for permeability modulation effect in this
assay. Surprisingly, none of the additional analogs of this parent
structure such as, Nva (30), terminal olefin rigidified analogs 28,
30, and 31 exhibited activity in our assay including (D)-Ally-Gly
(32). The data obtained to-date (Table 1) unambiguously suggests
that the P2 terminal olefin function is an imperative structure unit
for maintaining or enhancing the tight junction modulating activ-
ity. Potential dimerization of compound 19 was evaluated using
HPLC analysis.10 As predicted, this molecule is void of dimerization
and is stable as a monomer.
The effect of AT-1002 and compound 19 on cell viability was
measured using the CellTiter-GloÒ cell viability assay. After 3 h of
treatment, AT-1002 and Compound 19 did not significantly reduce
cell viability compared to untreated control cells (Fig. 2). These
data suggests that these compounds enhance LY permeability
due to their permeability modulation activity and not due to cell
viability reduction.
The data reported herein suggest that P2 functionalization can be
exploited to generate potent and stable analogs of AT-1002. Applica-
tionoftheP2singlemutationapproachdescribedhereinhasresulted
in compounds that are void of undesirable dimerization. In general,
peptidesreported in this study do notsignificantly reducecell viabil-
ity. Also, the effects are reversible as shown by compound removal
experiments (data not shown). The incorporationof allylglycine (19)
deliversapromisingpermeabilitymodulatorwhoseactivityandnon-
dimerizingpropertywarrantsfurtherstudiesasapotentialnewagent.
Additional studies including in vivo testing are underway with com-
pound 19 in our group and will be reportedin due course.
8. Cammish, L. E.; Kates, S. A. In Fmoc Solid Phase Peptide Synthesis: A Practical
Approach; Chan, W. C., White, P. D., Eds.; Oxford University: Oxford, 2000; pp
277–279.
9. (a) Hubatsch, I.; Ragnarsson, E. G.; Artursson, P. Nat. Protoc. 2007, 2, 2111; (b)
Sambuy, Y.; De Angelis, I.; Ranaldi, G.; Scarino, M. L.; Stammati, A.; Zucco, F.
Cell. Biol. Toxicol. 2005, 21, 1; (c) Yamashita, S.; Konishi, K.; Yamazaki, Y.; Taki,
Y.; Sakane, T.; Sezaki, H.; Furuyama, Y. J. Pharm. Sci. 2002, 91, 669.
10. Each peptide sample was analyzed using LC–MS with a Waters 2695 HPLC
system interfaced with
analytical separation was carried out using a Phenomenex Luna-C18 column
m) and a dual solvent system consisting of A (0.1% TFA in
a Waters Micromass ZQ mass spectrometer. An
(150 Â 4.6 mm, 5.0
l
water) and B (0.1% TFA in acetonitrile). A 15 min gradient from 0% to 40% B at a
flow rate of 1.0 mL/min was employed for general analysis purpose. The MS
measurement was performed using a positive electrospray ionization mode
and the cone voltage was optimized for the presence of maximum precursor
ion signal [M+H]+. The purity was evaluated by using LC–MS and comparing
the UV absorbance of the sample solution with the solvent blank using a
Waters 2695 HPLC equipped with
a Waters 2996 diodearray detector at
different wavelengths. The physicochemical data of AT-1002, compounds 35
and 19 are as follows: AT-1002: Rt = 8.1 min, [M+H]+ 708.2 (Calcd 708.3);
compound 35: Rt = 8.9 min, [(M+2H)/2]+ 707.6 (Calcd 707.3); compound 19:
Rt = 7.8 min, [M+H]+ 702.2 (Calcd 702.4).