A general method for making peptide therapeutics resistant to serine protease degradation: Application to dipeptidyl peptidase IV substrates

Article abstract of DOI:10.1021/jm400423p

Bioactive peptides have evolved to optimally fulfill specific biological functions, a fact which has long attracted attention for their use as therapeutic agents. While there have been some recent commercial successes fostered in part by advances in large-scale peptide synthesis, development of peptides as therapeutic agents has been significantly impeded by their inherent susceptibility to protease degradation in the bloodstream. Here we report that incorporation of specially designed amino acid analogues at the P1′ position, directly C-terminal of the enzyme cleavage site, renders peptides, including glucagon-like peptide-1 (7-36) amide (GLP-1) and six other examples, highly resistant to serine protease degradation without significant alteration of their biological activity. We demonstrate the applicability of the method to a variety of proteases, including dipeptidyl peptidase IV (DPP IV), dipeptidyl peptidase 8 (DPP8), fibroblast activation protein α (FAPα), α-lytic protease (αLP), trypsin, and chymotrypsin. In summary, the "P1′ modification" represents a simple, general, and highly adaptable method of generating enzymatically stable peptide-based therapeutics.

Full text of DOI:10.1021/jm400423p

Subscriber access provided by NEW MEXICO STATE UNIV
Article
A General Method for Making Peptide Therapeutics Resistant to Serine
Protease Degradation; Application to Dipeptidyl Peptidase IV Substrates
Kathryn R. Heard, Wengen Wu, Youhua Li, Peng Zhao, Iwona Woznica, Jack H.
Lai, Martin Beinborn, David G. Sanford, Matthew T Dimare, Amrita K Chiluwal,
Diane E Peters, Danielle Whicher, James L. Sudmeier, and William W Bachovchin
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 17 Sep 2013
Downloaded from http://pubs.acs.org on September 17, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A General Method for Making Peptide Therapeutics
Resistant to Serine Protease Degradation;
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Application to Dipeptidyl Peptidase IV Substrates
Kathryn R. Heard †, Wengen Wu †, Youhua Li †, Peng Zhao †, Iwona Woznica †, Jack H. Lai †,
Martin Beinborn ‡, David G. Sanford †, Matthew T. Dimare †, Amrita K. Chiluwal †, Diane E.
Peters †, Danielle Whicher †, James L. Sudmeier †, and William W. Bachovchin *†
† Department of Biochemistry, Tufts University Sackler School of Graduate Biomedical
Sciences, 136 Harrison Avenue, Boston, Massachusetts 02111
‡ Molecular Pharmacology Research Center, Molecular Cardiology Research Institute, Tufts
Medical Center, 800 Washington Street, Boston, Massachusetts 02111
KEYWORDS: alphaꢀlytic protease (αLP), amino acid analogue, brain natriuretic peptide (BNP),
chymotrypsin, diabetes, dipeptidyl peptidase 8 (DPP8), dipeptidyl peptidase IV (DPP IV),
enterostatin (ENT), fibroblast activation protein alpha (FAPα), glucagonꢀlike peptideꢀ1 (7ꢀ36)
amide (GLPꢀ1), glucoseꢀdependent insulinotropic peptide (GIP), halfꢀlife (t ), neuropeptide Y
½
(NPY), oxyntomodulin (OXM), peptide therapeutics, protease, serine protease, trypsin
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 42
1
2
3
4
5
6
7
Abstract
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Bioactive peptides have evolved to optimally fulfill specific biological functions ꢀꢀ a fact which
has long attracted attention for their use as therapeutic agents. While there have been some
recent commercial successes fostered in part by advances in largeꢀscale peptide synthesis,
development of peptides as therapeutic agents has been significantly impeded by their inherent
susceptibility to protease degradation in the bloodstream. Here we report that incorporation of
specially designed amino acid analogues at the P1’ position, directly Cꢀterminal of the enzyme
cleavage site, renders peptides, including glucagonꢀlike peptideꢀ1 (7ꢀ36) amide (GLPꢀ1) and six
other examples, highly resistant to serine protease degradation without significant alteration of
their biological activity. We demonstrate the applicability of the method to a variety of
proteases, including dipeptidyl peptidase IV (DPP IV), dipeptidyl peptidase 8 (DPP8), fibroblast
activation protein alpha (FAPα), alphaꢀlytic protease (αLP), trypsin, and chymotrypsin. In
summary, the “P1’ modification” represents a simple, general, and highly adaptable method of
generating enzymatically stable peptideꢀbased therapeutics.
2
ACS Paragon Plus Environment

Page 3 of 42
Journal of Medicinal Chemistry
1
2
3
4
INTRODUCTION
5
6
7
8
9
Naturally occurring bioactive peptides offer a number of advantages as therapeutic agents over
synthetic small molecules or antibodies.1 Honed over eons by evolution to have a particular
function, usually through specific interaction with one or more targets, peptides are relatively
free of adverse events caused by crossꢀreactivity with unintended targets. Owing to their smaller
size, peptides also provide superior tissue penetration than antibodies.1 Advances in peptide
chemistry and manufacturing have fostered the clinical development and application of
therapeutic peptides. In fact, the number of new peptides entering clinical testing has nearly
doubled in the past decade.2 The emergence of peptides as mainstream therapeutic agents has
been hindered, however, by their susceptibility to proteolytic degradation and inactivation in
circulation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The serine protease dipeptidyl peptidase IV (DPP IV) preferentially cleaves peptide bonds after
Pro or Ala residues penultimate to the Nꢀterminus3ꢀ6 and functions as the principal determinant of
the circulating halfꢀlife for many bioactive peptides. Several endogenous DPP IV substrates,
including glucagonꢀlike peptideꢀ1 (7ꢀ36) amide (GLPꢀ1) (t < 2 min)3, glucoseꢀdependent
½
insulinotropic factor (GIP) (t = 5ꢀ7 min)4, oxyntomodulin (OXM) (t = 12 min)7, neuropeptide
½
½
Y (NPY) (t = 12 min)8, brain natriuretic peptide (BNP) (t = 15ꢀ20 min)9, and enterostatin
½
½
(ENT) (t = 5 min)10, have biological properties that render them therapeutically useful. For
½
example, the numerous glucoseꢀlowering properties of the incretin hormone GLPꢀ1 make it an
ideal agent for the treatment of Type 2 diabetes mellitus (T2D).11ꢀ13 Similar to GLPꢀ1, the
hormones GIP and OXM act as glucoseꢀlowering agents and have been studied extensively as
potential treatments for T2D.11, 14ꢀ20 Furthermore, OXM and ENT suppress appetite and enhance
21ꢀ23
energy expenditure and have therefore been implicated as potential treatments for obesity.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 42
1
2
3
4
5
6
7
8
9
NPY, which possesses potent anxiolytic and antiꢀdepressant effects, has been proposed as a
therapeutic agent for anxiety, depression, and postꢀtraumatic stress disorder.24ꢀ26 BNP, which
plays an important role in the body’s natural defense against hypertension and plasma volume
expansion, is currently FDAꢀapproved for the treatment of acutely decompensated congestive
heart failure.27, 28 With such short halfꢀlives, however, therapeutic use of these peptides would
require multiple injections per day or intravenous infusion. Such methods of administration are
expensive, inconvenient, and invasive; rendering them impractical for the treatment of chronic
diseases that require daily medication.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Here we report that substitution of the P1’ residue, which lies directly Cꢀterminal of the enzyme
cleavage site, with an amino acid analogue containing a tertiaryꢀsubstituted βꢀcarbon atom
renders these peptides resistant to proteolytic degradation by DPP IV without significantly
altering their biological activity. Circular dichroism and receptor activation studies indicate that
the modified peptide analogues retain the overall secondary structure and receptor agonist
activity of their native counterparts. Furthermore, experiments in diabetic (db/db) mice revealed
that treatment with a P1’ꢀmodified version of the naturallyꢀoccurring hormone GLPꢀ1 elicited a
more potent and prolonged glucoseꢀlowering affect than treatment with the native peptide. The
“P1’ modification” appears to confer resistance to essentially all serine proteases, as evidenced
by our studies with DPP IV, dipeptidyl peptidase 8 (DPP8), fibroblast activation protein alpha
(FAPα), alphaꢀlytic protease (αLP), chymotrypsin, and trypsin. The modification also confers
resistance to the metalloprotease aminopeptidase P 2 (APP2), suggesting that it may effectively
block proteolysis by this class of enzyme as well. Kinetic studies suggest that the P1’
modification prevents degradation by blocking the enzyme’s catalytic machinery rather than by
interfering with substrate recognition or binding. In summary, the P1’ modification represents a
4
ACS Paragon Plus Environment

Page 5 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
general and adaptable method of generating enzymatically stable peptides that retain the
physiochemical properties and biological activity of their native counterparts.
7
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 42
1
2
3
4
RESULTS
5
6
7
8
9
Amino Acid Analogues. Three hexapeptides with the sequence APXSWS, where X was Leu,
Ile, or tertꢀLeu (1), were designed to investigate the substrate specificity of DPP IV at the P1’
position. The amino acid analogue 1 is an isomer of the amino acids Leu and Ile that contains a
tertiaryꢀsubstituted βꢀcarbon; a stereochemical arrangement that is not present in any naturally
occurring amino acid (Figure 1). As shown in Figure 2, the synthetic hexapeptides containing
the natural amino acid residues Leu or Ile at the P1’ positions were quickly hydrolyzed by DPP
IV. At 30 min, less than 70% of AP(Leu)SWS and AP(Ile)SWS remained intact. In contrast, the
hexapeptide with the amino acid analogue 1 at the P1’ position was completely resistant to DPP
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
IVꢀmediated
cleavage
over
the
30
min
incubation.
NH2
2
2
2
1
2
3
NH2
4
5
Figure 1. Structures of the amino acid analogues tertꢀLeu (1), βꢀdimethylꢀAsp (2), βꢀdimethylꢀ
Gln (3), NꢀFmocꢀLꢀβ, βꢀdimethylꢀAsp(tꢀbutyl) (4), and NꢀFmocꢀLꢀβ, βꢀdimethylꢀGlnꢀGly(tꢀ
butyl)ꢀOH (5).
6
ACS Paragon Plus Environment

Page 7 of 42
Journal of Medicinal Chemistry
1
2
3
4
Subsequent to this finding, the amino acid
5
6
7
8
analogues βꢀdimethylꢀAsp (2) and βꢀdimethylꢀGln
(3) were specially designed in our laboratory to
function as tertiaryꢀsubstituted βꢀcarbon atom
containing analogues of the naturally occurring
amino acid residues Asp and Gln, respectively
(Figure 1). NꢀFmocꢀLꢀβ, βꢀdimethylꢀAsp(tꢀbutyl)
(4) and NꢀFmocꢀLꢀβ, βꢀdimethylꢀGlnꢀGly(tꢀbutyl)ꢀ
OH (5) were synthesized as protected forms of 2
and 3, respectively, for incorporation into protease
substrate analogues (Figure 1 and Supporting
Information).
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2.
Degradation of synthetic
substrates by DPP IV. AP(Leu)SWS (ꢀ),
AP(Ile)SWS (ꢀ), and AP(1)SWS (ꢀ)
were incubated with purified porcine DPP
IV (10 mU/mL) at 37 oC. The amount of
intact peptide was quantified by HPLCꢀ
MS and expressed as a percentage of the
corresponding value observed before
exposure to enzyme digestion (time = 0).
Protease-resistant GLP-1 Analogues. In Vitro Degradation. The P1’ Glu residue of the
endogenous DPP IV substrate GLPꢀ1 was substituted by the amino acid analogue 1 to generate a
novel analogue (6). To conserve the acidic charge of the P1’ Glu residue, while maintaining the
stereochemical arrangement of the tertiaryꢀsubstituted βꢀcarbon, a second GLPꢀ1 analogue (7)
was synthesized with the amino acid analogue 2 at the P1’ position. The nomenclature and
amino acid sequences of native and synthetic peptides examined in this work, as well as their
corresponding P1’ꢀmodified analogues, are shown in Table 1. In vitro degradation assays were
performed to determine the halfꢀlife of native GLPꢀ1 and the P1’ꢀmodified GLPꢀ1 analogues 6
and 7 in the presence of recombinant human DPP IV. As shown in Figure 3A, native GLPꢀ1 was
quickly hydrolyzed by DPP IV (t ~ 0.3 h) (Table S1). Conversely, the analogues 6 and 7
½
remained intact over a 24 h incubation with the enzyme (Figure 3A).
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 42
1
2
3
4
5
6
Table 1. Amino acid sequences of native, synthetic, and P1’ꢀmodified protease substrates.
Amino acid analogues are indicated by the numbers 1 (tertꢀLeu), 2 (βꢀdimethylꢀAsp), and 3 (βꢀ
dimethylꢀGln).
7
8
9
Peptide/Analogue
Amino Acid Sequence
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
GLPꢀ1
6
HA EGTFTSDVSSYLEGQAAKEFIAWLVKGRꢀNH2
HA 1GTFTSDVSSYLEGQAAKEFIAWLVKGRꢀNH2
HA 2GTFTSDVSSYLEGQAAKEFIAWLVKGRꢀNH2
HA 2GTF1SDDSSY1EGQAAK2F1AW1VK2RꢀNH2
YA EGTFISDYSIAMDKIHQQDFVNWLLAQKGKKNDWKHNITQꢀNH2
YA 1GTFISDYSIAMDKIHQQDFVNWLLAQKGKKNDWKHNITQꢀNH2
YA 2GTFISDYSIAMDKIHQQDFVNWLLAQKGKKNDWKHNITQꢀNH2
HS QGTFTSDYSKYLDSRRAQDFVQWLMNTKRNRNNIAꢀNH2
HS 1GTFTSDYSKYLDSRRAQDFVQWLMNTKRNRNNIAꢀNH2
HS 3GTFTSDYSKYLDSRRAQDFVQWLMNTKRNRNNIAꢀNH2
YP SKPDNPGEDAPAEDMARYYSALRHYINLITRQRYꢀNH2
YP 1KPDNPGEDAPAEDMARYYSALRHYINLITRQRYꢀNH2
YP 1KPDNPGEDAPAEDMARYYSALRHYINLLTRPRYꢀNH2
SP KMVQGSGCFGRKMDRISSSSGLGCKVLRRHꢀNH2
SP 1MVQGSGCFGRKMDRISSSSGLGCKVLRRHꢀNH2
VP DPRꢀNH2
7
8
GIP
9
10
OXM
11
12
NPY
13
14
BNP
15
ENT
16
VP 1PRꢀNH2
LLPR VA NGSSFVTVꢀOH
17
LLPRVA 1GSSFVTVꢀOH
18
LLPR 1ANGSSFVTVꢀOH
19
Secondary Structure. The secondary structure of GLPꢀ1 and its analogues was examined by
CD spectroscopy in the far UV region (185ꢀ260 nm). GLPꢀ1, 6, and 7 were dissolved in a
sodium phosphate solution containing 30% trifluoroethanol (TFE). TFE induces and stabilizes
helix and βꢀhairpin formation and mimics the membrane environment the peptide encounters just
prior to binding its receptor.29 As is characteristic of peptides and proteins with significant αꢀ
helical secondary structure the spectrum of native GLPꢀ1 displayed minima at 222 and 208 nm
8
ACS Paragon Plus Environment

Page 9 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Characterization of P1’ꢀmodified GLPꢀ1 analogues. (A) GLPꢀ1 (ꢀ), 6 (ꢀ), and 7
(ꢀ) were incubated with purified human DPP IV (5 nM) at 37 C. The data represent the
o
mean ± SEM of 3 independent experiments. (B) GLPꢀ1, 6, and 7 were dissolved in 10 mM
sodium phosphate buffer containing 30% TFE and scanned from 260 to 185 nm on a CD
spectrophotometer. The data represent the average of 4 scans. (C) The binding affinity of
GLPꢀ1 (ꢀ), 6 (ꢀ), and 7 (ꢀ) was measured by [125I]ꢀExendinꢀ4 (9ꢀ39) competition binding
experiments with COSꢀ7 cells transiently expressing the human GLPꢀ1R. The data represent
the mean ± SEM of 3 independent experiments performed in duplicate. (D) GLPꢀ1R
activation by GLPꢀ1 (ꢀ), 6 (ꢀ), and 7 (ꢀ) was determined by measuring the stimulation of
cAMP production. The data represent the mean ± range of 2 independent experiments
performed in duplicate or triplicate. (E) Fasting blood glucose (FBG) measurements were
determined prior to and at the indicated times after IP administration to db/db mice of vehicle
(PBS) (ꢁ) or 8 ꢁg/mouse of GLPꢀ1 (ꢀ) or 7 (ꢀ). The data represent the mean ± SEM
percent change in FBG, compared to basal FBG levels, for 10 animals. (F) The change in
FBG area under the curve (AUC) was calculated for each data set presented in panel E. The
data represent the mean ± SEM change in FBG AUC (0 – 480 min) for 10 animals. Statistical
significance: #, p < 0.0001 compared to vehicle; ns: nonꢀsignificant compared to vehicle, ***, p
< 0.0001 as indicated.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 42
1
2
3
4
5
6
7
8
9
and a maximum at 190 nm (Figure 3B). The CD spectra of 6 and 7 were nearly identical to that
of GLPꢀ1 (Figure 3B), suggesting introduction of a tertiaryꢀsubstituted βꢀcarbon at the P1’
position does not significantly change the secondary structure of native GLPꢀ1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Biological Activity. Competitive binding assays were performed with the radioligand [125I]ꢀExꢀ
4 (9ꢀ39) to determine the binding affinity of native GLPꢀ1, 6, and 7 at the human GLPꢀ1R. As
shown in Figure 3C, native GLPꢀ1 bound the GLPꢀ1R with high affinity to yield an IC50 value of
1.5 nM (Table S2). GLPꢀ1’s binding affinity was reduced to 36 nM when Glu3 was replaced by
the uncharged amino acid analogue 1 to generate analogue 6 (Figure 3C and Table S2). The
binding affinity was restored to 7.6 nM when the charged amino acid analogue 2 was introduced
at the P1’ position to generate analogue 7 (Figure 3C and Table S2).
The in vitro biological activity of native and P1’ꢀmodified GLPꢀ1 analogues at the human GLPꢀ
1R was determined by measuring cAMP production in response to receptor stimulation. As
shown in Figure 3D, native GLPꢀ1 and the P1’ꢀmodified GLPꢀ1 analogues 6 and 7 activated the
GLPꢀ1R in a concentrationꢀdependent manner. Introduction of the amino acid analogue 1 or 2 at
the P1’ position in GLPꢀ1 had little if any impact on the peptide’s potency at the GLPꢀ1R (Figure
3D). The EC50 values obtained from fitted sigmoidal curves were 350, 260, and 170 pM for
GLPꢀ1, 6, and 7, respectively (Table S3).
The glucoseꢀlowering efficacy of native GLPꢀ1 and the P1’ꢀmodified GLPꢀ1 analogue 7 was
investigated in the db/db mouse model of T2D. Similar to human T2 diabetics, db/db mice
display elevated fasting blood glucose (FBG) levels. The FBG levels of db/db mice were
measured at various time points following ip administration of each peptide (8 ꢂg/mouse). As
shown in Figure 3E, thirty minutes after administration of native GLPꢀ1 the average FBG level
10
ACS Paragon Plus Environment

Page 11 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of db/db mice was reduced by 30%. The glucoseꢀlowering effect of native GLPꢀ1 was, however,
shortꢀlived. Ninety minutes after administration, the FBG levels of db/db mice administered
native GLPꢀ1 returned to the levels observed in db/db mice administered vehicle (PBS) (Figure
3E). The P1’ꢀmodified GLPꢀ1 analogue 7 elicited a more profound and longꢀlived glucoseꢀ
lowering effect than native GLPꢀ1. FBG levels of db/db mice were reduced by over 50% 60 min
after administration of 7 and remained reduced by 45% for up to 90 min after peptide
administration. At 240 min after administration, FBG levels had returned to basal levels in db/db
mice receiving any of the three treatments (PBS, GLPꢀ1, or 7). The area under the FBG curve
(0ꢀ480 min) for db/db mice administered either vehicle or 8 ꢁg of peptide (GLPꢀ1 or 7) was
calculated. In comparison to treatment with vehicle, treatment with 7 significantly (p < 0.0001)
reduced the 8ꢀh FBG AUC, whereas treatment with GLPꢀ1 did not (Figure 3F). In addition, the
8ꢀh FBG AUC was significantly (p < 0.0001) less in db/db mice treated with 7 than those treated
with native GLPꢀ1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Safety. To evaluate the safety of extended treatment with P1’ꢀmodified peptides analogue 7 (1
mg/kg) was administered by ip injection to C57BL/6 mice once daily for seven days. The
bodyweight of each mouse, as well as the food consumed per cage, was measured daily. Water
consumption was measured on the third and seventh days of treatment. As shown in Figure S1,
throughout the sevenꢀday treatment, the bodyweight, food consumption, and water consumption
of mice treated with analogue 7 was not statistically (P > 0.05) different than that of vehicleꢀ
treated animals.
On Day 7, following the final dose of vehicle or analogue 7, necropsies were performed and the
liver, kidneys, pancreas, heart, and brain were collected and weighed. No gross abnormalities
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 42
1
2
3
4
5
6
were noted in any of the twelve animals in the study. In addition, the organ weights of the
vehicleꢀtreated and peptideꢀtreated animals were not statistically different (p>0.05) (Figure S1).
7
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12
ACS Paragon Plus Environment

Page 13 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Degradation of native and P1’ꢀmodified DPP IV substrates. Peptides were
o
incubated with purified human DPP IV at 37 C. (A) GIP (ꢀ), 9 (ꢀ), and 10 (ꢀ) were
incubated with 10 nM DPP IV. (B) OXM (ꢀ), 11 (ꢀ), and 12 (ꢀ) were incubated with 50
nM DPP IV. (C) NPY (ꢀ), 13 (ꢀ), and 14 (ꢀ) were incubated with 5.0 nM DPP IV. (D)
BNP (ꢀ) and 15 (ꢀ) were incubated with 0.5 nM DPP IV. (E) ENT (ꢀ) and 16 (ꢀ) were
incubated with 10 nM DPP IV. The data in panels AꢀE represent a single experiment.
Protease-resistant Analogues of Other DPP IV Substrates. In Vitro Degradation. The
endogenous DPP IV substrates GIP, OXM, NPY, BNP, and ENT were synthesized with the
amino acid analogue 1 at the P1’ position to generate the analogues 9, 11, 13, 15, and 16,
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 42
1
2
3
4
5
6
7
8
9
respectively (Table 1). Take note that the carboxylic acid normally found at the Cꢀterminus of
native GIP, OXM, BNP, and ENT was replaced by an amide in “native” and “P1’ꢀmodified”
peptides to prevent degradation by carboxypeptidases in future in vivo studies. In addition, to
demonstrate the applicability of the method to receptorꢀselective agonists, an analogue (14) of
the Y1Rꢀselective peptide [Leu31Pro34]NPY was synthesized with 1 at the P1’ position (Table 1).
A GIP analogue (10) containing the amino acid analogue 2 at the P1’ position was synthesized so
as to conserve the acidic charge of the P1’ Asp residue present in native GIP (Table 1).
Likewise, an OXM analogue (12) containing the amino acid analogue 3 at the P1’ position was
synthesized so as to conserve the basic charge of the P1’ Gln residue present in native OXM
(Table 1). In vitro degradation assays revealed that the P1’ꢀmodified analogues 11, 12, 13, 14,
15, and 16 were completely resistant to DPP IVꢀmediated degradation for 24 h or more (Figure
4). The P1’ꢀmodified GIP analogues 9 and 10 were degraded by DPP IV, but at a significantly
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
slower rate than native GIP (t ~ 3 h) (Figure 4A and Table S1). The halfꢀlives of the analogues
½
9 and 10 were estimated to be > 30 h and 10 h, respectively. The halfꢀlives of native GLPꢀ1,
GIP, OXM, NPY, BNP, ENT and their respective P1’ꢀmodified analogues in the presence of
DPP IV are provided in Table S1.
Secondary Structure. The secondary structure of native GIP, OXM, and NPY and their P1’ꢀ
modified analogues 9, 10, 11, 12, 13, and 14 was examined by CD spectroscopy in the far UV
region (185ꢀ260 nm) as described above for native GLPꢀ1, 6, and 7. Similar to the results
obtained for GLPꢀ1 and its P1’ꢀmodified analogues, the CD spectra of each native peptide and its
respective P1’ꢀmodified analogue(s) were nearly identical (Figure S2).
14
ACS Paragon Plus Environment

Page 15 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Concentrationꢀresponse curves for GIP, GLPꢀ1, GCG, Y1, and Y2 receptor
activation. Halfꢀmaximal effective concentration (EC50) values were determined by
measuring activation of a cAMPꢀdependent reporter gene (panel A), stimulation of
cAMP production (panels B and C), or recruitment of βꢀarrestin to the receptor (panels
D and E). (A) Agonist activity of GIP (ꢀ), 9 (ꢀ), and 10 (ꢀ) at the GIPR. Agonist
activity of OXM (ꢀ), 11 (ꢀ), and 12 (ꢀ) at the GLPꢀ1R (B) or GCGR (C). Agonist
activity of NPY (ꢀ), 13 (ꢀ), and 14 (ꢀ) at the Y1R (D) or Y2R (E). The data
represent the mean ± SEM of two independent experiments performed in duplicate or
triplicate.
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 42
1
2
3
4
5
6
7
8
9
Biological Activity. The in vitro biological activity of native and P1’ꢀmodified forms of GIP,
OXM, and NPY were determined at their respective receptor(s) by measuring activation of a
cAMPꢀdependent reporter gene (Figure 5A), stimulation of cAMP production (Figure 5B and
5C), or βꢀarrestin recruitment to the receptor (Figure 5D and 5E). As shown in Figure 5, all
native and P1’ꢀmodified peptides activated their respective receptors in a concentrationꢀ
dependent manner. Introduction of the amino acid analogue 1 or 2 at the P1’ position in GIP
generated the analogues 9 and 10 and decreased the peptide’s potency at the GIPR by
approximately 20ꢀ or 10ꢀfold, respectively (Figure 5A). As previously reported, native OXM
acted as a dual agonist at the GLPꢀ1 and glucagon (GCG) receptors (Figure 5B and 5C).
Introduction of the amino acid analogue 1 at the P1’ position in OXM, to generate analogue 11,
reduced the potency of the peptide at the GLPꢀ1R and CGCR by approximately 2ꢀ and 20ꢀfold,
respectively (Figure 5B and 5C). Introduction of the amino acid analogue 3 at the P1’ position in
OXM, to generate analogue 12, did not change the peptide’s potency at the GCGR, but decreased
the peptide’s potency at the GLPꢀ1R by more than 10ꢀfold (Figures 5B and 5C). The NPY
analogues 13 and 14, both of which have the amino acid analogue 1 at the P1’ position, retained
agonist activity similar to that of native NPY at the Y1R (Figure 5D). The NPY analogue 13
also potently activated the Y2R, while the Y1Rꢀselective analogue 14 did not, even at
concentrations in the high nanomolar range (Figure 5E). Corresponding EC50 values for native
and P1’ꢀmodified analogues are provided in Table S3.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Application of the P1’ Modification to Other Enzyme Substrates. DPP IV Homologs: FAPα
and DPP8. The P1’ꢀmodified NPY analogues 13 and 14 were exposed to the DPP IV homologs
FAPα and DPP8 (Figure 6). Similar to DPP IV, these enzymes cleave on the carboxyl side of
30, 31
the Pro2 residue in NPY to generate the truncated metabolite NPY3ꢀ36
.
In the presence of
16
ACS Paragon Plus Environment

Page 17 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
FAPα the halfꢀlife of native NPY was 4.8 h, whereas the analogues 13 and 14 had halfꢀlives of
greater than 50 h, with approximately 70% of
both remaining intact after 48 h (Figure 6B and
Table S1). Interestingly, less than 50% of native
NPY was digested after an 8 h exposure to DPP8
(Figure 6A). The P1’ꢀmodified analogue 13, on
the other hand, remained completely intact over
an 8 h incubation with DPP8 (Figure 6A).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Metalloprotease: APP2.
NPY is also
metabolized by the metalloprotease APP2, which
cleaves after the Nꢀterminal Tyr1 residue to
32
generate NPY2ꢀ36
.
In the presence of APP2,
native NPY had a halfꢀlife of approximately 0.3 h
and was almost completely degraded to NPY2ꢀ36
after a 1 h incubation (Figure 6C and Table S1).
Under the same conditions, the P1’ꢀmodified
analogues 13 and 14 had greatly extended halfꢀ
lives of 11.7 and 11.0 h, respectively (Figure 6C
and Table S1).
Figure 6. Degradation of NPY and P1’ꢀ
modified analogues by DPP8, FAPα, or
APP2. NPY (ꢀ), 13 (ꢀ), and 14 (ꢀ)
were incubated with purified (A) DPP8
(100 nM), (B) FAPα (25 nM), or (C)
Serine Proteases: Trypsin, Chymotrypsin, αLP.
A synthetic endoprotease substrate (17) was
designed in our laboratory to contain both a postꢀ
o
APP2 (100 nM) at 37 C. The data
represent the mean ± range of two (A and
C) or mean ± SEM of three (B)
independent experiments.
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 42
1
2
3
4
5
6
7
8
9
Ala cleavage site for the archetypal serine
protease αLP and a postꢀArg cleavage site
for trypsin (Table 1). The substrate
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
analogues 18 and 19 were generated by
replacing the P1’ residue at the cleavage
sites for αLP and trypsin, respectively, in 17
with the amino acid analogue 1 (Figure 7
and Table 1). In the presence of purified
αLP or trypsin the unmodified endoprotease
substrate 17 had a halfꢀlife of approximately
0.4 h (Figure 7 and Table S1). Conversely,
the P1’ꢀmodified analogues 18 and 19
remained completely intact when incubated
with αLP or trypsin for 5 or 8 h, respectively
(Figure 7).
Figure 7. Degradation of unmodified and
P1’ꢀmodified substrates of αLP and trypsin.
(A) 17 (ꢀ) and 18 (ꢀ) were incubated with
o
purified αLP (10 nM) at 37 C. (B) 17 (ꢀ)
To further demonstrate the widespread
applicability of this method, a GLPꢀ1
analogue (8) containing the amino acid
and 19 (ꢀ) were incubated with purified
o
trypsin (10 nM) at 37 C. Sites of αLP (A)
and trypsin (B) cleavage are shown. The
data represent a single experiment.
analogues 1 or 2 at the P1’ position of trypsin and chymotrypsin cleavage sites was also
generated. The stability of the multiplyꢀmodified GLPꢀ1 analogue 8 was investigated in the
presence of the digestive enzymes trypsin and chymotrypsin (Figure 8). Unmodified native
GLPꢀ1 was completely degraded after an incubation of only 10 s with either trypsin or
chymotrypsin (Figure 8). In contrast, 8 remained completely intact over a 5 h incubation with
18
ACS Paragon Plus Environment

Page 19 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
trypsin (Figure 8A). The rate of degradation of 8 was significantly reduced in comparison to that
of native GLPꢀ1 in the presence of chymotrypsin, with 50% of 8 remaining intact after ~ 0.6 h
(Figure 8B and Table S1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kinetic Analysis of Peptide-DPP IV
Interaction. It is possible that P1’ꢀmodified
peptides are less susceptible to degradation by
DPP IV because (i) the modified substrates lose
affinity for the enzyme, or (ii) binding to the
activeꢀsite is conserved but subsequent catalytic
cleavage is compromised. Kinetic analyses
were performed to explore these options and
determine whether and to what extent P1’ꢀ
modified substrates inhibit DPP IV activity. In
brief, the initial rate of DPP IVꢀmediated
cleavage of the chromogenic substrate AlaꢀProꢀ
paraꢀnitroanilide (APꢀpNA) was measured in
Figure 8. Degradation of GLPꢀ1 and 8 by
trypsin and chymotrypsin. GLPꢀ1 (ꢀ) and
8 (ꢀ) were incubated with purified (A)
trypsin (100 nM) or (B) chymotrypsin (100
nM) at 37 C. Predicted sites of trypsin (A)
and chymotrypsin (B) cleavage within the
the presence of increasing concentrations of the
P1’ꢀmodified analogues 9, 13, 14, 15, or 16.
Modified peptides were found to competitively
inhibit DPP IV, with Ki values of 160 ± 70 for
9, 6.3 ± 0.8 for 13, 5.0 ± 0.6 for 14, and 61 ± 7
ꢁM for 15 (Figure S3 and Table S4). These
o
sequence of 8 are shown.
The data
represent the mean ± range of two
independent experiments.
values are in the range of reported Km values of DPP IV for the corresponding unmodified
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 42
1
2
peptides (Table S4).33ꢀ35 Although to the best of our knowledge the Km of ENT with DPP IV has
not been reported, the low Ki (4.5 ± 0.7 ꢁM) value of the P1’ꢀmodified ENT analogue (16)
determined in our experiments implies that the P1’ꢀmodified analogue binds with a relatively
high affinity to the DPP IV activeꢀsite (Figure S3E). Thus, it appears that while the enzyme
affinity of modified peptides is largely conserved, their catalytic cleavage is compromised. As
expected, native peptides had markedly higher Ki values for inhibition of DPP IV than their P1’ꢀ
modified counterparts. For example, the Ki for inhibition of DPP IV by ENT was 142 ± 43 ꢁM,
while the Ki for inhibition of DPP IV by 16 was only 4.5 ± 0.7 ꢁM (Table S4). This is likely due
to the fact that the native peptides are rapidly degraded by DPP IV and the resulting metabolites
do not act as competitive inhibitors of the enzyme.
3
4
5
6
7
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
ACS Paragon Plus Environment

Page 21 of 42
Journal of Medicinal Chemistry
1
2
3
4
DISCUSSION
5
6
7
8
The use of peptides as therapeutic agents is greatly limited by their susceptibility in vivo to
degradation and inactivation by a wide range of proteases. In this study we describe a simple
and general method of generating metabolically stable peptide analogues that retain potent
biological activity. The method involves substituting the P1’ residue, which is positioned
directly Cꢀterminal of the protease cleavage site, with an amino acid analogue containing a
tertiaryꢀsubstituted βꢀcarbon (ie. 1, 2, or 3). Application of this method to the endogenous DPP
IV substrates GLPꢀ1, GIP, OXM, NPY, BNP, and ENT generated novel P1’ꢀmodified peptide
analogues that were highly resistant to DPP IVꢀmediated degradation.
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The influence of the P1’ residue on susceptibility to DPP IVꢀmediated cleavage has been
previously examined with GLPꢀ1, GIP, and OXM. Green and colleagues found that replacing
the P1’ Glu residue in GLPꢀ1 with a Pro or Lys residue greatly enhanced the halfꢀlife of GLPꢀ1
in the presence of DPP IV, while replacing the P1’ residue with a Tyr or Phe residue only
moderately enhanced the halfꢀlife.36, 37 An investigation of the stability of Glu3ꢀsubstituted GIP
analogues in the presence of DPP IV revealed GIP analogues containing an Ala, Phe, Trp, or Tyr
residue at the P1’ position were quickly degraded by DPP IV, yielding halfꢀlives similar to that
of native GIP.38 Substitution of Glu3 with a Lys residue slightly enhanced the stability of GIP in
the presence of DPP IV, but only substitution with a Pro residue completely blocked DPP IVꢀ
mediated degradation of GIP.38 Substitution of the Gln3 residue in OXM with a Glu residue, in
combination with other Nꢀterminal substitutions (ie. DꢀHis1 or Ala2), rendered the peptide
resistant to DPP IVꢀmediated degradation over a 2 h incubation.39 When the Gln3 residue in
OXM was substituted with an Asp, Glu, Leu, noroleucine (Nle), or Pro residue, only the OXM
analogue containing an Asp residue at the P1’ position displayed significant resistance to DPP
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 42
1
2
IVꢀmediated degradation.40 To our knowledge, the influence of amino acid substitutions at the
P1’ position in NPY, BNP, or ENT on DPP IVꢀmediated degradation has never been investigated
and this is the first report of a DPP IVꢀresistant NPY, BNP, or ENT analogue.
3
4
5
6
7
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CD studies revealed that introduction of the amino acid analogues 1, 2, and/or 3 at the P1’
position in GLPꢀ1, GIP, OXM, and NPY did not significantly alter the secondary structure of the
peptides, which is required to maintain high affinity and potent activity at their respective
receptors (Figure S2).41ꢀ44 The effect of P1’ residue substitutions on agonist activity has been
previously investigated for GLPꢀ1, GIP, and OXM. GLPꢀ1 analogues containing an Ala or Asp
residue at the P1’ position were 30ꢀfold less potent and considerably less active than native GLPꢀ
1.45 Replacement of the Glu3 residue in GLPꢀ1 with a Pro, Phe, or Tyr residue also reduced
binding affinity and agonist activity at the GLPꢀ1R, and an analogue containing a basic Lys
residue at the P1’ position even acted as an antagonist at the GLPꢀ1R.36, 37 Conversely, in our
study, replacement of the P1’ residue in GLPꢀ1 with the amino acid analogue 1 or 2 did not
significantly alter the peptide’s binding affinity or agonist activity at the GLPꢀ1R. In fact, the
P1’ꢀmodified analogue 7 was ~2ꢀfold more potent than native GLPꢀ1 at the GLPꢀ1R and
demonstrated more potent and longꢀlived glucoseꢀlowering activity than the native peptide in
diabetic mice. Furthermore, daily administration of analogue 7 to lean mice for seven days did
not cause any gross abnormalities or significant changes in body weight, organ (liver, pancreas,
kidneys, heart, brain) weight, food consumption, or water consumption.
Siteꢀdirected mutagenesis studies of the GIPR have suggested that the Glu3 residue in GIP
interacts directly and uniquely with residue Arg183 in the GIPR to stimulate cAMP production.46
GIP analogues containing an Ala, Trp, or Tyr residue at the P1’ position were reported to display
slightly reduced activity at the GIPR (2ꢀ to 3ꢀfold) and caused notable inhibition (12ꢀ24%) of
22
ACS Paragon Plus Environment

Page 23 of 42
Journal of Medicinal Chemistry
1
2
GIPꢀstimulated cAMP production.38 Replacement of Glu3 in GIP with a Lys, Phe, Pro, or
Hydroxyproline (Hyp) residue produced potent GIPR antagonists.38, 47 In our study, replacement
of the P1’ Glu3 residue with the amino acid analogue 1 or 2 moderately reduced the potency of
GIP at the GIPR by 20ꢀ or 10ꢀfold, respectively. We suggest the GIP analogue 10 is a slightly
more potent GIPR agonist than analogue 9 because the amino acid analogue 2 in 10 conserves
the acidic charge of the Glu3 residue present in native GIP. The reduced potency of 10 in
comparison to native GIP may be due to a difference in how the carboxylate group of Asp in 10
and Glu in native GIP is positioned to interact with residues in the GIPR. We propose that a GIP
analogue containing a βꢀdimethylꢀGlu residue at the P1’ position may be more equivalent in
potency to native GIP than either 9 or 10.
3
4
5
6
7
8
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
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OXM is a dual agonist with activity at both the GLPꢀ1R and the GCGR.48ꢀ50 Several studies
have demonstrated that the Gln3 residue in OXM is essential for GCGR recognition and
activation. In comparison to native OXM, analogues containing a Glu, Asp, Asn, Leu, or
noroleucine (Nle) residue at the P1’ position displayed greatly reduced activity at the GCGR, but
retained potent activity at the GLPꢀ1R.40 In agreement with previously published studies, we
found that a basic charge at the P1’ position in OXM is necessary for the peptide to retain
activity comparable to native OXM at the GCGR. Introduction of the uncharged amino acid
analogue 1 at the P1’ position in OXM significantly reduced the peptide’s potency at the GCGR,
but not at the GLPꢀ1R. Conversely, introduction of the basic amino acid analogue 3 at the P1’
position significantly reduced activity at the GLPꢀ1R, but not at the GCGR. Loss of GLPꢀ1R
activity with the latter peptide derivative was surprising, since native OXM, which contains a
basic Gln residue at the P1’ position, is a potent activator of the GLPꢀ1R.
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 42
1
2
3
4
5
6
7
8
9
It is wellꢀestablished that both an intact Nꢀ and Cꢀterminus are required for NPY to bind the
Y1R subtype.43, 51 In fact, the binding affinity of NPY to the Y1R is decreased by 100ꢀfold or
>1000ꢀfold, respectively, when the Nꢀterminal Tyr amino acid residue or TyrꢀPro dipeptide is
removed.43 In contrast, an intact Nꢀterminus is not required for NPY to bind and activate the
Y2R subtype.52 The P1’ꢀmodified analogue 13 retained potent agonist activity at both the Y1R
and Y2R. Our P1’ꢀmodified derivative of the Y1Rꢀselective NPY analogue, referred to as
analogue 14, acted as a potent agonist at the Y1R, but, as expected, displayed no activity at the
Y2R.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The presence of the amino acid analogue 1 at the P1’ position in NPY also rendered the
analogues 13 and 14 resistant to cleavage by the DPP IV homologs DPP8 and FAPα and the
metalloprotease APP2. Interestingly, less than 50% of native NPY was digested after an 8 h
exposure to DPP8. A recently published study suggests DPP IVꢀgenerated metabolites can bind
to a secondary substrate binding site on the enzyme, outside the S2ꢀS2’ region, to inhibit
subsequent enzyme activity.53 Perhaps the DPP8ꢀgenerated metabolite NPY3ꢀ36 binds to a
secondary binding site on DPP IV to inhibit cleavage of the remaining fullꢀlength NPY. In the
NPY analogues 13 and 14 the P2’ position of the APP2 cleavage site is occupied by the bulky
amino acid analogue 1. The fact that APP2 is reported to disfavor P2’ residues with bulky sideꢀ
chains54 may explain its inability to degrade analogues 13 and 14.
Having demonstrated that the P1’ modification protected substrates of several
amino/dipeptidases from degradation, we explored whether this method of modification could
also prevent the degradation of endoprotease substrates. We found that a synthetic peptide
substrate containing the amino acid analogue 1 at the P1’ position of the αLP and trypsin
cleavage sites was resistant to αLPꢀ and trypsinꢀmediated degradation, respectively. In addition,
24
ACS Paragon Plus Environment

Page 25 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
introduction of the amino acid analogues 1 or 2 at the P1’ position of the trypsin and
chymotrypsin cleavage sites in GLPꢀ1 greatly extended the peptide’s halfꢀlife in the presence of
these digestive enzymes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kinetic analyses were performed to investigate the mechanism by which the P1’ modification
renders peptides resistant to DPP IVꢀmediated degradation. Kinetic analyses suggested that the
P1’ꢀmodified substrates (ie. 9, 13, 14, 15, and 16) act as competitive inhibitors of DPP IV
(Figure S3). The efficacy of P1’ꢀmodified analogues as competitive inhibitors of DPP IV varied,
with corresponding Ki’s ranging from low to high micromolar values (Table S4). Due to the
presence of conformationally restrained amino acid residues (ie. 1, 2, or 3) at the P1’ position, it
is likely that modified peptides analogues bind within the activeꢀsite of the enzyme in a
constrained conformation. We propose that in this conformation the peptide has enough
flexibility to establish the interactions necessary to retain high binding affinity at the enzyme
activeꢀsite. The peptide conformation may be too restricted, however, to establish the
interactions necessary for complete hydrolysis of the scissile bond by DPP IV. In modeling
studies the hexapeptide NPY1ꢀ6 was found to be positioned similarly within the activeꢀsites of
DPP IV and FAPα.55 Therefore, it is likely that the P1’ modification confers resistance to FAPαꢀ
and DPP IVꢀmediated degradation by a similar mechanism. It may be necessary to obtain the
crystal structure of a P1’ꢀmodified analogue in the activeꢀsite of DPP IV in order to elucidate the
exact mechanism by which the P1’ modification renders peptides resistant to enzymatic
degradation.
Conclusion
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 42
1
2
3
4
5
6
7
8
9
More than 60 biologically active hormones, growth factors, chemokines, and neuropeptides
have been identified as physiological substrates of DPP IV. The results presented herein suggest
that any of these substrates could be made resistant to DPP IVꢀmediated degradation by
introduction of the P1’ modification. Moreover, the proposed method works when applied to
peptide substrates of other proteases, such as FAPα, trypsin, and chymotrypsin. It should be
possible to synthesize and substitute a tertiaryꢀsubstituted βꢀcarbon analogue for almost every
naturally occurring amino acid ꢀꢀ a general platform for enhancing the enzymatic stability of
peptideꢀbased therapeutics.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
26
ACS Paragon Plus Environment

Page 27 of 42
Journal of Medicinal Chemistry
1
2
3
4
EXPERIMENTAL SECTION
5
6
7
8
9
Peptide Synthesis. The following peptides were synthesized and purified by the Tufts
University Core Facility (TUCF): NH2ꢀAlaꢀProꢀLeuꢀSerꢀTrpꢀSer, NH2ꢀAlaꢀProꢀIleꢀSerꢀTrpꢀSer,
NH2ꢀAlaꢀProꢀTleꢀSerꢀTrpꢀSer, GLPꢀ1, OXM, NPY, ENT, 6, 7, 8, 11, 13, 14, 16, 17, 18, and 19.
Native GIP and BNP, and the modified peptides 9, 10, 12, and 15 were synthesized and purified
by CS Bio (Menlo Park, CA). The purity (> 95%) of all peptides was confirmed by HPLCꢀMS.
The modified amino acid FmocꢀLꢀtertꢀLeu was purchased from Novabiochem. NꢀFmocꢀLꢀβ, βꢀ
dimethylꢀAsp(tꢀbutyl) (4) and NꢀFmocꢀLꢀβ, βꢀdimethylꢀGlnꢀGly(tꢀbutyl)ꢀOH (5) were
synthesized inꢀhouse (Supporting Information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Vitro Degradation Assays. In vitro degradation assays were performed to determine the
stability of native and modified peptides in the presence of various enzymes. For all degradation
assays peptides (0.1 mM) were incubated at 37 °C with purified enzyme. At specified times,
samples were collected from the degradation reactions and enzymatic activity was terminated
with the addition of 10% (v/v) 0.1 M HCl.
Synthetic hexapeptides were incubated in DPP IV assay buffer (50 mM HEPES, 0.14 M NaCl,
pH 8.0) with 10 mU/mL DPP IV purified from porcine kidney (SigmaꢀAldrich, USA). Native
and P1’ꢀmofidied analogues of GLPꢀ1, GIP, OXM, NPY, BNP, and ENT were incubated in DPP
IV assay buffer with purified recombinant human DPP IV (R&D Systems) at the concentrations
indicated in the corresponding Figure. Native and P1’ꢀmodified NPY analogues were incubated
in DPP IV assay buffer with 25 nM purified recombinant human FAPα (R&D Systems) or 100
nM purified recombinant human DPP8 (BPS Biosciences). Additionally, native and P1’ꢀ
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 42
1
2
3
4
5
6
modified analogues of NPY were incubated in APP2 assay buffer (100 mM HEPES, 500 mM
MnCl2, pH 8.0) with 100 nM purified recombinant human APP2 (R&D Systems).
7
8
9
The unmodified (17) and P1’ꢀmodified synthetic substrate (18) of the enzyme αLP were
incubated in αLP assay buffer (50 mM TrisꢀHCl, 100 mM KCl, pH 8.75) with 10 nM αLP
purified from Lysobacter enzymogenes.56 Similarly, the unmodified (17) and P1’ꢀmodified
synthetic substrate (19) of the enzyme trypsin were incubated in trypsin assay buffer (50 mM
TrisꢀHCl, 20 mM CaCl2, pH 8.0) with 10 nM purified porcine pancreatic trypsin (Sigma, USA).
Native GLPꢀ1 and 8 were incubated with 100 nM purified porcine pancreatic trypsin in trypsin
assay buffer, or with 10 nM purified bovine pancreatic chymotrypsin (Sigma, USA) in
chymotrypsin assay buffer (100 mM TrisꢀHCl, 10 mM CaCl2, pH 8.0).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Samples were analyzed by reverseꢀphase HPLCꢀMS on a Thermo Finnigan LCQ Duo
instrument equipped with a Zorbax Cꢀ18 ECLYPSE column (2.2 x 50 mm, 3.5 ꢁm) and a UV
detector. With the exception of ENT digests, peptide fragments were separated by linear
gradient elution at a flow rate of 0.2 mL/min beginning at 80% solvent A/20% solvent B and
shifting to 30% solvent A/70% solvent B over 20 min, where solvent A is water (0.1% TFA) and
solvent B is acetonitrile (0.08% TFA). The eluents were monitored by UV absorption and mass
spectrometry. ENT peptide fragments were separated by linear gradient elution beginning at
100% solvent A/0% solvent B and shifting to 50% solvent A/50% solvent B. The amount of
intact peptide was quantified by integrating the corresponding area under the curve in each
extracted ion chromatogram. This value was expressed as “Percent Peptide Intact”, relative to
the corresponding value before exposure to enzyme digestion (time 0). The data were fit to a
singleꢀphase exponential equation in GraphPad Prism 5 (GraphPad Software Inc., San Diego,
CA) to determine halfꢀlife values.
28
ACS Paragon Plus Environment

Page 29 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Circular Dichroism. Peptides were dissolved to a nominal concentration of 30 ꢁM (as
determined by UV absorbance) in 10 mM sodium phosphate buffer (pH 7.0) containing 30%
TFE. Actual peptide concentrations were calculated by measuring the UV absorbance at 280 nm
and using a molar extinction coefficient of 6970 Mꢀ1cmꢀ1 for GLPꢀ1 and its analogues, 13980 Mꢀ
1cmꢀ1 for GIP and its analogues, 8250 Mꢀ1cmꢀ1 for OXM and its analogues, and 7450 Mꢀ1cmꢀ1 for
NPY and its analogues. The concentration of BNP, which lacks any aromatic amino acid
residues, was determined by BCA assay (Thermo Scientific, USA). FarꢀUV spectra were
collected on a Jasco Jꢀ810 circular dichroism spectropolarimeter (Jasco Inc., Easton, MD) at 25
°C in a 1 mm quartz cuvette. Four scans of spectral data were accumulated from 260 to 190 nm
with a response time of 2 s, and from 185 to 200 nm with a response time of 16 s. In all cases, a
bandwith of 1 nm, data pitch of 0.5 nm, and scanning speed of 10 nm/min were employed. For
data analysis, the solvent signal was subtracted and millidegree values were converted to mean
residue ellipticity (deg cm2 dmolꢀ1) using the following formula:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[Θ]MRW = (Θobs ꢃ 100 ꢃ Mr) / (c ꢃ l ꢃ NA);
where [Θ]MRW is mean residue ellipticity, Θobs is observed ellipticity in degrees, Mr is the
peptide’s molecular weight, c is the peptide concentration in mg/mL, l is the pathlength in cm,
and NA is the number of amino acids in the peptide. GraphPad Prism 5 software was used to fit
and smooth curves and to graph mean residue ellipticity as a function of wavelength.
Receptor Binding Assay. To determine the binding affinity of native and modified GLPꢀ1
analogues at the GLPꢀ1R competitive binding experiments were performed as previously
described.57 Briefly, COSꢀ7 cells transiently expressing the GLPꢀ1R were treated with peptides
together with 17 pM of the radioligand [125I]ꢀexendin(9ꢀ39). Positive control wells contained the
29
ACS Paragon Plus Environment