N. Nishizawa et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Fig. 1. Structures of the N-terminal moieties (R) of Kp analogs.
stability, the residual ratio of Kp-10 [metastin(45–54)], 1, and 16 after
incubation in Sprague-Dawley (SD) rat serum at 37 °C was analyzed via
high-performance liquid chromatography-tandem mass spectrometry
(LC/MS/MS). The serum stability assay was initiated with the addition
of 10 µL of aqueous peptide solution (10 μg/mL) to 990 μL of pre-
incubated SD rat serum. The mixed solution was incubated at 37 °C and
100-μL aliquots were sampled at 0, 5, 15, 30, 60, and 120 min. Each
aliquot was mixed with 10 μL of 0.2% aqueous solution of formic acid/
methanol (1:4, v/v) and 150 μL of acetonitrile. After centrifugation of
the mixture at 15000 rpm for 5 min, 200 μL of supernatant was mixed
with 20 μL of 0.2% aqueous solution of formic acid/methanol (1:4, v/v)
containing an internal standard and 300 μL of 0.4% aqueous solution of
acetic acid. Twenty microliters of supernatant from each sample was
collected after centrifugation at 15000 rpm for 5 min and injected on
LC/MS/MS.
Compound 16 showed good agonist activity and superior biological
stability in rat serum, but had low solubility in water (< 10 mg/mL)
and saline (< 0.1 mg/mL). To improve the physicochemical profile of
16, derivatives with an N-terminal hetero aromatic, alicyclic, or ali-
phatic group (17–27) were synthesized and subjected to RP-HPLC
analysis to estimate the hydrophobicity of peptide from their retention
times.26 After substitution with six-membered heteroaromatic ring
systems, the 2-pyridylcarbonyl group (17) showed good agonist ac-
tivity, but had a retention time similar to that of 16 on RP-HPLC, and
the 3-pyridylcarbonyl group (18) attenuated agonist activity. Finally,
19, with a 4-pyridylcarbonyl group, showed agonist activity as potent
as 16 and had a short retention time on RP-HPLC. In fact, 19 had very
good solubility in H2O (> 20 mg/mL) and saline (> 1 mg/mL).
Among peptides with
a five-membered heteroaromatic ring
(20–22), only 22 had a shorter retention time, but also had a slightly
decreased agonist activity. Compound 23, which has a phenylacetyl
group at the N-terminus, exhibited more than 10 times lower potency
compared with 16. Peptides with alicyclic and aliphatic moieties, cy-
clohexanoyl (24), propionyl (25), isobutyryl (26), and cyclopropanoyl
(27), exhibited good agonist activities. In particular, the cyclopropa-
noyl group (27) expressing σ-aromaticity had a higher activity com-
parable with the peptide containing a benzoyl group (16), though these
peptides were less hydrophilic. The results suggested that arylcarbonyl
and α-branched structures were much more beneficial for receptor ac-
tivation than hydrophobicity, aromaticity, and hydrogen bonding
ability of the N-terminal moiety.
The residual ratio of each peptide after incubation is shown in
Table 2. Metastin(45–54) degraded quickly (with a half-life of less than
5 min) and was undetectable after 15 min. Regarding the nonapeptide
analog (1), which was a metastin(45–54) derivative with several sub-
stitutions of amino acid to increase its metabolic stability, 70% of the
initial concentration remained after 2 h of incubation. Compound 16,
the pentapeptide analog with Trp residue in addition to the same sub-
stitution as 1, displayed excellent stability and 91% of intact peptide
remained after a 2-h incubation. The efficient strategy for further im-
proving metabolic stability was reducing the peptide length of the Kp
analog containing Trp54 substitution. In studies on nonapeptide ana-
logs, replacing Phe54 with Trp54 improved the mean residence time
(MRT) values in rat intravenous pharmacokinetic studies.14 The im-
provement of MRT appears to be attributed to the enhancement of
protease resistance in the circulation.
Our successful discovery of pentapeptide analogs [R-Phe-azaGly-
Leu-Arg(Me)-Trp-NH2] without N-terminal aromatic modification con-
trasts the previous findings of others,17 who showed the importance of
the N-terminal aromatic ring for KISS1R activation in their SAR study of
pentapeptide Kp analogs (R-Phe-Gly-Leu-Arg-Trp-NH2). This incon-
sistency led us to hypothesize that the azaGly and/or Arg(Me) residues
affected molecular fitting to the KISS1R.
Table 2
Residual ratio of Kp analogs after incubation in SD rat serum.
The testosterone-suppressive effect of pentapeptide analogs was
evaluated in male SD rats after a 6-day sustained subcutaneous ad-
ministration. Compound 16 showed low solubility in aqueous media,
for the in vivo study; therefore, we selected 19 because it had water
solubility suitable for a test compound. The EC50 value of 19 for rat
KISS1R (1.2 × 10−9 M) was 7-fold higher than that of the nonapeptide,
1 (1.7 × 10−10 M). In previous studies, continuous administration of 1
and leuprolide significantly reduced plasma testosterone levels at a
and leuprolide were 1.1 and 3.2 ng/mL, respectively).13 In this study,
plasma testosterone suppression by leuprolide at a higher dose of
Time (min)
Metastin(45–54)
1
16
0
100
100
98
97
95
89
70
100
98
97
99
95
91
5
1.5
15
30
60
120
< 0.5
< 0.5
< 0.5
< 0.5
a
The stability index of synthetic metastin derivatives, percent of compound
remaining after incubation at initial concentrations of 0.1 μg/mL, was de-
termined at 37 °C in IDS/SD rat serum (n = 2).
2 nmol/h was used as
a positive control (Fig. 2A). Testosterone
3
Nishizawa, Naoki
