Linking (Pyr)1apelin-13 pharmacokinetics to efficacy: Stabilization and measurement of a high clearance peptide in rodents

Article abstract of DOI:10.1016/j.ab.2018.12.022

Apelin, the endogenous ligand for the APJ receptor, has generated interest due to its beneficial effects on the cardiovascular system. Synthesized as a 77 amino acid preproprotein, apelin is post-translationally cleaved to a series of shorter peptides. Though (Pyr)¹apelin-13 represents the major circulating form in plasma, it is highly susceptible to proteolytic degradation and has an extremely short half-life, making it challenging to quantify. Literature reports of apelin levels in rodents have historically been determined with commercial ELISA kits which suffer from a lack of selectivity, recognizing a range of active and inactive isoforms of apelin peptide. (Pyr)¹apelin-13 has demonstrated beneficial hemodynamic effects in humans, and we wished to evaluate if similar effects could be measured in pre-clinical models. Despite development of a highly selective LC/MS/MS method, in rodent studies where (Pyr)¹apelin-13 was administered exogenously the peptide was not detectable until a detailed stabilization protocol was implemented during blood collection. Further, the inherent high clearance of (Pyr)¹apelin-13 required an extended release delivery system to enable chronic dosing. The ability to deliver sustained doses and stabilize (Pyr)¹apelin-13 in plasma allowed us to demonstrate for the first time the link between systemic concentration of apelin and its pharmacological effects in animal models.

Full text of DOI:10.1016/j.ab.2018.12.022

Accepted Manuscript
1
Linking (Pyr) apelin-13 pharmacokinetics to efficacy: Stabilization and measurement
of a high clearance peptide in rodents
Joelle M. Onorato, Carrie Xu, Xue-Qing Chen, Anne V. Rose, Claudia Generaux,
Kimberley Lentz, Petia Shipkova, Susan Arthur, James K. Hennan, Roy Haskell,
Michael C. Myers, R. Michael Lawrence, Heather J. Finlay, Michael Basso, Jeffrey
Bostwick, Gayani Fernando, Ricardo Garcia, Samuel Hellings, Mei-Yin Hsu, Rongan
Zhang, Lei Zhao, Peter Gargalovic
PII:
S0003-2697(18)31120-5
https://doi.org/10.1016/j.ab.2018.12.022
YABIO 13219
DOI:
Reference:
To appear in: Analytical Biochemistry
Received Date: 31 October 2018
Revised Date: 19 December 2018
Accepted Date: 28 December 2018
Please cite this article as: J.M. Onorato, C. Xu, X.-Q. Chen, A.V. Rose, C. Generaux, K. Lentz, P.
Shipkova, S. Arthur, J.K. Hennan, R. Haskell, M.C. Myers, R.M. Lawrence, H.J. Finlay, M. Basso, J.
Bostwick, G. Fernando, R. Garcia, S. Hellings, M.-Y. Hsu, R. Zhang, L. Zhao, P. Gargalovic, Linking
1
(Pyr) apelin-13 pharmacokinetics to efficacy: Stabilization and measurement of a high clearance peptide
in rodents, Analytical Biochemistry (2019), doi: https://doi.org/10.1016/j.ab.2018.12.022.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Linking (Pyr)1Apelin-13 Pharmacokinetics to Efficacy: Stabilization and Measurement of a High
Clearance Peptide in Rodents
(Subject category: mass spectrometry, peptides)
Joelle M. Onorato*, Carrie Xu, Xue-Qing Chen, Anne V. Rose, Claudia Generaux, Kimberley
Lentz, Petia Shipkova, Susan Arthur, James K. Hennan, Roy Haskell1, Michael C. Myers, R.
Michael Lawrence, Heather J. Finlay, Michael Basso, Jeffrey Bostwick, Gayani Fernando,
Ricardo Garcia, Samuel Hellings, Mei-Yin Hsu, Rongan Zhang, Lei Zhao and Peter Gargalovic
Drug Discovery
Bristol-Myers Squibb Company
Princeton, NJ 08543-5400
1Present address: Arvinas LLC, New Haven, CT 06511
Running Title: Stabilization and Determination of (Pyr)1Apelin-13 in Rodent Plasma
*Send Proofs and Correspondences to:
Joelle Onorato
Bristol-Myers Squibb Company
PO Box 5400
Princeton, NJ 08543-5400
Tel:
(609)-252-4292
E-mail: joelle.onorato@bms.com
Abbreviations: LOQ, limit of quantitation; LC/MS/MS, liquid chromatography/mass spectrometry/mass
spectrometry; TFA, trifluoroacetate; FA, formic acid; PBS, phosphate buffered saline; DMF,
dimethylformamide; Fmoc, fluorenylmethyloxycarbonyl; HCTU, 2-(6-chloro-1-H-benzotriazol-
1-yl)-1,1,3,3-tetramethyluronium; Et2O, diethyl ether; Pbf, 2,2,4,6,7-
pentamethyldihydrobenzofuran-5-sulfonyl; Trt, trityl; BOC, tert-butyloxycarbonyl; tBu, tert-
butyl; HPLC, high performance liquid chromatography; LC-MS, liquid chromatography mass
spectrometry; MeCN, acetonitrile; NMM, n-Methyl morpholine; TFA, trifluoroacetic acid; IS,
internal standard; QC, quality control; PK, pharmacokinetic; PD, pharmacodynamic; ELISA, enzyme-
linked immunosorbent assay; EIA, Enzyme Immunoassay; SC, subcutaneous; IV, intravenous; IP,
1

ACCEPTED MANUSCRIPT
intraperitoneal; PMSF, phenylmethanesulfonyl; BSA, bovine serum albumin; PC, phosphatidylcholine;
GDO, glyceryl dioleate; PI, protease inhibitor
Abstract
Apelin, the endogenous ligand for the APJ receptor, has generated interest due to its beneficial
effects on the cardiovascular system. Synthesized as a 77 amino acid preproprotein, apelin is post-
translationally cleaved to a series of shorter peptides. Though (Pyr)1apelin-13 represents the major
circulating form in plasma, it is highly susceptible to proteolytic degradation and has an extremely short
half-life, making it challenging to quantify. Literature reports of apelin levels in rodents have historically
been determined with commercial ELISA kits which suffer from a lack of selectivity, recognizing a range
of active and inactive isoforms of apelin peptide. (Pyr)1apelin-13 has demonstrated beneficial
hemodynamic effects in humans, and we wished to evaluate if similar effects could be measured in pre-
clinical models. Despite development of a highly selective LC/MS/MS method, in rodent studies where
(Pyr)1apelin-13 was administered exogenously the peptide was not detectable until a detailed stabilization
protocol was implemented during blood collection. Further, the inherent high clearance of (Pyr)1apelin-
13 required an extended release delivery system to enable chronic dosing. The ability to deliver sustained
doses and stabilize (Pyr)1apelin-13 in plasma allowed us to demonstrate for the first time the link between
systemic concentration of apelin and its pharmacological effects in animal models.
Keywords (max 6)
Apelin-13, mass spectrometry, ELISA, rodent, extended release, beagle dog, stabilization
2

ACCEPTED MANUSCRIPT
1. Introduction
The physiological role of apelin and its receptor APJ is not well understood, though pre-clinical
and clinical evidence suggests a broad range of effects including a potential important role in the
pathogenesis of heart failure and cardiovascular homeostasis [1-9]. Characterization and measurement of
circulating apelin peptides in plasma has been hindered by very low levels as well as a lack of specific
assays for their detection. Reported endogenous circulating levels of immuno-reactive apelin in humans
vary by as much as an order of magnitude [6, 10]. This same variation in basal levels of immuno-reactive
apelin is observed in rodents, where reported concentrations range from sub-pg/mL to ng/mL. For
example, the measured baseline levels of apelin in Wistar rats determined by two different labs was
reported as 45.5 ng/mL when measured by EIA and 0.16 pg/mL when measured by ELISA [11, 12].
Acknowledging that many factors, including differences in animals/environment/handling, could
contribute to variations in measurement of any endogenous molecule, the ambiguity associated with
immunoassays is well recognized. Though sensitive, they often suffer from non-selective or poorly
characterized reagents. Certainly caution should be used when interpreting results from commercially
available apelin ELISA kits, as we have found them to cross-react with a range of apelin fragments,
including shorter C-terminal fragments of apelin lacking biological activity. Investigators have also
reported non-selectivity of in-house generated anti-apelin antibodies; an example is a polyclonal antibody
generated against apelin-17 shown to cross-react with 13-, 12-, 11-, 10- and 5-amino acid apelin
fragments [13, 14]. The uncertain specificity of readily available apelin assays kits has made
interpretation of circulating apelin levels difficult, whether measured as endogenous or following
peripheral administration. Mesmin and co-workers demonstrated the disconnect between assays by
comparing the determination of immuno-reactive apelin to specific apelin isoforms determined by
LC/MS/MS in identical human plasma samples [15]. They demonstrated that immuno-reactive plasma
apelin (208-466 pg/mL) was not detectable in the same plasma samples analyzed by LC/MS/MS, despite
an LOQ of 30-150 pg/mL. This further highlights the issues associated with using non-specific
3

ACCEPTED MANUSCRIPT
immunoassays to quantify individual apelin peptides and suggests caution should be used when
comparing apelin levels reported by different investigators. The selectivity of mass spectrometry has
enabled a more confident evaluation of apelin isoforms present in various systems. Pyroglutamyl apelin-
13 ((Pyr)1apelin-13) has been identified using mass spectrometry as the major circulating form in human
plasma and the oxidized peptide as the major form in human heart [16, 17], and thus is considered to be
the most abundant “active” form of the peptide in terms of potential beneficial cardiovascular effects.
Further adding to the challenges associated with measuring apelin itself, circulating peptides in
general suffer from rapid clearance as well as a susceptibility to proteolytic hydrolysis. The rate of
hydrolysis can significantly vary between species, as described herein. In the case of apelin, little is
known about the specific proteases responsible for its cleavage. Though ACE2 has been identified as a
major enzyme in the apelin degradation pathway, it is likely that multiple proteases are involved, which
complicates attempts at stabilization through targeted inhibition [18]. Mesmin and Zhen have both
characterized the human ex-vivo proteolytic degradation patterns of apelin, and both labs were able to
sufficiently stabilize apelin in human plasma using either protease inhibitor (PI) (Mesmin) or plasma
acidification (Zhen) [10, 15, 16]. While these investigators were able to accurately quantify different
apelin peptides in human plasma, stabilization of apelin in other species has not been demonstrated to
date.
To study the effects of (Pyr)1apelin-13 in pre-clinical models, we thus faced the challenge of
stabilizing the peptide in plasma with enough confidence to allow accurate quantification. Following
development and optimization of a blood collection protocol, plasma levels of (Pyr)1apelin-13 were
successfully measured and correlated to hemodynamic response. Further, alternative delivery routes were
tested and used to administer (Pyr)1apelin-13 such that plasma concentrations could be sustained over
days or weeks. Our study highlights the importance of thoughtful and thorough assay development in the
particular matrix of interest, and is the first report to accurately quantify levels of (Pyr)1 apelin-13 in
rodent plasma and link its concentrations to pharmacodynamic response.
4

ACCEPTED MANUSCRIPT
2. Materials and methods
2.1 Chemicals and reagents
(Pyr)1apelin-13 (TFA salt) was purchased from GenScript® USA, Inc. (Piscataway, NJ).
13C/15N-(Pyr)1apelin-13 (Pro 3 and Pro 12 labelled with 13C/15N for an added mass of 12 Da) was
purchased from Innovagen (Sweden). Apelin peptides were either purchased from GenScript® or
synthesized in-house as described below. Apelin-36 (Human) Extraction-Free EIA Kit (catalog #EK-
057-15), Apelin-36 (Rat, Mouse) Extraction-Free EIA Kit (catalog #EK-057-28) and Apelin-12 (Human,
Rat, Mouse, Bovine) EIA Kit (catalog #EK-057-23) were purchased from Phoenix Pharmaceuticals, Inc.
(Burlingame, CA). Control animal plasma (K2EDTA) was purchased from BioIVT (Westbury, NY). Rat
(Sprague-Dawley) whole blood was either generated in-house or purchased from BioIVT, shipped on ice
and used on the day of delivery. Mouse (C57Bl/6), dog (beagle) and human blood was obtained in-house
and used on the day of collection. Guanidine hydrochloride, PMSF (phenylmethanesulfonyl fluoride),
cOmplete® PI cocktail, bovine serum albumin (BSA), rat serum albumin (RSA) and formic acid were
purchased from Sigma (St. Louis, MO). HALT® PI cocktail was purchased from ThermoFisher
Scientific (Waltham, MA). All liquid chromatography solvents (water, methanol, acetonitrile,
isopropanol) were LC/MS grade. Glyceryl dioleate (GDO) was purchased from Pfaltz & Bauer
(Waterbury, CT). Lipoid S100 (phosphatidylcholine from soybean) was purchased from Lipoid GmbH
(Newark, NJ).
2.2 Institutional approvals
All procedures with animals were conducted under guidelines set by the BMS Institutional
Animal Care and Use Committee. Blood samples from BMS employee volunteers were obtained in
accordance with the Corporate Policy on Environment, Health and Safety.
5

ACCEPTED MANUSCRIPT
2.3 Synthesis of apelin peptides
1
A representative example of (Pyr) apelin-13 (H-pGluRPRLSHKGPMPF-OH) peptide synthesis is
outlined below:
1
The synthesis of (Pyr) apelin-13; H-pGluRPRLSHKGPMPF-OH was conducted on 0.1 mmol
scale. All amino acids were double coupled using HCTU/NMM in DMF. A starting resin of Fmoc-Phe-
Wang resin (100-200 mesh; BACHEM), 0.76 mmol/g, 132 mg resin was employed. Amino acid stock
solutions were prepared as 0.2 M solutions in DMF. To a polypropylene solid-phase reaction vessel was
added Wang-Phe-Fmoc resin (132 mg, 0.100 mmol). The resin was washed with DMF (3 x 2.0 mL) and
the mixture was agitated for 10 minutes before the solvent was drained. Fmoc deprotections were
conducted with 20% piperidine/DMF (2 x 5 mL, 2.0 min). Couplings were run with 5 equivalents of
amino acids (2.5 mL of 0.2 M solution in DMF, 2.5 mL of 0.4 M HCTU/0.8 M NMM/DMF; 15 minutes
for each amino acid) and were run twice. The growing peptide on resin was washed between coupling
cycles with DMF (3 x 5 mL, 0.5 minutes) with a final wash of the finished peptide on resin with
dichloromethane (5 mL) followed by drying under a nitrogen atmosphere. After the completion of the
peptide synthesis, the dried resin was transferred from the Symphony reaction vial to a plastic fritted
funnel with a plastic tight closing lid. This plastic fritted funnel had a tip that could be removed after
cleavage to filter the remaining resin away from the cleavage cocktail containing the synthesized peptide.
The cleavage cocktail was prepared (9.4 mL TFA, 0.25 mL triisopropylsilane, 0.25 mL water) and stirred
for 1 minute to ensure proper mixing. The cleavage cocktail (2.5 mL) was transferred to the plastic fritted
1
funnel containing the Wang-(Pyr) apelin-13 resin. The funnel vial was capped and shaken vigorously for
4 hours at room temperature. After 4 hours the tip was removed and the solution filtered into a 20 mL
scintillation vial. The resin was washed with 100% TFA (2 x 1 mL) into the vial and the contents of the
vial were concentrated on a Genevac over a 1.5 hour time period. The resulting solid/liquid residue was
diluted with Et2O (5 mL) and spun down on the Genevac and the Et2O was removed via pipet. This Et2O
6

ACCEPTED MANUSCRIPT
washing process was repeated twice more and the resulting white solid was stored overnight in a freezer
for subsequent preparative reverse phase HPLC purification. The white solid was warmed to room
temperature and a small portion was dissolved in H2O:MeCN (3:1). LC-MS analysis on a Waters
Acquity LC-MS system indicated that the product formed was the primary component. The white solid
was then taken up in H2O:MeCN (2:1; 14 mL) and distributed between 7 vials for preparative reverse
phase HPLC purification on a Phenomenex Luna C18 5 micron 21.2 x 250 mm column using the
following gradient conditions (HPLC solvents contained 0.05% TFA): 15% MeCN to 35% MeCN over
18 minutes, ramp to 100% MeCN for 4 minutes (retention time = 12.5 minutes). The preparative
fractions were pooled, concentrated and analyzed by LC/MS to confirm the product mass of M+2H+ =
767.
Common amino acids used are listed below with side-chain protecting groups indicated inside
parenthesis:
pGlu-OH = pyroglutamic acid; Fmoc-Arg(Pbf)-OH; Fmoc-Gly-OH; Fmoc-His(Trt)-OH; Fmoc-Leu-OH;
Fmoc-Lys(Boc)-OH; Fmoc-Met-OH; Fmoc-Pro-OH; Fmoc-Ser(tBu)-OH;
2.4 Evaluation of commercially available EIA/ELISA cross-reactivity with apelin peptides
Evaluation of the apelin peptide fragment cross-reactivity of 3 commonly used ELISA kits for the
determination of human apelin-36 (#EK-057-15), rat and mouse apelin-36 (#EK-057-28) and human, rat,
mouse and bovine apelin-12 (#EK-057-23) was performed according to the manufacturer’s instructions.
1
Briefly, (Pyr) apelin-13, apelin-36, apelin-13, apelin-8 (apelin-13 minus N-terminal amino acids 1-5),
apelin-7 (apelin-13 minus N-terminal amino acids 1-6) and an apelin C-terminal truncated peptide
(apelin-13 lacking 5 C-terminal amino acids) were added to an immunoplate pre-coated with secondary
antibody. Peptides were diluted in assay buffer and tested in duplicate at concentrations ranging from 1
pM to 10 µM. To the peptide samples was added biotinylated peptide and anti-apelin primary antibody
and the plate was incubated for 2 hours at room temperature. Following incubation the well contents
7

ACCEPTED MANUSCRIPT
were removed and the wells washed 4X with 1X assay buffer. Streptavidin-horseradish peroxidase (100
µL) was added and the plate incubated for 1 hour at room temperature. The plate was washed 4X with
1X assay buffer, then 100 µL of substrate (3, 3`, 5, 5`-tetramethylbenzidine, TMB) solution was added,
and the plate was incubated for 1 hour at room temperature. The samples were quenched with 2N HCl
(100 µL) and the absorbance at 450 nm read in a microtiter plate reader.
2.5 LC/MS/MS analysis
LC/MS/MS analysis of (Pyr)1apelin-13 was performed on an API5500 mass spectrometer
interfaced to a Thermo Flux uHPLC pump and a CTC Pal autosampler. LC was performed under
gradient conditions on a Waters BEH300 2.1 x 50 mm column heated to 65°C. The mobile phase
consisted of solvent A = 0.1% formic acid in water and solvent B = 0.1% formic acid in
acetonitrile:methanol (1:1, v/v) at a flow rate of 0.6 mL/min. The gradient used was as follows: hold 5%
B for 0.2 minutes, 5-50% B in 1.8 minutes, 50-95% B in 0.5 minutes (hold 0.75 min), 95-50% B (hold
1.75 minutes)*, re-equilibrate to 5% B. MS settings (positive ionization mode) were set to curtain gas 40,
GS1/GS2 45, temperature 450°C, ionspray voltage 4500V, declustering potential 66V, entrance potential
6V and collision energy 17V. The MS/MS transitions monitored were m/z 384.2
→424.8 for
(Pyr)1apelin-13 and m/z 387.4
→
426.8 for the IS. Data were acquired using Analyst® software. *A 50%
B hold and column wash after eluting (Pyr)1apelin-13 and prior to column equilibration was necessary to
eliminate significant peptide carry-over.
2.6 (Pyr)1apelin-13 quantification
Stock solutions of (Pyr)1apelin-13 were prepared monthly in 0.1% BSA and stored at -80°C.
Dilution solutions were prepared from the stock on the day of analysis.
Quantification of (Pyr)1apelin-13 was accomplished by spiking plasma (either purchased or
generated in-house depending on the experiment) with peptide at a concentration range of 0.018-3912
nM. Quality control samples were generated by spiking plasma with (Pyr)1apelin-13 at 2.24 and 60.4 nM.
8

ACCEPTED MANUSCRIPT
Plasma from all species to be used for blanks, QC’s or standards was incubated at 37°C for at least 4
hours to ensure degradation of any endogenous (Pyr)1apelin-13. Following finalization of the
stabilization cocktail components, rodent plasma was spiked with stabilization cocktail at the same ratio
used for unknown samples then used to generate calibration curves and QC’s. Calibration standards and
QCs were prepared fresh on the day of analysis and prepared in parallel with unknown plasma samples as
described below. The concentration of (Pyr)1apelin-13 in unknown samples was determined from a
calibration curve generated by plotting the peak area ratio of (Pyr)1apelin-13/IS vs. concentration; the
acceptance criteria for analysis was an accuracy of ±15% for all standards (±20% at the LOQ) and QCs.
2.7 (Pyr)1apelin-13 ex vivo plasma stability experiments
The stability of (Pyr)1apelin-13 in rat, mouse, dog, monkey and human plasma was evaluated by
spiking 0.6 mL of plasma (at 37°C or on ice) with 30 µL of a 100 µg/mL (Pyr)1apelin-13 solution (in
0.1% BSA) to make the final concentration of peptide ~5 µg/mL. At the designated time-points 50 µL of
plasma was removed and added to a 96-well plate containing 150 µL of ice cold IS in 0.1% formic acid
(FA) in isopropanol. The plate was vortexed for 2 minutes then stored at -80°C. After the last time-point
was collected the plate was vortexed for 2 minutes then centrifuged for 10 minutes at 4000 RPM. The
supernatant was removed, transferred to a clean plate, and dried under heated nitrogen. The samples were
then reconstituted in 150 µL of 0.1% FA in 90:10 water:methanol and analyzed by LC/MS/MS as
described above.
PI were prepared as follows. For the Roche cOmplete® cocktail, a stock solution was made by
dissolving 1 tablet in 0.625 mL water with vigorous vortexing (the manufacturer’s recommended stock
solution was 1 tablet in 2 mL, but a more concentrated stock solution was prepared to minimize plasma
dilution). For the Roche cOmplete® Mini cocktail, 1 tablet was dissolved in 1 mL of either mouse or rat
plasma (the manufacturer’s recommended stock solution was 1 tablet in 1.5 mL). The effect of Roche
cOmplete® and Halt PI cocktails on the rodent plasma stability of (Pyr)1apelin-13 was tested by adding a
9

ACCEPTED MANUSCRIPT
volume of PI stock solution to plasma to make 5, 10 or 20% PI by volume; inhibitors were added prior to
spiking the peptide. After adding peptide the stability experiments were conducted as described above.
2.8 (Pyr)1apelin-13 whole blood recovery
The stability and recovery of (Pyr)1apelin-13 in rat, mouse, dog and human whole blood was
determined by spiking small volumes of concentrated peptide solution into 0.5 mL of blood (used on the
day of collection and kept on ice) at varying concentrations, typically ranging from 100-500 nM. The
blood was then gently rocked to mix and centrifuged at 4000 RPM for 5-10 minutes to separate plasma.
Blood recovery experiments were performed in triplicate. In some experiments, blood was left on ice for
extended periods then centrifuged to separate plasma. PI and additives, if tested, were added to whole
blood as described below prior to addition of (Pyr)1apelin-13. Blood recovery experiments were repeated
at least once on an independent day using a fresh collection of whole blood.
Halt® single-use cocktail was used fresh daily at 4X the manufacturer’s recommendation (0.04
µL Halt®/µL plasma). PMSF stock solutions were made fresh the day of each experiment at the
recommended concentration of 10 mM in isopropanol. Stock solutions were kept on ice and used within 2
hours of preparation. For stability experiments, PMSF was added to plasma to make the final
concentration 0.25 mM, which is in the recommended concentration range of 0.1-1 mM.
Plasma generated from the same lot of whole blood used for peptide spiking experiments was
used for analytical double blanks and for generation of calibration curves and QCs.
Calibration curve standards were prepared fresh on the day of each experiment.
2.9 Evaluation of (Pyr)1apelin-13 delivery via SC osmotic pump to mice and rats
Rats or mice were subcutaneously implanted with micro-osmotic pumps (ALZET 2004) pre-filled
with (Pyr)1apelin-13 in PBS. The concentration of (Pyr)1apelin-13 in the pump was chosen based on the
desired dose (25, 50 or 100 mg/kg/day) and a pump delivery volume of 0.25 µL/hour. Animals
10

ACCEPTED MANUSCRIPT
(n=6/group) were implanted and at designated time-points blood was collected via retro-orbital bleed into
tubes containing stabilization cocktail (see below). At the end of each study the pumps were recovered
and inspected for visual integrity. In selected studies, the remaining (Pyr)1apelin-13 solution in the pump
was recovered and the concentration determined by LC/MS/MS for dosing solution accuracy.
Blood (typically 150 µL) was collected using the following optimized collection procedure.
Microtainer® (Becton Dickinson, Franklin Lakes, NJ) collection tubes were pre-filled the morning of the
study with stability cocktail: HALT® (from a fresh vial, 0.04 µL/µL blood), 0.25 mM PMSF (final
concentration in blood, i.e. for 150 µL blood 3.8 µL of a 10 mM freshly prepared stock in isopropanol
was added) and 25% guanidine HCl v/v (i.e. for 150 µL blood 37.5 µL of a 6M stock was added). The
tubes were kept on ice until blood collection. Blood was collected directly into the Microtainer® tubes
containing stabilization cocktail and kept on ice until centrifugation for plasma separation in a refrigerated
centrifuge. Plasma was transferred to a collection plate and frozen at -80°C until analysis.
2.10 Evaluation of (pyr1)apelin-13 pharmacology in rat pressure-volume (PV) loop model
Male Sprague-Dawley rats (Harlan Laboratories) were anesthetized with 4% isoflurane and
surgical anesthesia was maintained with 1.5-2% isoflurane. Animals were catheterized with a Millar
SPR-671 (1.4F) pressure catheter advanced from a femoral artery into the abdominal aorta for
measurement of aortic blood pressure. A 2F Millar pressure-volume transducer was advanced from the
right carotid artery into the left ventricle for measurement of cardiac left ventricular pressure and
conductance. The right jugular vein was cannulated for administration of (Pyr)1apelin-13 and vehicle.
Following at least a 5 minute equilibration period, an infusion of vehicle containing 0.1% BSA in PBS at
0.1 mL/kg/min was started and data was collected for 10 minutes. At the end of the vehicle infusion, an
infusion of (Pyr)1apelin-13 (600 µg/kg/min, 0.1 mL/kg/min) was begun and maintained for 15 minutes,
followed by 15 minutes of wash-out. The plasma exposure of (Pyr)1apelin-13 was measured in a satellite
group of animals (i.e. animals receiving peptide IV but not monitored for hemodynamics). Blood samples
11

ACCEPTED MANUSCRIPT
were collected from the arterial cannula at 5, 10 and 15 minutes after the start of drug infusion and 2
and10 minutes after terminating the infusion (17 and 25 minutes) using the optimized blood collection
protocol described above. Blood was kept on ice until centrifugation for plasma separation. Plasma was
transferred to a collection plate and frozen immediately at -80°C until LC/MS/MS analysis.
Hemodynamic parameters evaluated were stroke volume, cardiac output, and maximum rate of
ventricular pressure change (dP/dt max), mean aortic blood pressure and heart rate. Raw data for each
parameter was averaged every 10 seconds for the duration of the infusion and plotted as the percent
change from baseline over 10 minutes. Detection of statistically significant differences from vehicle was
determined by one-way ANOVA (GraphPad Prism version 5.0 for Windows, GraphPad Software, San
Diego, CA) with Dunnett’s test for multiple comparisons (average % change data) or Tukey’s multiple
comparison test (baseline data). A p value of < 0.05 was considered statistically significant.
2.11 SC/IP administration of (Pyr)1apelin-13 in rats and beagle dogs
Sprague-Dawley rats were dosed IP or SC (4 mL/kg) with (pyr1)apelin-13 at 50 mg/kg in either
0.1% serum albumin in PBS or a lipid-based extended release vehicle containing a mix of GDO and
Lipoid S100 (phosphatidylcholine from soybean) (see below). At the designated time-points blood was
collected via tail-vein bleed into Microtainer® tubes containing stabilization cocktail as described above.
Plasma was obtained within 1 hour of collection by centrifugation in a refrigerated centrifuge at
approximately 1300 x g for 10 minutes. The plasma was collected in a 96-well plate and frozen at -80°C
until analysis.
Beagle dogs (Marshall Bioresources, Charles River, NY) chronically instrumented with a
Cardiovascular Total Telemetry Implant, Model T27G-9A (Konigsberg Instruments Inc., Pasadena, CA)
were administered (Pyr1)apelin-13 in either 0.1% serum albumin or lipid extended release vehicle (see
below) at 10 mg/kg by SC administration using a dosing volume at 0.25 mL/kg. At the designated time-
points 1 mL of blood was collected via the jugular vein into tubes containing K2EDTA and kept on ice
12

ACCEPTED MANUSCRIPT
until centrifugation. Plasma was obtained within 1 hour of collection by centrifugation in a refrigerated
centrifuge at approximately 1300 x g for 10 minutes. The plasma was transferred to a 96-well plate and
frozen at -80°C until analysis.
The composition of the lipid extended release vehicle was chosen based on evaluation of the
vehicle for delivery of peptides/macromolecules in other therapeutic areas within drug discovery
(unpublished). The GDO-Lipoid S100 vehicle (68:32 w/w) was prepared by combining 13.64 grams of
GDO with 6.41 grams of Lipoid S100 and stirring the mixture for ~2 days at 40°C. For rat and dog
studies, (Pyr)1apelin-13 was prepared at 40 mg/mL in 15% ethanol/85% GDO-Lipoid S100 mixture (v/w)
by dissolving (Pyr)1apelin-13 in ethanol (sonicate 15 minutes) then adding GDO-Lipid S100 vehicle and
stirring for 30 minutes. The resulting formulation was viscous but could be administered via syringe.
3. Results and Discussion
3.1 Selectivity evaluation of commercially available apelin ELISA kits
We used high resolution LC/MS to identify the major degradation products of (Pyr)1apelin-13.
Similar to the results reported by Mesmin and co-workers in human plasma (and therefore not revisited
here in detail) the major (Pyr)1apelin-13 fragments identified in mouse and rat plasma resulted from
sequential N-terminal proteolysis of the peptide. It is important to note that while apelin-36 and apelin-13
are biologically active, further N-terminal proteolysis of apelin-13 leads to significant loss of biological
activity, with apelin-11 and smaller fragments being inactive (Murza et al. and data not shown) [19]. The
most abundant peptide fragment identified in our studies was apelin-7, but many other fragments were
detected, indicating a complex proteolytic degradation pathway. The majority of circulating apelin levels
reported in the literature were determined using commercially available apelin ELISAs [7, 11, 12, 20].
Our preference for using mass spectrometry to measure apelin was based on the assumption that
13

ACCEPTED MANUSCRIPT
commercial ELISAs cross-react with multiple apelin peptides, including smaller degradation fragments of
apelin-13 lacking biological activity. To verify this, various apelin peptide fragments were synthesized
and tested using selected commercially available ELISAs reported in literature [7, 11, 12, 20]. Indeed,
the commercial assays recognized apelin peptides of varying lengths. For example, human apelin-36
ELISA showed cross-reactivity with apelin-13 and also smaller biologically inactive fragments, apelin-8
and apelin-7 (Figure 1A) but did not cross-react with a (Pyr)1apelin-C truncated peptide. Similar
observations were made for the rat and mouse apelin-36 ELISA (Figure 1B). These results point to the
seven C-terminal amino acids as the location of the antibody epitope (Figure 1C). Evaluation of apelin-
12 ELISA (#EK-057-23) demonstrated similar cross-reactivity (data not shown). Considering the
common utilization of commercial immunoassays due to their convenience and sensitivity, this lack of
selectivity could at least partially explain the wide range of values reported in the literature [10]. For
those reasons we selected the LC/MS/MS approach, as controlled systemic administration of (Pyr)1apelin-
13 and accurate measurement of systemic exposure was key to establishing robust and reliable PK/PD.
14

ACCEPTED MANUSCRIPT
B
A
#EK-057-15
#EK-057-28
C
Location of ELISA epitope
(Pyr)1Apelin-13
pGlu
Arg Pro Arg Leu Ser His Lys Gly Pro Met Pro Phe
-C
N-
-12 -11 -10 -9 -8 -7
(Pyr)1Apelin-C-terminal truncated peptide (not recognized by ELISA antibody)
pGlu
Arg Pro Arg Leu Ser His Lys
N-
Figure 1: Cross-reactivity evaluation of selected commercially available apelin ELISA kits
A) The cross-reactivity of apelin peptides with human apelin-36 (#EK-057-15) ELISA.
Apelin-13, apelin-8, apelin-7 and a (Pyr)1apelin-13 C-terminal truncated peptide were evaluated
with the human apelin-36 ELISA kit as described in Materials and Methods. Data shown is the
average±range of two replicates.
B) The cross-reactivity of apelin peptides with rat and mouse apelin-36 (#EK-057-28)
ELISA. (Pyr)1apelin-13, apelin-36 and apelin-7 were evaluated with the mouse apelin-36 ELISA
kit as described in Materials and Methods. Data shown in the graph is the average±range of 2
replicates.
C) Location of the Phoenix ELISA’s antibody epitope (top) and structure of the
(Pyr)1apelin-13 C-terminal truncated peptide (bottom).
15

ACCEPTED MANUSCRIPT
3.2 (Pyr)1apelin-13 plasma stability across different species
The stability of (Pyr)1apelin-13 in rat, mouse, dog, monkey and human plasma at 37°C was
determined by spiking the peptide into plasma and monitoring its disappearance over time (Figure 2A).
(Pyr)1apelin-13 was much less stable in rodent plasma (levels below limit of quantitation by 1 hour)
compared to dog, monkey and human plasma (>50% of initial concentration for at least 1 hour). The
effect of various PI cocktail combinations on the stability of apelin were tested. The stabilization
experiments were conducted on ice to mimic plasma collection protocols used in pre-clinical studies. One
of the most effective commercially available PI cocktails was the Roche cOmplete® tablet. In
combination with ice, cOmplete® was able to stabilize (Pyr)1apelin-13 in mouse plasma such that 90% or
greater remained at 60 minutes (Figure 2B). Similar results were observed in rat plasma (data not
shown). (The behavior of (Pyr)1apelin-13 in mouse and rat blood was very similar; thus, results from
both species are shown interchangeably throughout this report.)
A
B
110
100
90
100
80
60
40
20
0
No Inhibitor
80
20% COmplete®
10% COmplete®
5% COmplete®
10% COmplete®, Mini
Rat
Mouse
Dog
Human
70
60
Monkey
0
20
40
Time (minutes)
60
0
20
40
60 80 100 120 140
Time (minutes)
Figure 2: Plasma Stability Profile of (Pyr)1Apelin-13 Across Species and Effect of PI Cocktail on
Stability in Mouse Plasma
16

ACCEPTED MANUSCRIPT
A) (Pyr)1apelin-13 (5 µg/mL) was spiked into rat, mouse, dog, monkey and human plasma,
incubated at 37°C and the stability was monitored by LC/MS/MS as described in materials and
methods. Data represents the mean±SD of 3 replicates. B) The stability of (Pyr)1apelin-13 (5
µg/mL) in mouse plasma on ice was determined by LC/MS/MS in the presence of 5, 10 or 20% (by
volume) cOmplete® (purchased as regular or “mini” tablets) PI cocktail. The plasma was then
spiked with (Pyr)1apelin-13, kept on ice for 2 hours, and the plasma concentration determined by
LC/MS/MS at the specified time-points as described in materials and methods.
The significantly faster degradation of (Pyr)1apelin-13 in rodent blood and plasma compared to
dog, monkey and human (an interesting and new observation to us) was presumably due to species
differences in the protease profile. Likewise, that PI cocktails such as cOmplete® effectively prevented
proteolysis in rodent plasma but not in rodent blood suggests that protease activity in blood is greater than
in plasma. The potential uptake of (Pyr)1apelin-13 by red blood cells was also considered, but the
recovery of peptide from whole blood compared to lysed blood was comparable (data not shown).
Ultimately, introduction of a denaturation agent (guanidine HCl) to blood provided a faster resolution to
the stability problem vs. an exhaustive search for the “right” combination of PI’s. It is notable, however,
that a combination of both guanidine HCl and a PI cocktail was required to achieve complete and
reproducible recovery of (Pyr)1apelin-13 from rodent blood.
3.3 (Pyr)1apelin-13 whole blood recovery and optimization of stabilization additives
Despite the promising in vitro plasma stability results obtained with cOmplete® PI, follow-up in
vivo studies in rodents administered (Pyr)1apelin-13 and where collected rodent plasma was treated with
the cocktail did not show detectable peptide. It was suspected that degradation of (Pyr)1apelin-13 in
whole blood prior to plasma separation was an issue. To investigate this, recovery of (Pyr)1apelin-13
from plasma vs. whole rodent blood was compared. Mouse or rat plasma and whole blood (iced) either
untreated or pre-treated with a PI cocktail (for these experiments the cOmplete® tablets were replaced
17

ACCEPTED MANUSCRIPT
with HALT® single-use PI cocktail for the convenience of a ready-to-use solution) and/or denaturing
agent was spiked with (Pyr)1apelin-13. The plasma or blood was allowed to rest on ice for 15-90 minutes
(to cover the blood collection time span expected from various rodent model protocols) then plasma was
recovered, prepared and analyzed by LC/MS/MS. The recovery of (Pyr)1apelin-13 from mouse plasma (±
HALT® and guanidine HCl) compared to mouse blood (± HALT® and guanidine HCl) is shown in
Figure 3A. In plasma, HALT® was sufficient to allow recovery of >75% of (Pyr)1apelin-13, with no
additional benefit from the denaturing agent guanidine-HCl. In contrast, guanidine-HCl significantly
improved the recovery of (Pyr)1apelin-13 from whole blood vs. blood treated only with HALT® (Figure
3A). Based on these observations, the concentration of guanidine HCl was titrated to optimize recovery
of (Pyr)1apelin-13 from blood (Figure 3B).
B
A
*
**
*
**
5%
25%
% Guanidine HCl
25%
PI’s
10%
Figure 3: Recovery of (Pyr)1Apelin-13 from Mouse Blood and Optimization of Stabilization
Additives
A) The recovery of (Pyr)1apelin-13 from mouse plasma and blood was determined by
spiking the peptide at 153 ng/mL (100 nM) into C57Bl/6 mouse blood (fresh, iced) and mouse
plasma (isolated from the same lot of blood). Plasma and blood were either untreated or pre-
treated with PI (Halt® cocktail, 8 µL into 200 µL plasma/blood) or PI + guanidine HCl (G-HCl,
18

ACCEPTED MANUSCRIPT
25% v/v of a 6M stock). Plasma (spiked, or separated from spiked blood) was prepared and
analyzed by LC/MS/MS as described in materials and methods. Recovery (calculated as %
theoretical) is plotted as the mean±SD of 3 replicates. *Statistically different (p<0.05) from plasma
control by unpaired students t-test. ** Statistically different (p<0.05) from blood control by
unpaired students t-test. B) The optimal concentration of G-HCl for the recovery of (Pyr)1apelin-
13 from rat blood was determined by adding increasing concentrations of G-HCl (5, 10, 25% v/v or
25% G-HCl + Halt® cocktail + PMSF (0.25 mM) to blood then spiking the blood with (Pyr)1apelin-
13 at 100 ng/mL (65 nM). The blood was kept on ice for 90 minutes, then prepared and analyzed
by LC/MS/MS as described in materials and methods. Recovery (calculated as % theoretical) is
plotted as the mean±SD of 3 replicates.
Maximum recovery of (Pyr)1apelin-13 was achieved by treating iced blood with a cocktail of 25% v/v
guanidine-HCl (1.5M concentration in plasma) + Halt® PI cocktail (a mix of 6 broad-spectrum inhibitors)
+ PMSF (a serine and cysteine PI). This optimized cocktail was equally effective at stabilizing
(Pyr)1apelin-13 in both mouse and rat whole blood. Although the increased recovery of (Pyr)1apelin-13
using 25% guanidine-HCl was not statistically significant compared to lower concentrations the trend
towards better recovery compelled us to choose 25% guanidine-HCl. Further, we found that inclusion of
PMSF, a broad spectrum PI, with HALT® brought the recovery of (Pyr)1apelin-13 consistently near
100%.
After the composition of the blood collection stabilization cocktail was finalized it was tested to
ensure reproducible and concentration dependent recovery of (Pyr)1apelin-13 (Figure 4A). Additionally,
the stability of (Pyr)1apelin-13 in blood containing stabilization cocktail was monitored over time to
ensure complete coverage over the variable blood collection times associated with different rodent model
protocols (Figure 4B).
19

ACCEPTED MANUSCRIPT
A
B
400
300
200
100
0
0
15
30
45
60
75
90
Time on Ice Prior to Centrifugation
(minutes)
Figure 4: Reproducibility of (Pyr)1Apelin-13 Stabilization Cocktail
A) Mouse whole blood (iced) pre-treated with Halt® (8 µL into 200 µL plasma/blood),
PMSF (10 µL 0.5 mM stock into 200 µL plasma/blood) and guanidine-HCl (25% v/v of a 6M stock )
was spiked with (Pyr)1apelin-13 at 0.65, 1.6, 3.3, 13, 65 and 163 nM. Blood was centrifuged to
retrieve plasma, and the plasma was prepared and analyzed by LC/MS/MS as described in
materials and methods. B) The extended stability of (Pyr)1apelin-13 in mouse blood pre-treated
with stabilization cocktail was determined by spiking the peptide into blood at 326 nM (500 ng/mL).
Blood was kept on ice and time-points were removed at 0, 15, 30, 60 and 90 minutes, at which time
plasma was collected following centrifugation. The plasma was prepared and analyzed by
LC/MS/MS as described in materials and methods.
Notably, we found that pre-treatment of dog blood and plasma with PI and denaturing agents was
unnecessary and that standard blood/plasma collection procedures were sufficient provided that
blood/plasma was kept on ice and frozen as soon as possible. In terms of stability (Pyr)1apelin-13
behaved very similarly in dog and human whole blood. The recovery of (Pyr)1apelin-13 from mouse and
20

ACCEPTED MANUSCRIPT
rat blood (iced and pre-treated with stabilization cocktail) compared to iced, untreated blood from dogs
and humans is shown in Figure 5.
*
**
Rat
Mouse
Mouse +
Cocktail + Cocktail
Dog Human
Rat
Figure 5: Recovery of (Pyr)1Apelin-13 from Whole Blood Across Species
Blood from mice, rats, dogs and human volunteers (all collected fresh in-house and kept on
ice) was spiked with (Pyr)1apelin-13 at 65 nM. The blood was then iced for approximately 15
minutes and centrifuged for plasma collection. The plasma was prepared and analyzed by
LC/MS/MS as described in materials and methods. Data shown was calculated as the mean±SD of
at least 3 replicates. *Statistically different (p<0.05) from mouse blood control by students t-test
(unpaired). ** Statistically different (p<0.05) from rat blood control by students t-test (unpaired).
3.4 Measurement of (Pyr)1apelin-13 plasma levels in pressure-volume loop rats
(Pyr)1apelin-13 was evaluated in a rat pressure-volume loop model at 600 µg/kg/min as described
in materials and methods. Catherized rats received an IV infusion of vehicle (0.1% BSA in PBS) for 10
minutes followed by (Pyr)1apelin-13 for 15 minutes. In order to avoid disruption of hemodynamics,
plasma exposure was determined in satellite animals. As expected, (Pyr)1apelin-13 was cleared very
21

ACCEPTED MANUSCRIPT
rapidly (t1/2 = 3 minutes) following discontinuation of infusion (Figure 6A). Figure 6B shows the
beneficial effect on cardiac output in rats receiving (Pyr)1apelin-13 compared to vehicle.
A
B
10000
1000
100
10
15 minute infusion
(Pyr)1apelin-13
Vehicle
20
15
10
5
Rat 1
Rat 2
Rat 3
Rat 4
Rat 5
Rat 6
0
1
0
5
10
15
20
25
30
-5
10
20
30
Time (minutes)
Time (minutes)
Figure 6: Measurement of (Pyr)1Apelin-13 in Plasma of Pressure-Volume Loop Rats
A) (Pyr)1apelin-13 was administered IV to catheterized pressure-volume loop rats as
described in materials and methods. Plasma exposure was determined in satellite animals; blood
was collected into tubes containing the stabilization cocktail from the arterial cannula at 5, 10 and
15 minutes after the start of infusion and 2 minutes and 10 minutes after terminating the infusion
(17 and 25 minutes). The collected plasma was prepared and analyzed by LC/MS/MS as described
in materials and methods. B) Cardiac output was determined in rats receiving either vehicle or
(Pyr)1apelin-13 (600 µg/kg/min) as described in materials and methods; data is plotted as the
mean±SD of 6 rats. The difference in cardiac output between vehicle and (Pyr)1apelin-13 was
determined to be significant by one-way ANOVA.
3.5 SC osmotic pump delivery of (Pyr)1apelin-13 to rodents
We also wished to explore prolonged effects of (Pyr)1apelin-13 with chronic dosing and thus
investigated several options for administering the peptide over extended periods. A pilot exposure study
was conducted where (Pyr)1apelin-13 was delivered to mice via SC osmotic pump at 25 mg/kg/d, 50
22

ACCEPTED MANUSCRIPT
mg/kg/d and 100 mg/kg/d. Blood was collected at 2 weeks using the stabilization cocktail protocol and
the plasma analyzed for (Pyr)1apelin-13. An identical experiment had previously been performed where
blood was collected from mice receiving (Pyr)1apelin-13 via SC pump and the resulting plasma was
treated with cOmplete® PI cocktail but (Pyr)1apelin-13 was not detected. However, in the follow-up
study which implemented the blood stabilization cocktail, plasma (Pyr)1apelin-13 was detectable and the
measured exposure is shown in Figure 7A. Based on these results, a dose of 50 mg/kg/d was chosen for a
2 week proof-of-concept study. Levels of (Pyr)1apelin-13 were in the range expected based on the pilot
PK study (Figure 7A). Similarly, a SC osmotic pump was also used to successfully deliver (Pyr)1apelin-
13 to rats (50 mg/kg/d) over a 28 day study period (plasma exposures are shown in Figure 7B).
B
A
250
200
150
100
50
0
Figure 7: SC Osmotic Pump Delivery of (Pyr)1Apelin-13 to Rodents
A) (Pyr)1apelin-13 was administered at 50 mg/kg/d to mice for PK or proof-of-concept
(POC) evaluation (n=15 or 30) or B) rats (n=6) via SC osmotic pump as described under materials
and methods. Blood was collected into tubes containing the stabilization cocktail by retro-orbital
bleed (mice) or tail vein (rats) at designated times and the collected plasma prepared and analyzed
by LC/MS/MS as described in materials and methods.
23

ACCEPTED MANUSCRIPT
Evident in Figure 7 is the variability in (Pyr)1apelin-13 plasma exposure between animals when
delivered via SC pump. We investigated the source of this variability, looking at factors such as
reproducibility of pump delivery, stability of (Pyr)1apelin-13 in vehicle at 37°C over time and effect of
different vehicles (i.e. PBS vs. PBS with 0.1% BSA) but could not identify one obvious source. We
suspect the variation is a consequence of delivering a high clearance and proteolytically unstable peptide
into the SC space, where the protease content and activity is not well understood. Additionally, it cannot
be ruled out that differences in SC depot dissolution may contribute to systemic variability.
3.6 Evaluation of alternative delivery routes for (Pyr)1apelin-13 administration
Because of the variability associated with (Pyr)1apelin-13 SC pump delivery and the surgery
required for osmotic pump implantation, alternative delivery routes were investigated. First, the SC
(injection) and intraperitoneal (IP) PK profiles of (Pyr)1apelin-13 were evaluated. SC and IP
administration (vehicle = 0.1% rat serum albumin in PBS) to rats provided a longer duration of exposure
than IV, though the half-life was still relatively short (SC t1/2 ~13 minutes, IP t1/2 ~ 16 minutes). SC
administration resulted in higher and more prolonged exposure compared to IP (Figure 8A). Despite
improving the duration of exposure of (Pyr)1apelin-13 compared to IV, SC administration could not
provide adequate plasma exposure over 24 hours without frequent multiple dosing (which is undesirable
for hemodynamic measurements).
Recently, lipid formulations that generate lyotropic liquid crystalline depots in situ have
demonstrated sustained release delivery of a diverse variety of bioactive molecules [21-26] . The
composition of liquid crystal systems varies and can involve a combination of lipid and co-solvent, such
as a mixture of GDO and phosphatidylcholine (PC) in ethanol [26, 27]. The GDO-PC system has been
extensively studied and has been demonstrated to take up water immediately following administration and
form stiff gels at the injection site [26, 27]. Diffusion is extremely limited within the highly-structured
interior of the gel such that the peptide is retained within the corresponding depot [25, 28]. As the gel
24

ACCEPTED MANUSCRIPT
components are biodegradable via various lipid metabolic pathways, release occurs as structural integrity
of the liquid crystals is lost and the matrix erodes, thus producing an extended plasma profile devoid of a
burst release at early time-points [21, 25, 28, 29]. Application of these lipid systems is not limited to
hydrophobic molecules, which generally dissolve well into the injected solution, but have also
demonstrated sustained release properties for water soluble molecules, including peptides, as well [21, 25,
28, 29]. Such parenteral depots should also serve to sequester compounds from biological processes that
would otherwise lead to degradation and loss of activity at the site of injection before systemic exposure
has been achieved. Additionally, apelin peptides have demonstrated binding affinity to anionic micelles
[30-32]. This raises the possibility that release of lipid-bound apelin into the blood stream could infer a
stabilizing effect. Irrespective of mechanism, there seemed sufficient evidence to warrant trial of a SC
injection of (Pyr)1apelin-13 in such a GDO-PC lipid vehicle (GDO-Lipoid S100). Figure 8B shows the
plasma profile of (Pyr)1apelin-13 following SC injection of the extended release formulation to rats.
Using the lipid formulation the half-life of (Pyr)1apelin-13 in rats was increased to 11 hours. Notably,
while half-life was improved the measured plasma exposure of (Pyr)1apelin-13 between animals was still
variable (consistent with SC delivery contributing to variability). Despite the persistent variability in
(Pyr)1apelin-13, the lipid formulation provided an attractive surgery-free and safe (no irritation was
observed at the site of injection) alternative delivery option for rat models.
B
A
10000
1000
100
10
SC
IP
1
0
20
40
60
80
100
Time (minutes)
25

ACCEPTED MANUSCRIPT
Figure 8: Evaluation of (Pyr)1Apelin-13 Alternate Delivery Routes for Rat Models
A) (Pyr)1apelin-13 (50 mg/kg) in PBS was administered SC or IP to rats. Blood was
collected into tubes containing stabilization cocktail via tail vein bleed and the collected plasma
prepared and analyzed by LC/MS/MS as described in materials and methods. Data shown is the
mean±SD (n=3). B) (Pyr)1apelin-13 (50 mg/kg) in an extended release vehicle (GDO-Lipoid S100)
was administered SC to rats. The vehicle was prepared as described in materials and methods.
Blood was collected into tubes containing stabilization cocktail via tail vein bleed and the collected
plasma prepared and analyzed by LC/MS/MS as described in materials and methods. Data shown
is the mean±SD (n=5).
The extended release vehicle proved useful in other species as well. Despite the improved
stability profile of (Pyr)1apelin-13 in dogs vs. rodents, high clearance in dogs prevented achieving
prolonged plasma coverage through traditional dosing routes. Therefore, the extended release vehicle
was tested for SC delivery of (Pyr)1apelin-13 to dogs. As observed in rats, there were no signs of
irritation at the site of injection. The resulting plasma exposure compared to traditional SC is shown in
Figure 9. The extended release vehicle improved the half-life of (Pyr)1apelin-13 in dogs from 3.6 to 11
hours and enabled evaluation of (Pyr)1apelin-13 in dog models where repeat dosing throughout the day
can disturb hemodynamic measurements and therefore confound data interpretation (such as in telemetry-
instrumented dogs).
26

ACCEPTED MANUSCRIPT
Figure 9: Evaluation of (Pyr)1Apelin-13 SC Extended Release Vehicle in Dogs
(Pyr)1apelin-13 (10 mg/kg) in 0.1% BSA in PBS or in an extended release vehicle (GDO-
Lipoid S100) was administered SC to beagle dogs. Blood was collected via the jugular vein and
kept on ice until centrifugation. Data shown is the mean from either 5 animals (SC) or 2 animals
(SC extended release). For all studies, collected plasma was prepared and analyzed by LC/MS/MS
as described in materials and methods.
4. Conclusions
Cardio-protective effects of peripherally administered apelin in rodents have been reported by
many researchers [6, 7, 12, 20, 33-37]. The link between reported efficacy and plasma exposure, however,
remains unclear as apelin plasma levels were either not determined or measured with non-specific
immunoassays. Development of a successful blood treatment protocol for the stabilization of
(Pyr)1apelin-13 was instrumental in allowing us to establish the relationship between circulating levels of
(Pyr)1apelin-13 and PD response in acute rodent models. SC injection in GDO-Lipoid S100 vehicle
27

ACCEPTED MANUSCRIPT
allowed administration of (Pyr)1apelin-13 to rats and telemetry instrumented dogs with minimal dosing
disruption, which was critical as hemodynamic measurements in animals can be compromised or masked
by excessive handling and stress. This permitted us to monitor by telemetry, with confidence, the safety
profile of (Pyr)1apelin-13 at sustained plasma concentrations. The ability to administer (Pyr)1apelin-13
both acutely and chronically and to measure circulating levels with confidence led to an increased
understanding of the PK/PD relationship and safety profile in multiple animal models, thus providing
valuable insights into the mechanism of action of (Pyr)1apelin-13.
5. References
1.
2.
3.
4.
5.
6.
7.
Wysocka, M.B., K. Pietraszek-Gremplewicz, and D. Nowak, The Role of Apelin in Cardiovascular
Diseases, Obesity and Cancer. Front Physiol, 2018. 9: p. 557.
Mughal, A. and S.T. O'Rourke, Vascular effects of apelin: Mechanisms and therapeutic potential.
Pharmacol Ther, 2018. 190: p. 139-147.
Luo, X., et al., Apelin/APJ system: A critical regulator of vascular smooth muscle cell. J Cell
Physiol, 2018. 233(7): p. 5180-5188.
Wu, L., L. Chen, and L. Li, Apelin/APJ system: A novel promising therapy target for pathological
angiogenesis. Clin Chim Acta, 2017. 466: p. 78-84.
Lu, L., et al., Apelin/APJ system: A bifunctional target for cardiac hypertrophy. Int J Cardiol, 2017.
230: p. 164-170.
Charles, C.J., Update on apelin peptides as putative targets for cardiovascular drug discovery.
Expert Opin Drug Discov, 2011. 6(6): p. 633-44.
Falcao-Pires, I., et al., Correlation between plasma levels of apelin and myocardial hypertrophy in
rats and humans: possible target for treatment? Expert Opin Ther Targets, 2010. 14(3): p. 231-
41.
8.
9.
Barnes, G., A.G. Japp, and D.E. Newby, Translational promise of the apelin--APJ system. Heart,
2010. 96(13): p. 1011-6.
Chapman, N.A., D.J. Dupre, and J.K. Rainey, The apelin receptor: physiology, pathology, cell
signalling, and ligand modulation of a peptide-activated class A GPCR. Biochem Cell Biol, 2014.
92(6): p. 431-40.
10.
11.
12.
Mesmin, C., et al., MS-based approaches to unravel the molecular complexity of proprotein-
derived biomarkers and support their quantification: the examples of B-type natriuretic peptide
and apelin peptides. Bioanalysis, 2012. 4(23): p. 2851-63.
Bulbul, M., et al., Peripheral apelin-13 administration inhibits gastrointestinal motor functions in
rats: The role of cholecystokinin through CCK1 receptor-mediated pathway. Neuropeptides,
2017. 63: p. 91-97.
Akcilar, R., et al., The effects of apelin treatment on a rat model of type 2 diabetes. Adv Med Sci,
2015. 60(1): p. 94-100.
28

ACCEPTED MANUSCRIPT
13.
De Mota, N., et al., Apelin, a potent diuretic neuropeptide counteracting vasopressin actions
through inhibition of vasopressin neuron activity and vasopressin release. Proc Natl Acad Sci U S
A, 2004. 101(28): p. 10464-9.
14.
15.
Reaux, A., et al., Physiological role of a novel neuropeptide, apelin, and its receptor in the rat
brain. J Neurochem, 2001. 77(4): p. 1085-96.
Mesmin, C., et al., Liquid chromatography/tandem mass spectrometry assay for the absolute
quantification of the expected circulating apelin peptides in human plasma. Rapid Commun
Mass Spectrom, 2010. 24(19): p. 2875-84.
16.
17.
Zhen, E.Y., R.E. Higgs, and J.A. Gutierrez, Pyroglutamyl apelin-13 identified as the major apelin
isoform in human plasma. Anal Biochem, 2013. 442(1): p. 1-9.
Maguire, J.J., et al., [Pyr1]apelin-13 identified as the predominant apelin isoform in the human
heart: vasoactive mechanisms and inotropic action in disease. Hypertension, 2009. 54(3): p. 598-
604.
18.
Wang, W., et al., Angiotensin-Converting Enzyme 2 Metabolizes and Partially Inactivates Pyr-
Apelin-13 and Apelin-17: Physiological Effects in the Cardiovascular System. Hypertension, 2016.
68(2): p. 365-77.
19.
20.
21.
22.
23.
24.
25.
26.
Murza, A., et al., Stability and degradation patterns of chemically modified analogs of apelin-13
in plasma and cerebrospinal fluid. Biopolymers, 2014. 102(4): p. 297-303.
Pan, C.S., et al., Apelin antagonizes myocardial impairment in sepsis. J Card Fail, 2010. 16(7): p.
609-17.
Otte, A., et al., Liquid crystalline drug delivery vehicles for oral and IV/subcutaneous
administration of poorly soluble (and soluble) drugs. Int J Pharm, 2018. 539(1-2): p. 175-183.
Ki, M.H., et al., A new injectable liquid crystal system for one month delivery of leuprolide. J
Control Release, 2014. 185: p. 62-70.
Guo, C., et al., Lyotropic liquid crystal systems in drug delivery. Drug Discov Today, 2010. 15(23-
24): p. 1032-40.
Stevenson, C.L., D.B. Bennett, and D. Lechuga-Ballesteros, Pharmaceutical liquid crystals: the
relevance of partially ordered systems. J Pharm Sci, 2005. 94(9): p. 1861-80.
Clogston, J. and M. Caffrey, Controlling release from the lipidic cubic phase. Amino acids,
peptides, proteins and nucleic acids. J Control Release, 2005. 107(1): p. 97-111.
Oradd, G., et al., Phase diagram of soybean phosphatidylcholine-diacylglycerol-water studied by
x-ray diffraction and 31P- and pulsed field gradient 1H-NMR: evidence for reversed micelles in
the cubic phase. Biophys J, 1995. 68(5): p. 1856-63.
27.
Yaghmur, A., Larsen, S.W., Schmitt, M., Ostergaard, J., Larsen, C., Jensen, H., Urtti, A., Rappolt,
M., In situe characterization of lipidic bupivacaine-loaded formulations. Soft Matter, 2011. 7(18):
p. 8291-8295.
28.
29.
Clogston, J., et al., Controlling release from the lipidic cubic phase by selective alkylation. J
Control Release, 2005. 102(2): p. 441-61.
Wadsater, M., et al., Formation of highly structured cubic micellar lipid nanoparticles of soy
phosphatidylcholine and glycerol dioleate and their degradation by triacylglycerol lipase. ACS
Appl Mater Interfaces, 2014. 6(10): p. 7063-9.
30.
31.
32.
Shin, K., et al., Apelin conformational and binding equilibria upon micelle interaction primarily
depend on membrane-mimetic headgroup. Sci Rep, 2017. 7(1): p. 15433.
Langelaan, D.N. and J.K. Rainey, Headgroup-dependent membrane catalysis of apelin-receptor
interactions is likely. J Phys Chem B, 2009. 113(30): p. 10465-71.
Langelaan, D.N., et al., Structural insight into G-protein coupled receptor binding by apelin.
Biochemistry, 2009. 48(3): p. 537-48.
29