NOVEL COMPOUND FOR TREATMENT OF SEVERE HYPOGLYCEMIA
The present invention relates to a compound with improved solubility and
physical and chemical stabilities over human glucagon for use in treating diabetes and/or
5 obesity.
Human glucagon, which has the following amino acid sequence: His-Ser-GlnGly-
Thr-Phe-Thr-Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-Ser-Arg-Arg-Ala-Gln-Asp-PheVal-
Gln-Trp-Leu-Met-Asn-Thr (SEQ ID NO: 1), is a 29 amino acid peptide hormone
produced in the pancreas. When blood glucose begins to fall, glucagon signals the liver
10 to break down stored glycogen into glucose for release into the bloodstream, causing
blood glucose level to rise.
In a subject with diabetes, hypoglycemia can arise as a side effect of diabetes
treatment. In addition, the natural glucagon response to hypoglycemia in diabetics may
be impaired, making it harder for glucose levels to return to the normal range. Ifleft
15 untreated, severe or acute hypoglycemia can cause serious issues such as seizures,
unconsciousness, brain damage, or even death.
Administration of glucagon is an established therapy for treating acute
hypoglycemia. Emergency glucagon administration can restore normal glucose levels
within minutes of administration. Glucagon prepared for administration, however, has
20 several problems. In aqueous buffers at or near physiological pH, glucagon has poor
solubility. When formulated at low or high pH, glucagon also demonstrates poor
chemical stability and poor physical stability such as gelation and soluble aggregate
formation. To minimize these problems, current commercial glucagon products are
provided as a lyophilized powder with instructions to reconstitute at the time of
25 administration. In an emergency situation, reconstituting a lyophilized powder is
burdensome and inconvenient. Thus, it is desirable to provide a compound for
therapeutic use that maintains the biological performance of human glucagon under
physiological conditions while also exhibiting sufficient aqueous solubility, chemical
stability and physical stability under non-physiological conditions.
30 Glucagon analogs with amino acid substitutions to improve solubility and stability
in acidic and physiological pH buffers are disclosed in W02008086086. There is still a
need for a compound that maintains the biological performance of human glucagon under
wo 2015/094875 PCT /US2014/069643
-2-
physiological conditions while also exhibiting sufficient solubility and chemical and
physical stabilities under non-physiological conditions.
Accordingly, the present invention provides a compound which maintains wildtype
glucagon activity but also exhibit sufficient solubility as well as chemical and
5 physical stability. The present invention also provides a compound which is suitable for
pump and/or emergency administration. In addition, the present invention provides a
compound that can be administered in combination with a fast-acting insulin analog in a
dual-chamber pump to provide closed-loop glycemic control.
10
The present invention provides a compound comprising the amino acid sequence
His-Ala-Gln-Gly-Thr-Phe-Leu-Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-(Aib )-Lys-
Lys-Ala-Gln-Glu-Phe-Val-Glu-Trp-Leu-Leu-Lys-Thr-Gly-Pro-Ser-Ser-Gly-Ala-Pro-ProLys-
Ser-Lys-NH2 (SEQ ID NO: 2). The present invention also provides a compound
consisting of the amino acid sequence
His-Ala-Gln-Gly-Thr-Phe-Leu-Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-(Aib )-Lys-
15 Lys-Ala-Gln-Glu-Phe-Val-Glu-Trp-Leu-Leu-Lys-Thr-Gly-Pro-Ser-Ser-Gly-Ala-Pro-Pro-
Lys-Ser-Lys-NH2 (SEQ ID NO: 2). Unexpectedly, it has been found that the compound
of the present invention exhibits increased aqueous solubility, increased chemical
stability, and reduced fibrillation as compared to human glucagon in aqueous solution. In
addition, the compound of the present invention demonstrates enhanced solubility at pH
20 in the 5-7 range. The compound of the present invention also provides similar activity as
human glucagon- e.g., potency, time of action, and selectivity at the glucagon receptor as
compared to human glucagon. Thus, the compound of the present invention is suitable to
treat hypoglycemia, including severe or acute hypoglycemia. The improved properties of
the compound of the present invention also allow for the preparation of glucagon in
25 aqueous solutions for pump administration and severe hypoglycemia treatment. .
The present invention further provides a method of treating hypoglycemia in a
subject comprising administering a compound comprising the amino acid sequence of
SEQ ID NO: 2. The present invention also provides a method of treating hypoglycemia in
a subject comprising administering a compound consisting of the amino acid sequence of
30 SEQ ID N0:2. The present invention a further provides a compound comprising the
amino acid sequence of SEQ ID NO: 2 for use in therapy. The present invention also
provides a compound consisting of the amino acid sequence of SEQ ID NO: 2 for use in
wo 2015/094875 PCT /US2014/069643
-3-
therapy. The present invention also provides a compound comprising the amino acid
sequence of SEQ ID NO: 2 for use in the treatment of hypoglycemia. The present
invention also provides a compound consisting of the amino acid sequence of SEQ ID
NO: 2 for use in the treatment of hypoglycemia. The present invention provides a
5 compound comprising the amino acid sequence of SEQ ID NO: 2 for use in the
manufacture of a medicament for the treatment of hypoglycemia. The present invention
also provides a compound consisting of the amino acid sequence of SEQ ID NO: 2 for
use in the manufacture of a medicament for the treatment of hypoglycemia.
The present invention provides a pharmaceutical composition comprising a
10 compound comprising an amino acid sequence of SEQ ID NO: 2 and a pharmaceutically
acceptable buffer. The present invention also provides a pharmaceutical composition
comprising a compound consisting ofthe amino acid sequence of SEQ ID NO: 2 and a
pharmaceutically acceptable buffer. The present invention also provides a
pharmaceutical composition comprising a compound comprising an amino acid sequence
15 of SEQ ID NO: 2 and a histidine buffer. The present invention also provides a
pharmaceutical composition comprising a compound consisting of an amino acid
sequence of SEQ ID NO: 2 and a histidine buffer. The present invention also provides a
pharmaceutical composition comprising a compound comprising the amino acid sequence
of SEQ ID NO: 2 and histidine. The present invention also provides a pharmaceutical
20 composition comprising a compound consisting of the amino acid sequence of SEQ ID
NO: 2 and histidine. The present invention also provides a pharmaceutical composition
comprising a compound comprising an amino acid sequence of SEQ ID NO: 2 and
histidine-buffered saline. The present invention also provides a pharmaceutical
composition comprising a compound consisting of the amino acid sequence of SEQ ID
25 NO: 2 and histidine-buffered saline. The pharmaceutical composition is preferably an
aqueous solution. As used herein, the term "pharmaceutically acceptable buffer" is
understood to encompass any of the standard pharmaceutical buffers known to those
skilled in the art. Pharmaceutically acceptable buffer for parenteral administration
include, for example, physiological saline, phosphate-buffered saline, citrate-buffered
30 saline, and histidine-buffered saline. Standard pharmaceutical formulation techniques
may be employed such as those described in Remington's Pharmaceutical Sciences, Mack
Publishing Company, Easton, P A.
wo 2015/094875 PCT /US2014/069643
-4-
The compound of the present invention can be administered using any standard
route of administration, such as parenterally, intravenously, subcutaneously,
intramuscularly, or transdermally. In an embodiment, the compound of the present
invention is administered subcutaneously or intramuscularly.
5 The pharmaceutical composition can have a pH that is physiologically acceptable.
In an embodiment, the pharmaceutical composition can have a pH ranging from about 4
to about 8. More preferably, the pharmaceutical composition can have a pH of about 5 to
about 6.
A dose for the compound of the present invention can range from about 0.01 mg
10 to about 100 mg. The dose can range from about 0.01 mg to about 10 mg. The dose can
also range from about 0.1 mg to about 3 mg. In addition, the dose can range from about
0.01 mg to about 0.03mg.
The compound of the present invention can be provided as part of a kit. In an
embodiment, the kit is provided with a device for administering the compound to a
15 human subject. More preferably, the kit comprises a syringe and needle for administering
the compound. Most preferably, the compound is pre-formulated in aqueous solution
within the syringe.
The compound of the present invention can also be used in a pump system, such
as an insulin pump or a hi-hormonal (e.g., insulin-glucagon) pump system.
20 As used herein, the term "effective amount" is understood to mean an amount that
produces a desired therapeutic effect without causing unacceptable side effects when
administered to a subject. For example, an "effective amount" of a compound ofthe
present invention is the quantity that would result in greater control of blood glucose
concentration than in the absence oftreatment. An "effective amount" of a compound of
25 the present invention administered to a subject may depend on the type and severity of
the disease and on the characteristics of the subject including, without limitation, general
health, age, sex, body weight, tolerance to drugs, and the severity of inability to regulate
blood glucose.
As used herein, the term ''treating" is understood to mean amelioration of the
30 symptoms associated with a specific disorder or condition, such as hypoglycemia.
wo 2015/094875 PCT /US2014/069643
-5-
The amino acid sequences of the present invention contain the standard single
letter or three letter codes for the twenty naturally occurring amino acids. Additionally,
"Aib" is alpha amino isobutyric acid.
As used herein, "fibrillation" refers to gelation and soluble aggregate formation
5 observed when glucagon is formulated at low or high pH.
Example 1: Peptide Synthesis
The compound of SEQ ID NO: 2 is generated by solid-phase peptide synthesis on
a Protein Technologies Inc. Symphony. Synthesis (0.125 mmol scale) is performed on
10 Fmoc-Rink amide polystyrene resin (Rapp Polymere Tubingen, Germany) with
substitution approximately0.68 mmol/g. The synthesis is performed using the Fmoc
main-chain protecting group strategy. Amino acid side-chain derivatives used are:
Asp(O-tert-butyl, OtBu), Gln(Trityl, Trt), Glu(OtBu), His(Trt), Lys(tert-butoxy-carbonyl,
Boc), Ser(OtBu), Thr(OtBu), Trp(Boc), and Tyr(OtBu). Coupling is carried out with
15 approximately 10 equivalents of amino acid activated with diisopropylcarbodiimide
(DIC) and hydroxybenzotriazole (HOBt) (1: 1:1 molar ratio) in dimethylformamide
(DMF). Coupling is carried out for 90 minutes to 4 hours at room temperature.
Concomitant cleavage from the resin and side chain protecting group removal is
carried out in a solution containing trifluoroacetic acid (TFA): triisopropylsilane: 1,2-
20 ethanedithiol : water: thioanisole 90:4:2:2:2 (v/v) for 2 hat room temperature. The
solution is filtered and peptide is precipitated with cold diethyl ether and centrifuged at
4000 rpm for 3 min (cold ether washing repeated for three times). Crude peptide is
redissolved in 40 mL of water containing 10% acetic acid and purified on a C18 reversedphase
high performance liquid chromatography (HPLC) column (Waters SymmetryPrep
25 7 Jlm, 19 x 300 mm) at a flow rate of 18 mL/min. Sample is eluted with a linear AB
gradient of 15 to 55% B over 100 minutes where A= 0.05% TFA/ H20 and B = 0.04%
TF A/acetonitrile. Product generally elutes at about 26-28 %acetonitrile. Peptide purity
and molecular weight is confirmed on an Agilent 1100 Series liquid chromatographymass
spectrometry (LC-MS) system with a single quadrupole MS detector. Analytical
30 HPLC separation is done on a Waters SymmetryShield RP18, 3.5 Jlm, 4.6 mm x 100 mm
column with a linear AB gradient of 10 to 100% B over 15 minutes in which A= 0.05%
TFA/H20 and B = 0.04% TFA/ 40% H20/ 60% acetonitrile and the flow rate is 0.7
wo 2015/094875 PCT /US2014/069643
-6-
mL/min (wavelength of 220 11m). The compound is purified to > 95 %purity and is
confirmed to have molecular weight corresponding to the calculated value within 1
atomic mass unit (amu).
The TFA salt is converted to the acetate salt using AG 1-X8 Resin (Bio-RAD,
5 acetate form, 100-200 mesh, 3.2 meq/dry g, moisture content 39-48% by wt) (anion
exchange resin). For example, 422 mg peptide is dissolved in 120 mL of30%
Acetonitrile/H20. 40 g resin (about 100 fold molar ratio to positive charges of the
peptide) is added. The mixture solution is mixed by rotary stirring at room temperature
for 1 hour. The mixture solution is filtered, and the resin is washed 5 times with 30%
10 ACN/H20. The original solution and washed solution are combined and lyophilized.
Solubility and Chemical stability
The compound of SEQ ID NO: 2 is dissolved in H20 to a 10 mg/mL concentration
(peptide content), filtered through a 0.22 Jlm filter (Millex, SLGV004SL), and then
15 diluted to 1 mg/mL in Buffer P5 (10 mM Histidine, 150 mM NaCl in H20, pH 5.0) or
Buffer P6 (10 mM Histidine, 150 mM NaCl in H20, pH 6.0). Each solution is
transferred to three vials and autoclaved. Samples are then maintained at 4°C, 30°C and
40°C. Samples are visually assessed at different time points for turbidity and phase
separation. Stability of the compound is assessed by analytical reverse phase HPLC (RP-
20 HPLC) on a Phenomenex Aeris Widepore, 3.6Jlm, XB-C18 4.6 x 100 mm column
(P/NO OOD-4482-EO) heated at 60°C with a AB (A=0.05% TFA/H20; B= 0.04%
TF A/acetonitrile ) gradient of 5 % B isocratic over 5 min, 5 to 25 % B over 20 minutes,
25 to 30 % B over 30 min, and 30 to 45 % B over 10 min with a flow rate of 1.2 mL/min
(wavelength of220 11m).
25 The compound of SEQ ID NO: 2 maintains good solubility at 4°C, 30°C and 40°C
30
at pH 5 (Buffer P5) and pH 6 (Buffer P6) over 4 weeks both by visual assessment and by
RP-HPLC. Physical appearance is clear to colorless, with no opalescences and no
particles. Recovery by RP-HPLC is quantitative as demonstrated in Table 1.
wo 2015/094875 PCT /US2014/069643
-7-
TABLE 1
Buffer Total Peak 4°C Total 30°C 40°C 30°C vs 40°C vs
Area day 0 Peak Area Total Total 4°C 4°C
4wk Peak Area Peak Area Recovery Recovery
4wk 4wk 4wk 4wk
P5 4903 4933 4932 4932 100% 100%
P6 4885 4918 4911 4960 100% 101%
The compound of SEQ ID NO: 2 also maintains chemical stability at pH 5 (Buffer
P5) and pH 6 (Buffer P6) when maintained at 4°C, 30°C and 40°C for 4 weeks. As
5 demonstrated in Table 2, assessment of the samples by RP-HPLC indicates main peak
changes ofless than 1% for both, at pH 5 (Buffer P5) and pH 6 (Buffer P6) after 4 weeks
at 30° vs. 4 °C, less than 6% at pH 5 (Buffer P5) after 4 weeks at 40°C vs. 4 °C, and less
than 4% at pH 6 (Buffer P6) after 4 weeks at 40°C vs. 4 °C.
10 TABLE2
Buffer %Main 4°C 30°C 40°C
Peak %Main %Main %Main
DayO Peak Peak Peak
P5 99.55 98.01 97.07 92.64
P6 99.02 98.60 97.71 95.00
Physical stability test using Thioflavin T binding assay
Fibrillation is a common problem when glucagon is formulated in aqueous
solution. To assess the level of fibrillation of the compound of the present invention, a
15 Thioflavin T binding assay is performed.
The compound of SEQ ID NO: 2 is dissolved in various test buffers at 1 mg/mL
in 2.5 mL Fisher vials with a flat bottom (Fisher FS60965D) containing a flea sized
stirring bar (Fishers Catalog# 1451364). Test buffers are prepared in H20 and are all
adjusted to pH 6.0:
20 Buffer 1 = 20 mM Histidine
Buffer 2= 10 mM Histidine, 150 nM NaCl
Buffer 3= 10 mM Histidine, 300 mM sorbitol
Buffer 4= 10 mM Histidine, 0.02% Tween 80
Buffer 5= 10 mM Histidine, 300 mM Arginine
25 Buffer 6= 10 mM Histidine, 300 mM sucrose
wo 2015/094875 PCT /US2014/069643
-8-
In addition, human glucagon (SEQ ID NO: 1) is dissolved in a 12 mg/mL glycerol
solution at pH 2.8 to a final glucagon concentration of 1 mg/mL. All samples are
mechanically stressed at 25 oc in a magnetic stir plate set at 300 rpm. Aliquots of the
5 different samples (100 JlL each aliquot and done in triplicates) are taken at time points 0,
40 and 120 hours, and are added to a plate followed by 10 JlL of a 1 mM Thioflavin T (
T35516-25G, Sigma Aldrich) (1 mM stock solution in H20, pH 2.8) Samples are
incubated for 30 min. Fluorescence is measured using a Spectramax M5 (Moleculer
Devices) using 440 11m as the excitation wavelength, and the emission wavelength is set
10 at 480 11m with a 475 11m cut off and automatic sensitivity adjustment. Raw data is
collected with Softmax Pro 5.4.1 (Molecular Devices) and imported to Excel. The
average of the 3 wells per each time point becomes the reported fluorescence units shown
in Table 3 below:
15 TABLE3
Sample t=Oh t=40h t = 120 h
SEQ ID NO: 2 in Buffer 1 74.6 76.9 89.2
SEQ ID NO: 2 in Buffer 2 130.1 137.3 137.0
SEQ ID NO: 2 in Buffer 3 68.1 71.4 72.4
SEQ ID NO: 2 in Buffer 4 51.9 56.2 57.4
SEQ ID NO: 2 in Buffer 5 113.9 120.0 123.2
SEQ ID NO: 2 in Buffer 6 78.0 74.1 88.9
Human glucagon (SEQ ID NO: 1) in 12 mg/mL
glycerol, pH 2.8 37.9 620.4 1416.3
Buffer 1 33.4 36.4 34.2
Buffer 2 30.8 32.7 31.7
Buffer 3 34.5 36.5 36.2
Buffer4 27.0 28.8 26.8
Buffer 5 43.8 46.1 44.7
Buffer 6 51.0 52.8 55.0
12 mg/mL glycerol, pH 2.8 28.2 31.4 29.4
As shown in Table 3, the compound of SEQ ID NO: 2 maintains physical stability at
25°C and pH 6 in the presence of mechanical stress as assessed by both visual assessment
and Thioflavin T binding assays. The compound of SEQ ID NO: 2 did not demonstrate
wo 2015/094875 PCT /US2014/069643
-9-
fibrillation as measured by the Thioflavin T binding assay.
Effects of compound on blood glucose levels in C57 /Bl6 male mice
To determine the effects of the compound of SEQ ID NO: 2 on blood glucose
5 levels, the compound is administered to C57 /B 16 mice. Three month old male C57BL6
mice (Harlan Laboratories) are used. Animals are individually housed in a temperaturecontrolled
(24°C) facility with a 12 hour light/dark cycle, and have free access to food
and water. After 1 week acclimation to the facility, mice are randomized to treatment
groups (n= 4/group). Test compound is formulated in Buffer 2 (see Physical stability test
10 using Thioflavin T binding assay). On the morning of test, food is removed at 08:00AM.
Two hours after food is removed, the test compound is given subcutaneously at 0, 1, 3 or
10 Jlg/kg doses. Blood glucose is measured at time 0, 15, 30, 60 and 120 minutes after
test compound administration with an ACCU-CHECK® (Roche Diagnostics) glucometer.
Table 4 shows the glucose values at different time points. Results are expressed as mean
15 ± standard error mean (SEM) of 4 mice per group.
ED50 is calculated on the 30 minute glucose measurements. Blood glucose levels
at 10 Jlg/kg of compound of SEQ ID NO: 2 is taken as the maximum value. For the
compound of SEQ ID NO: 2, the ED 50 is 3.28 Jlg/kg (95% confidence interval). The
results demonstrate that the compound of SEQ ID NO: 2 is able to increase blood
20 glucose.
TABLE4
Blood glucose levels (mg/dL) after dosing with compound of
SEQIDNO: 2
Time (minutes) 0 Jlg/kg 1Jlg/kg 3Jlg/kg lOJlg/kg
0 141.0 ± 2.0 165.3 ± 16.7 148.0 ± 4.6 155.4 ± 10.1
15 185.1 ± 11.5 212.0 ± 12.6 234.1 ± 9.8 270.0 ± 12.7
30 193.5 ± 14.6 203.1 ± 13.1 242.1 ± 8.3 308.5 ± 2.6
60 170.8 ± 6.2 169.9 ± 10.1 169.4 ± 7.3 212.4 ± 8.3
120 140.2 ± 3.0 141.0 ± 4.2 148.2 ± 10.9 152.4 ± 1.5
wo 2015/094875 PCT /US2014/069643
-10-
Human Glucagon Receptor Binding Assay
The binding of the compound of SEQ ID NO: 2 is determined by using a
293HEK cell line overexpressing the human glucagon receptor (hGR) (Lok Setal. Gene
140 (2), 203-209 (1994); GenBank: L20316).
5 Crude plasma membranes are prepared using cells from suspension or adherent
culture. The cell pellets are lysed on ice in a hypotonic homogenization buffer (25 mM
Tris HCl, pH 7.5, 1 mM MgCh, and Roche Complete™ Inhibitors without EDTA
(Roche, 11873580001)) with DNAase at 20 Jlgfml (Invitrogen, 18047-019). The cell
suspension is homogenized with a glass dounce homogenizer using a Teflon pestle for 25
10 strokes. The homogenate is centrifuged 1800 X gat 4 oc for 15 min. The supernatant is
collected and the pellet is resuspended in hypotonic homgenization buffer (without
DNAse) andre-homogenized. The mixture is centrifuged at 1800 X g for 15 min. The
second supernatant is combined with the first supernatant and centrifuged at 1800 X g for
15 min to clarify. This clarified supernatant is further centrifuged at 25000 X g for 30
15 min at 4 °C. The membrane pellet is resuspended in hypotonic homogenization buffer
(without DNAse) and stored as frozen aliquots at -80 oc until use.
Human glucagon is radioiodinated by 1251-lactoperoxidase procedure and purified
by reversed phase HPLC at Perkin-Elmer/NEN (NEX207). The specific activity is about
2200 Ci/mmol. Kn determination is performed by homologous competition instead of
20 saturation binding due to high propanol content in the 1251-labelled glucagon material.
The Kn is estimated to be 1.24 11M and is used to calculate Ki values for all compounds
tested.
The receptor binding assay is carried out using a Scintillation Proximity Assay
(SPA) (Sun, S., Almaden, J., Carlson, T.J., Barker, J. and Gehring, M.R. Assay
25 development and data analysis of receptor-ligand binding based on scintillation proximity
assay. Metab Eng. 7:38-44 (2005)) with wheat germ agglutinin (WGA) beads (PerkinElmer)
previously blocked with 1% fatty acid free bovine serum albumin (BSA) (Gibco,
7.5% BSA). Glucagon (SEQ ID NO: 1) and compound (SEQ ID NO: 2) are dissolved in
dimethyl sulfoxide (DMSO) at a concentration of2 mM and stored frozen at -20 °C.
30 Glucagon and compound of SEQ ID NO: 2 are serially diluted into DMSO. 10
JlL of diluted samples are transferred into Coming 3632 clear bottom assay plates
containing 40 JlL assay Binding Buffer (25 mM 4-(2-hydroxyethyl)-1-
wo 2015/094875 PCT /US2014/069643
-11-
piperazineethanesulfonic acid (HEPES), pH 7.4, 2.5 mM CaCh, 1 mM MgCh, 0.1%
fatty acid free BSA, 0.003% Tween20, and Roche Complete Inhibitors without EDTA)
or cold glucagon (non-specific binding (NSB) at 1 JlM final). 90 JlL membranes (3
Jlgfwell), 50 JlL 125!-labelled Glucagon (0.15 11M final concentration in reaction), and 50
5 JlL ofWGA beads (150 Jlgfwell) are added. DMSO concentration does not exceed 4.2%.
Plates are sealed, mixed end over end, and read with a MicroBeta® scintillation counter
after 12 hours of settling time at room temperature.
Results are calculated as a percent of specific 125I -labelled glucagon binding in the
presence of compound. The absolute IC50 concentration of compound is derived by non-
10 linear regression of percent specific binding of 125I -labelled glucagon vs. the
concentration of sample added (8.5x10-12 to 0.5x10-7 mol/L). The IC50 dose is converted
to Ki using the Cheng-Prusoff equation (Cheng Y., PrusoffW. H., Biochem. Pharmacal.
22: 3099-3108 (1973)). The Ki ofthe compound of SEQ ID NO: 2 was 0.397 ± 0.062
11M (n=7) for hGR binding (Ki for human glucagon was 1.66 ± 0.0911M (n=47) for hGR
15 binding). This data demonstrates that the compound of SEQ ID NO: 2 binds to hGR with
increased affinity compared to human glucagon and may activate that receptor, in tum
triggering glucagon-dependent physiological responses.
Mouse Glucagon Receptor Binding Assay
20 To determine whether the compound of SEQ ID NO: 2 binds to the mouse
glucagon receptor (mGR), a binding assay as essentially described in the Human
Glucagon Receptor Binding Assayis performed. Crude plasma membranes are prepared
from 293HEK cells in suspension culture expressing a cloned mGR. ((Burcelin R, Li J,
Charron MJ. Gene 164 (2), 305-10 (1995) GenBank: L38613). Membrane pellets are
25 prepared as described in the Human Glucagon Receptor Binding Assay, resuspended in
homogenization buffer and stored as frozen aliquots at -80 oc until use.
Human glucagon is radioiodinated by 125!-lactoperoxidase procedure and purified
by reversed phase HPLC at Perkin-Elmer/NEN (NEX207). The specific activity is about
2200 Ci/mmol. Kn determination is performed by homologous competition instead of
30 saturation binding due to high propanol content in the 125!-labelled glucagon material. The
Kn is estimated to be 2.05 11M and is used to calculate Ki values for all compounds tested.
wo 2015/094875 PCT /US2014/069643
-12-
The SPA receptor binding assay and calculation of the results are carried out as
described in the Human Glucagon Receptor Binding Assay. The Ki of the compound of
SEQ ID NO: 2 was 0.494 ± 0.054 11M (n=7) for mGR binding(Ki for human glucagon
was 1.37 ± 0.0711M (n=33) for mGR binding). This data demonstrates that compound of
5 SEQ ID NO: 2 binds to mGR with increased affinity compared to human glucagon and
may activate that receptor, in tum triggering glucagon-dependent physiological responses.
Glucagon-Like-Peptide 1 Receptor Binding Assay
To determine whether the compound of SEQ ID NO: 2 binds to the human
10 glucagon-like peptide 1 receptor (hGLP-1R), a binding assay as essentially described in
the Human Glucagon Receptor Binding Assayis performed. Crude plasma membranes
are prepared from 293HEK suspension cells expressing a cloned human glucagon-like
peptide 1 receptor (hGLP-1R) (Graziano MP, Hey PJ, Borkowski D, Chicchi GG, Strader
CD, Biochem Biophys Res Commun. 196 (1):141-6 (1993) GenBank: NM_002062)
15 isolated from 293HEK membranes. Membrane pellets are prepared as described in the
Human Glucagon Receptor Binding Assay, resuspended in homogenization buffer and
stored as frozen aliquots at -80 oc until use.
Glucagon-like peptide 1 amide (GLP-1 amide) (SEQ ID NO: 3) is
radioiodinated by the 125!-lactoperoxidase procedure and purified by reversed phase
20 HPLC at Perkin-Elmer/NEN (NEX308). The specific activity is about 2200 Ci/mmol.
Kn determination is performed by homologous competition instead of saturation binding
due to high propanol content in the 125!-labelled GLP-1 material. The Kn is estimated to
be 0.329 11M and is used to calculate Ki values for all compounds tested.
The SPA receptor binding assay and calculation of the results are carried out as
25 described in the Human Glucagon Receptor Binding Assaywith the exception that
radioiodinated GLP-1 amide is used instead of the radioiodinated glucagon of the Human
Glucagon Receptor Binding Assay.
The Ki of the compound of SEQ ID NO: 2 was 1221 ± 7811M (n=6) for hGLP-1R
binding and the Ki of the glucagon (SEQ ID NO: 1) was 2098 ± 91 (n=17) (Ki for human
30 GLP-1 7-36 amide was 0.427 ± 0.16911M (n=64) for hGLP-1R binding). This data
wo 2015/094875 PCT /US2014/069643
-13-
demonstrates that the compound of SEQ ID NO: 2 binds to hGLP-1R with low affinity
and thus does not initiate GLP-1R-mediated physiological responses.
Glucose-Dependent Insulinotropic Peptide Receptor Binding Assay
5 To determine whether the compound of SEQ ID NO: 2 binds to the glucosedependent
insulinotropic peptide receptor (GIP-R), a binding assay as essentially
described in the Human Glucagon Receptor Binding Assayis performed. Crude plasma
membranes are prepared from suspension Chinese Hamster Ovary cells (CHO-S)
expressing human GIP-R (R (Usdin,T.B., Gruber,C., Modi,W. and Bonner,T.I., GenBank:
10 AAA84418.1) using cells from suspension culture. Membrane pellets are prepared as
described in the Human Glucagon Receptor Binding Assay, resuspended in
homogenization buffer and stored as frozen aliquots at -80 oc until use.
GIP (SEQ ID NO: 4) is radioiodinated by the I-125-lactoperoxidase procedure
(Markalonis, J.J., Biochem. J. 113:299 (1969)) and purified by reversed phase HPLC at
15 Perkin-Elmer!NEN (NEX-402). The specific activity is 2200 Ci/mmol. Kn
determination is performed by homologous competition using cold human GIP instead of
saturation binding. The Kn is estimated to be 0.17411M and is used to calculate Ki values
for all compounds tested.
The SPA receptor binding assay and calculation of the results are carried out as
20 described in the Human Glucagon Receptor Binding Assaywith the exception that
radioiodinated GIP is used instead of the radioiodinated glucagon of the Human Glucagon
Receptor Binding Assay.
The Ki of the compound of SEQ NO: 2 was 1054 ± 16711M (n=7) for human
GIP-R binding and the Ki of glucagon (SEQ ID NO: 1) was >3010 (n=1) (Ki for human
25 GIP 1-42 was 0.279 ± 0.02l11M, n=2). This data demonstrates that the compound of
SEQ ID NO: 2 binds to hGIP-R with low affinity and thus does not initiate hGIP-Rmediated
physiological responses.
30
Human Glucagon Receptor Stimulated cAMP Functional Assay.
The hGR stimulated cAMP functional assay uses the same cloned hGR expressing
cell line as used for the hGR binding assay described above in the Human Glucagon
Receptor Binding Assay. Cells are stimulated with glucagon, buffer controls, or Test
wo 2015/094875 PCT /US2014/069643
-14-
samples, and the cAMP generated within the cell is quantitated using the CisBio cAMP
Dynamic 2 HTRF Assay Kit (62AM4PEC). Briefly, cAMP levels within the cell are
detected by binding to the cAMP-d2 capture antibody in the presence of cell lysis buffer.
A second detection antibody provided in the kit, anti-cAMP Cryptate, is added to create a
5 competitive sandwich assay. When the detection antibody complex formed there is an
increase in the signal that is measured on a Perkin-Elmer Envision® instrument.
The hGR-HEK293 cells are harvested from sub-confluent tissue culture dishes
with Enzyme-Free Cell Dissociation Solution (Specialty Media 5-004-B). The cells are
pelleted at 100 X g at room temperature for 5 minutes then washed twice with phosphate
10 buffered saline (PBS). The washed cell pellet is resuspended at 1x 107 cells/ml in
Recovery™ Freeze Media (Gibco 2044) and frozen in liquid nitrogen. On the day of
treatment, a frozen aliquot of cells is transferred into pre-warmed Resuspension Cell
Media (DMEM, Gibco (31053P) containing 0.5% defined FBS (Hyclone SH30070); 20
mM HEPES, pH 7.4; and 2 mM Glutamine). The cells are then pelleted at 100 X gat
15 room temperature for 5 minutes. The supernatant is removed and the cell pellet is
resuspended in Cell Media (DMEM, Gibco (31053P) with 0.1% fatty acid-free bovine
serum albumin, BSA, 7.5%, (Gibco 15620); 20 mM HEPES, pH 7.4, and 2 mM
Glutamine) at 1.25x105 cells/mi. Test samples are prepared as 2 mM stocks in DMSO
and frozen at -20°C until needed. Glucagon, buffer controls and compound of SEQ ID
20 NO: 2, are serially diluted into DMSO followed by a step-down dilution into Compound
Dilution Media (Assay Media (DMEM, Gibco 31053P with 0.1% fatty acid-free bovine
serum albumin, BSA, 7.5%, (Gibco 15620); 20 mM HEPES, pH 7.4, and 2 mM
Glutamine) that contains 500 JlM IBMX). The reaction is performed in 40 JlL, by adding
20 JlL of cells (2500 cell/well) or cAMP standard curve samples to 96 Well plate Half
25 Area Black plates (Costar 3694), followed by addition of 20 JlL of either 2X
concentrated glucagon, buffer controls or compound of SEQ ID NO: 2 in Compound
Dilution Media. Final DMSO concentration does not exceed 1.1 %, and final IBMX
concentration is 250 JlM. The reaction is stopped by addition of 20 JlL of the cAMP-d2-
capture antibody (CisBio) diluted into the CisBio lysis buffer then gently mixed in
30 TITERTEK shaker. After 5 minutes oflysis, 20 JlL of the detection antibody, anti-cAMP
Cryptate (CisBio ), is added and mixed at 600 rpm for 1 minute, then shaken at 300 rpm.
The lysed cell and antibody mixtures are read after 1 hour at room temperature using the
wo 2015/094875 PCT /US2014/069643
-15-
Perkin-Elmer Envision®. Envision® units were converted to pmol/L cAMP/well using
the cAMP standard curve. The picomoles of cAMP generated in each well is converted
to a percent of the maximal response observed with the glucagon control. A relative EC50
value is derived by non-linear regression analysis using the percent maximal response vs.
5 the concentration (0.17x 1 o-12 to 1 X 1o-8M) of peptide added.
10
The compound of SEQ ID NO: 2 bound hGR with an EC5o of0.0356 ±0.007111M
(n=8) (EC5o for human glucagon was 0.0142±0.001811M, n= 6). This data demonstrates
that the compound of SEQ ID NO: 1 binds and activates hGR and can thereby initiate
glucagon receptor-mediated physiological responses.
wo 2015/094875
Sequence Listing
Human glucagon:
PCT /US2014/069643
-16-
His-Ser-Gln-Gly-Thr-Phe-Thr-Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-
5 Ser-Arg-Arg-Ala-Gln-Asp-Phe-Val-Gln-Trp-Leu-Met-Asn-Thr(SEQ
ID NO: 1)
Example 1
His-Ala-Gln-Gly-Thr-Phe-Leu-Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-
10 (Aib)-Lys-Lys-Ala-Gln-Glu-Phe-Val-Glu-Trp-Leu-Leu-Lys-ThrGly-
Pro-Ser-Ser-Gly-Ala- Pro- Pro-Lys-Ser- Lys-NH2 (SEQ ID NO: 2)
Human GLP-1:
15 His-Ala-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-
Gly-Gln-Ala-Ala-Lys-Glu-Phe-Ile-Ala-Trp-Leu-Val-Lys-Gly-ArgNH2
(SEQ ID NO: 3)
20 Human GIP:
Tyr-Ala-Glu-Gly-Thr-Phe-Ile-Ser-Asp-Tyr-Ser-Ile-Ala-Met-AspLys-
Ile-His-Gln-Gln-Asp-Phe-Val-Asn-Trp-Leu-Leu-Ala-Gln-LysGly-
Lys- Lys-Asn-Asp-Trp-Lys-His-Asn- Ile-Thr-Gln (SEQ ID N0:4)
X19933SequenceListing11SEP2014ST25
SEQUENCE LISTING
<110> Eli Lilly and Company
<120> NOVEL COMPOUND FOR TREATMENT OF SEVERE HYPOGLYCEMIA
<130> X19933
<150> 61/917597
<151> 2013-12-18
<160> 4
<170> PatentIn version 3.5
<210> 1
<211> 29
<212> PRT
<213> Homo sapiens
<400> 1
His Ser Gln Gly Thr Phe Thr Ser Asp Tyr Ser Lys Tyr Leu Asp Ser
1 5 10 15
Arg Arg Ala Gln Asp Phe Val Gln Trp Leu Met Asn Thr
20 25
<210> 2
<211> 40
<212> PRT
<213> Artificial Sequence
<220>
<223> Synthetic Construct
<220>
<221> MISC_FEATURE
<222> (16)..(16)
<223> Xaa at position 16 is 2-aminoisobutyric acid
<220>
<221> MOD_RES
<222> (40)..(40)
<223> AMIDATION
<400> 2
His Ala Gln Gly Thr Phe Leu Ser Asp Tyr Ser Lys Tyr Leu Asp Xaa
1 5 10 15
Lys Lys Ala Gln Glu Phe Val Glu Trp Leu Leu Lys Thr Gly Pro Ser
20 25 30
Ser Gly Ala Pro Pro Lys Ser Lys
35 40
<210> 3
<211> 30
<212> PRT
<213> Homo sapiens
Page 1
X19933SequenceListing11SEP2014ST25
<220>
<221> MOD_RES
<222> (30)..(30)
<223> AMIDATION
<400> 3
His Ala Glu Gly Thr Phe Thr Ser Asp Val Ser Ser Tyr Leu Glu Gly
1 5 10 15
Gln Ala Ala Lys Glu Phe Ile Ala Trp Leu Val Lys Gly Arg
20 25 30
<210> 4
<211> 42
<212> PRT
<213> Homo sapiens
<400> 4
Tyr Ala Glu Gly Thr Phe Ile Ser Asp Tyr Ser Ile Ala Met Asp Lys
1 5 10 15
Ile His Gln Gln Asp Phe Val Asn Trp Leu Leu Ala Gln Lys Gly Lys
20 25 30
Lys Asn Asp Trp Lys His Asn Ile Thr Gln

WE CLAIM:
1. A compound comprising the amino acid sequence ofHis-Ala-Gln-Gly-Thr-PheLeu-
Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-(Aib)-Lys-Lys-Ala-Gln-Glu-Phe-Val-
5 Glu-Trp-Leu-Leu-Lys-Thr-Gly-Pro-Ser-Ser-Gly-Ala-Pro-Pro-Lys-Ser-Lys-NH2
(SEQ ID NO: 2).
2. The compound of Claim 1, wherein the compound consists of the amino acid
sequence of His-Ala-Gln-Gly-Thr-Phe-Leu-Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-
10 (Aib)-Lys-Lys-Ala-Gln-Glu-Phe-Val-Glu-Trp-Leu-Leu-Lys-Thr-Gly-Pro-Ser-Ser-
15
Gly-Ala-Pro-Pro-Lys-Ser-Lys-NH2 (SEQ ID NO: 2).
3. A pharmaceutical composition comprising the compound of Claim 1 or Claim 2
and a pharmaceutically acceptable buffer.
4. The pharmaceutical composition of Claim 3, wherein the pharmaceutically
acceptable buffer is histidine-buffered saline.
5. A method of treating hypoglycemia in a subject comprising administering an
20 effective amount of a compound of Claim 1 or Claim 2.
25
30
6. A compound of Claim 1 or Claim 2 for use in a therapy.
7. A compound of Claim 1 or Claim 2 for use in the treatment ofhypoglycemia.
8. A compound of Claim 1 or Claim 2 for use in the manufacture of a medicament
for the treatment of hypoglycemia.