Glucagon-related peptides from phylogenetically ancient fish reveal new

approaches to the development of dual GCGR and GLP1R agonists for Type 2

diabetes therapy

Galyna V. Graham, J. Michael Conlon*, Yasser H. Abdel-Wahab, Peter R. Flatt

Diabetes Research Group, School of Biomedical Sciences, Ulster University, Cromore

Road, Coleraine, Northern Ireland BT52 1SA, UK

*Corresponding author:

Prof. J. M. Conlon

School of Biomedical Sciences

Ulster University,

Cromore Road,

Coleraine,

Northern Ireland BT52 1SA, UK

E-mail: m.conlon@ulster.ac.uk

Short title: Dual agonist glucagon-related peptides

Key words: insulinotropic; antidiabetic; glucagon, GLP-1, oxyntomodulin, exendin-4,

paddlefish

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

Abstract

The insulinotropic and antihyperglycaemic properties of glucagons from the sea

lamprey (Petromyzontiformes), paddlefish (Acipenseriformes) and trout (Teleostei) and

oxyntomodulin from dogfish (Elasmobranchii) and ratfish (Holocephali) are compared

with those of human glucagon and GLP-1 in mammalian test systems. All fish peptides

produced concentration-dependent stimulation of insulin release from BRIN-BD11 rat

and 1.1 B4 human clonal β-cells and isolated mouse islets. Paddlefish glucagon was the

most potent and effective peptide. The insulinotropic activity of paddlefish glucagon

was significantly (P < 0.01) decreased after incubating BRIN-BD11 cells with the

GLP1R antagonist, exendin-4(9-39) and the GCGR antagonist [des-His1,Pro4 ,Glu9]

glucagon amide but GIPR antagonist, GIP(6-30)Cex-K40[Pal] was without effect.

Paddlefish and lamprey glucagons and dogfish oxyntomodulin (10 nmol/L) produced

significant (P < 0.01) increases in cAMP concentration in Chinese hamster lung (CHL)

cells transfected with GLP1R and human embryonic kidney (HEK293) cells transfected

with GCGR. The insulinotropic activity of paddlefish glucagon was attenuated in

CRISPR/Cas9-engineered GLP1R knock-out INS-1 cells but not in GIPR knock-out

cells. Intraperitoneal administration of all fish peptides, except ratfish oxyntomodulin,

to mice together with a glucose load produced significant (P < 0.05) decreases in

plasma glucose concentrations and paddlefish glucagon produced a greater release of

insulin compared with GLP-1. Paddlefish glucagon shares the sequences Glu15-Glu16

and Glu24-Trp25-Leu26-Lys27-Asn28-Gly29 with the potent GLP1R agonist, exendin-4 so

may be regarded as a naturally occurring, dual-agonist hybrid peptide that may serve as

a template design of new drugs for type 2 diabetes therapy.

2

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

1. Introduction

Patients with obesity-related diseases such as Type 2 diabetes mellitus (T2DM) often

cannot be treated effectively with a single pharmaceutical agent [1]. The therapeutic

potential of unimolecular dual- and triple-agonist peptides based upon the primary

structures of the incretin hormones glucagon-like peptide-1 (GLP-1) and glucose-

dependent insulinotropic peptide (GIP) together with glucagon is becoming

increasingly apparent. Multi-agonist peptides that target (A) the GLP-1 receptor

(GLP1R) and the glucagon receptor (GCRG), (B) the GLP1R and the GIP receptor

(GIPR), and (C) the GLP1R, GCGR, and GIPR are currently in development [2,3].

Similarly, an enzymatically stable GIP/xenin hybrid peptide has been shown to enhance

β-cell function and improves glucose homeostasis in mice that have been rendered

insulin resistant and glucose intolerant by being fed a high fat diet [4]. A major

advantage of administration of these multi-agonist peptides over their constituent

monotherapies is their improved actions on blood glucose control, appetite,

suppression, weight loss and hyperlipidemia/hypercholesterolemia although some

patients have exhibited adverse effects including  diarrhoea, nausea and decreased

appetite [3,5].

An alternative strategy to the use of synthetic hybrid peptides is to exploit the

potential of naturally occurring dual agonist peptides. Oxyntomodulin is a C-terminally

extended form of glucagon that was first identified as an impurity in a commercially

available preparation of bovine/porcine glucagon [6] and is released from L-cells in the

intestine in response to nutrients [7]. A specific receptor for oxyntomodulin has not

been identified but the peptide will activate both the GCGR and the GLP1R, although

with an affinity that is appreciably less than glucagon and GLP-1 respectively.

3

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Consequently, the peptide will stimulate insulin release and improve glucose tolerance

when administered to mice [8,9]. Oxyntomodulin has received particular attention

because of its ability at pharmacological doses to inhibit appetite and food intake [10].

A GIP-oxyntomodulin hybrid peptide acting through GIP, glucagon and GLP-1

receptors exhibits promising weight reducing and anti-diabetic properties during long-

term administration to high fat fed mice [11]. Recently it has been shown that glucagon

from the European common dogfish Scyliorhinus canicula (Elasmobranchii) [12] acts

as a dual agonist at the glucagon and GLP-1 receptors and peptidase-resistant analogs

showed potent insulin releasing activity in vitro and antihyperglycaemic activity in

mice [13,14]. Similarly, the insulinotropic and glucose-lowering actions of zebrafish

GIP in mice involve interaction with both the GLP-1 and GIP receptors [15].

Glucagon has been isolated from the principal islets (Brockmann bodies) of a

wide range of teleost species and the genes encoding proglucagon in several teleosts

have been characterized (reviewed in [16]). The peptide has also been purified from

several phylogenetically more ancient fishes. Glucagon has been isolated from the islet

organ of the lampreys Petromyzon marinus, Lampetra fluviatilis, and Geotria australis

(Petromyzontiformes) [reviewed in (17)], from the pancreas of the ray Torpedo

marmorata (Elasmobranchii), and from the pancreata of several primitive

Actinopterygian (ray-finned) fish: the N. American paddlefish Polyodon spathula

(Polyodontidae) and the sturgeons Huso dauricus and Scaphirhynchus albus

(Acipenseriformes), the bichir Polypterus senegalis (Polypteriformes), the alligator gar

Lepisosteus spatula (Lepisosteiformes), and the bowfin Amia calva (Amiiformes)

(reviewed in [18]). Components comprising glucagon extended from its COOH-

terminal region by additional amino acid residues that may represent orthologs of

mammalian oxyntomodulin have been purified from the pancreata of S. canicula and

4

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

the Pacific ratfish Hydrolagus colliei (Holocephali) [12,19]. It is estimated that the

Petromyzontiformes diverged from the line of evolution leading to mammals between

500 and 550 million years ago (MYA) and the Elasmobranchii diverged around 420

MYA. The divergence of the Holocephali and the Elasmobranchii is also ancient (410

MYA) and the Polyodontidae are believed to have diverged from the Acipenseriformes

approximately 180 MYA [20].

The biological activities and physiological roles of the proglucagon-derived

peptides have not been investigated the ancient fishes but, in teleosts, the

responsiveness of hepatocytes to glucagon is limited to high concentrations while

physiological concentrations of GLP-1 effectively regulate hepatic metabolism by

stimulating gluconeogenesis and glycogenolysis. In contrast to mammals, GLP-1 shows

very weak insulinotropic activity in teleosts [21]. In the light of the promising anti-

diabetic effects of dogfish glucagon [13,14], the aim of the present study was to

compare the insulinotropic and antihyperglycemic properties of synthetic replicates of

glucagons from the sea lamprey [22], paddlefish [23], and rainbow trout Onchorynchus

mykiss [24] and oxyntomodulins from dogfish [12] and ratfish [19] with human

glucagon and GLP-1. Effects in vitro were determined using wild-type and genetically

engineered established insulin-producing cell lines and isolated mouse islets. Their

glucose-lowering and insulin stimulating actions in vivo were assessed following acute

administration to NIH Swiss mice together with a glucose load. The primary structures

and molecular masses of the peptides used in this study are shown in Table 1.

2. Materials and methods

2.1. Peptides

5

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

All peptides were supplied in crude form by EZBiolab Inc. (Carmel, IN, USA). The

peptides were purified to > 98% homogeneity by reversed-phase HPLC on a (2.2 cm x

25 cm) Vydac 218TP1022 (C-18) column (Grace, Deerfield, IL, USA). The

concentration of acetonitrile in the eluting solvent was raised from 21% to 56 % over

60 min using a linear gradient. Absorbance was measured at 214 nm and the flow rate

was 6 ml/min. The identities of the peptides were confirmed by matrix-assisted laser

desorption ionization time-of-flight mass (MALDI-TOF) spectrometry using a Voyager

DE PRO instrument (Applied Biosystems, Foster City, USA) as previously described

[25].

2.2. In vitro insulin release studies using BRIN-BD11 and 1.1B4 cells

The abilities of the purified synthetic peptides (0.01 nmol L-1 - 3 µmol L-1; n = 8) to

stimulate the rate of release of insulin from BRIN-BD11 rat clonal β-cells [26] and

1.1B4 human-derived pancreatic β-cells [27] was determined as previously described

[28] In a second series of experiments, the effects of 1 µmol L-1 concentrations of the

(A) GLP-1 receptor antagonist, exendin-4(9-39) [29], (B) glucagon receptor antagonist,

[desHis1,Pro4,Glu9]glucagon amide [30], and (C) glucose-dependent insulinotropic

peptide (GIP) receptor antagonist, GIP(6-30)Cex-K40[palmitate] [31] on the insulin-

releasing activity of the peptides (0.1 µmol/L) were studied by incubating BRIN-BD11

cells for 20 min at 37 ˚C in Krebs-Ringer Bicarbonate (KRB) buffer, pH 7.4

supplemented with 5.6 mmol L-1 glucose as previously described [15]. All samples

were stored at −20 °C prior to measurement of insulin release by radioimmunoassay

[29].

6

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

2.3. Insulin release studies using isolated mouse islets

Collagenase-isolated pancreatic islets from adult, male National Institutes of Health

(NIH) Swiss mice (Harlan Ltd, Bicester, UK) [30] were cultured for 48 h at 37 °C in an

atmosphere of 5% CO2 and 95% air as previously described [15, 28] The islets were

incubated with peptides (10 nmol L-1 and 1 µmol L-1; n =4) and alanine (10 mmol L-1; n

= 4)) for 1 h at 37˚C in KRB buffer supplemented with 16.7 mmol L-1 glucose. Insulin

release was measured by radioimmunoassay and the insulin content was measured by

extraction with acid-ethanol as previously described [28].

2.4 Effects on cAMP production

CHL Chinese hamster lung cells transfected with the human GLP-1 receptor (GLP1R)

[29] and HEK293 human embryonic kidney cells transfected with the human

glucagon receptor (GCGR) [34] were cultured, seeded, and incubated with KRB

buffer containing 0.1% (w/v) bovine serum albumin, 5.6 mmol L-1 glucose, 200 μmol

L-1 3-isobutyl-1-methylxanthine (IBMX) and test peptides (10 nmol L-1 and 1 µmol L-1)

as previously described [15] After 20 min, the medium was removed and cells were

lysed before measurement of cAMP using a Parameter cAMP assay kit (R&D Systems,

Abingdon, UK) according to the manufacturer’s recommended protocol.

2.5. Insulin release studies using CRISPR/Cas9-engineered INS-1 cells

Wild-type INS-1 832/3 rat clonal pancreatic β-cells and CRISPR/Cas9-engineered cells

with knock-out of either the GLP-1 receptor (GLP1R KO) or the GIP receptor (GIPR

7

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

KO) [35] were cultured as described for BRIN-BD11 cells but with the addition of

1 mmol L-1 sodium pyruvate and 50 mmol L-1 2-mercaptoethanol to the medium [15].

The cells were incubated with 10 nmol L-1 and 1 μmol L-1 concentrations of GLP-1,

glucagon, GIP, and fish peptides for 20 min 37 °C in KRB buffer supplemented with

5.6 mmol L-1 glucose and insulin release was measured by radioimmunoassay.

2.6. In vivo insulin release studies

All experiments involving animals were performed in accordance with the UK Animals

(Scientific Procedures) Act 1986, EU Directive 2010/63EU, the European Convention

for the Protection of Vertebrate Animals used for Experimental and other Scientific

Purposes' (Council of Europe No 123, Strasbourg 1985 and approved by Ulster

University Animal Ethics Review Committee. Acute in vivo studies were carried out

using 8-10 week-old male National Institutes of Health (NIH) Swiss mice (n = 6)

maintained in an environmentally controlled room (12 h:12 h light/darkness cycle, 22 ±

2 oC) with free access to a standard rodent diet and water. Peptides were dissolved in

0.9 % NaCl-water and stored as a 1 mmol L-1 stock solution. After an overnight fast,

glucose (18 mmol kg body weight-1), either alone or together with peptide (25 nmol kg

body weight-1), was administrated intraperitoneally in a volume of 100 µL to the mice

as previously described [28]. Blood samples (approx. 60 µL) were collected from a tail

vein at the time points shown in Fig. 7. The animals receiving peptides did not exhibit

any behavioural differences from control animals when monitored over a 24 h period.

8

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

2.7. Statistical analysis

Results were analysed in GraphPad PRISM (Version 5.0) and presented as mean ±

SEM. Statistical analyses were carried out by unpaired Student's t test (non-parametric,

with two-tailed P values and 95% confidence interval) and by one way ANOVA with

Bonferroni post-hoc test wherever applicable. Results were considered significant if P <

0.05. Area under the curve (AUC) analysis was performed using the trapezoidal rule

with baseline correction.

3. Results

3.1. In vitro insulin-releasing activity

The rate of insulin release from BRIN-BD11 rat clonal β-cells in the presence of 5.6

mmol L-1 glucose alone was 1.02 ± 0.02 ng 106 cells-1 20 min-1. The effects of

incubation with increasing concentrations of the fish glucagon-related peptides, human

glucagon, and human GLP-1 on insulin release are shown in Fig. 1. All peptides

produced significant and concentration-dependent increases the rate of insulin release

compared with the rate in the presence of 5.6 mmol L-1 glucose alone. As shown in

Table 2, the most potent peptide was paddlefish glucagon which produced a significant

stimulation of insulin release (P < 0.05) from BRIN-BD11 cells at a threshold

concentration of 30 pmol L-1 compared with 10 pmol L-1 for human GLP-1 and 10 nmol

L-1 for human glucagon. Paddlefish glucagon was the most effective peptide producing

a response at 3 µmol L-1 concentrations that was not significantly different from that of

GLP-1 (Table 2).

The rate of insulin release from 1.1B4 human clonal β-cells in the presence of

16.7 mmol L-1 glucose alone was 0.08 ± 0.01 ng 106cells-1 20 min-1. Incubation with all

9

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

fish peptides significantly (P < 0.001) stimulated insulin release at concentrations of 1

µmol L-1 compared with the rate in the presence of 16.7 mmol L-1 glucose alone (Fig.

2). Only paddlefish glucagon (P < 0.001) and dogfish oxyntomodulin (P < 0.05)

produced a significant increase at 10 nmol L-1. The rates of insulin release elicited by 1

µmol L-1 concentrations of glucagons from lamprey and paddlefish and by dogfish

oxyntomodulin were not significantly different to that produced by human GLP-1.

The rate of insulin release from mouse islets incubated for 60 min with 16.7

mmol L-1 glucose was 9.7 ± 1.3 % of the total insulin content. Incubation with all

peptides, except human and trout glucagon, at 1µM concentrations significantly

increased the rate of insulin release compared with the rate with 16.7 mmol L-1 glucose

alone (Fig. 3). Lamprey glucagon and dogfish oxyntomodulin were also effective at 10

nmol L-1. The stimulatory effects of 1 µmol L-1 concentrations of glucagons from

lamprey and paddlefish and dogfish oxyntomodulin on insulin release were not

significantly different from the effect produced by 1 µmol L-1 GLP-1.

3.2. Receptor antagonist studies

The in vitro insulinotropic activities of all peptides, except trout and human glucagons

and GIP, were significantly decreased when BRIN-BD11 cells were co-incubated with

the GLP-1 receptor antagonist, exendin-4(9-39) (Fig. 4A). Co-incubation with the

glucagon receptor antagonist [des-His1,Pro4,Glu9]glucagon amide significantly

attenuated the action of all peptides except ratfish oxyntomodulin, GLP-1 and GIP

(Fig. 4B). Human GIP was the only peptide whose stimulatory effect on insulin release

was attenuated by the GIP receptor antagonist, GIP(6-30)Cex-K40[palmitate] (Fig. 4C).

10

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

3.3. Effects on cAMP production

At the high concentration of 1 µmol L-1, all peptides except trout glucagon and GIP

significantly stimulated cAMP production in CHL cells transfected with the human

GLP-1 receptor (GLP1R) [29] (Fig. 5A). However, at a concentation of 10 nmol L-1

only incubations with lamprey and paddlefish glucagons, dogfish oxyntomodulin, and

GLP-1 produced significant (P < 0.001) increases in cAMP production. Similarly, at 1

µmol L-1 concentration significant stimulation of cAMP production in HEK293 cells

transfected with the human glucagon receptor (CGCR) [34] was observed after

incubating the cells with all peptides, except ratfish oxyntomodulin and GIP (Fig. 5B).

At the lower concentration of 10 nmol L-1, only lamprey and paddlefish glucagons and

dogfish oxyntomodulin among the fish peptides significantly (P < 0.05) stimulated the

activation of adenylate cyclase.

3.4. Insulin releasing activities using CRISPR/Cas9-engineered INS-1 cells

Incubation of wild type INS-1 cells with glucagon, GLP-1, GIP, and all fish peptides

(10 nmol L-1 and 1 μmol L-1) significantly increased the rate of insulin release

compared with the rate in the presence of 5.6 mmol L-1 glucose alone (Fig. 6A). The

stimulatory effect on insulin release in response to incubations with lamprey glucagon,

dogfish oxyntomodulin, and GLP-1 (10 nmol L-1 and 1 µmol L-1) was abolished in the

GLP-1 KO cells and the responses to paddlefish glucagon and ratfish oxyntomodulin

were significantly attenuated compared with the effects in wild-type INS-1 cells (Fig.

6B). In contrast, the insulin release from GIP KO cells was significantly less that from

the wild-type INS-1 cells only in the case of incubations with 10 nmol/L and 1 μmol/L

11

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

human GIP (Fig. 6C).

3.5. In vivo insulin release studies

Blood glucose concentrations of overnight fasted NIH Swiss mice were significantly

lowered at time points 15, 30 and 60 min after receiving intraperitoneal administration

of glucose (18 mmol kg body weight-1) along with 25 nmol kg body weight-1 of lamprey

glucagon, dogfish oxyntomodulin, and paddlefish glucagon (Fig. 7A) and trout

glucagon (Fig. 7C) compared with animals receiving glucose only. The integrated

responses to the lamprey, dogfish, and paddlefish peptides were not significantly

different from the response to GLP-1 (Fig. 7E). Concomitant with the lower blood

glucose levels, the concentrations of plasma insulin were significantly greater 15 and 30

min after administration of lamprey glucagon, dogfish oxyntomodulin, and paddlefish

glucagon (Fig. 7B), and trout glucagon (Fig 7D) compared with animals receiving

glucose only. The integrated insulin response to paddlefish glucagon was significantly

greater than the response to human GLP-1 (Fig. 7F).

4. Discussion

This study has extended earlier work with dogfish glucagon [13,14] to show that in

mammalian test systems the insulinotropic actions of glucagons from the sea lamprey

and paddlefish and oxyntomodulin from the dogfish are mediated through activation of

both the glucagon and the GLP-1 receptors. In contrast, the effect of ratfish

oxyntomodulin is mediated predominantly, if not exclusively, through interaction with

the GLP-1 receptor and the effect of trout glucagon through interaction with the

glucagon receptor. No activation of the GIPR receptor by any fish peptide studied was

indicated. These conclusions are supported by consistent data involving the use of

12

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15161718

19

20

21

22

23

24

25

26

specific receptor antagonists (Fig. 4), cells transfected with GLP1R and GCGR (Fig. 5),

and CRISPR/Cas9-engineered GLP1R knock-out and GIPR knock-out cells (Fig. 6).

Paddlefish glucagon was the most effective peptide producing a near-maximal increase

in the rate of release of insulin from BRIN-BD11 rat clonal β-cells at a concentration of

3 µmol/L that was not significantly different from that produced by human GLP-1

(Table 2). Along with dogfish oxyntomodulin, paddlefish glucagon produced the

greatest responses in 1.1B4 human clonal β-cells (Fig. 2) and, along with lamprey

glucagon, in isolated mouse islets (Fig. 3). When administered intraperitoneally to

overnight-fasted mice together with a glucose load, glucagons from lamprey and

paddlefish and oxyntomodulin from dogfish were equally effective as human GLP-1 in

lowering blood glucose concentrations and paddlefish glucagon produced a

significantly greater insulin response than an equivalent dose of GLP-1 (Fig. 7).

In an extensive series of articles from the laboratory of Merrifield, the amino

acids in glucagon responsible for receptor binding and signal transduction were

identified. It was established that His1, Asp9, Ser11, and Ser16 are involved in receptor

activation and Ser2, Ser8 and Asp15 are important in receptor binding [36]. In addition, it

was shown that the positively charged residues Lys12, Arg17, and Arg18 are necessary to

ensure high biological potency by stabilizing the ligand-receptor interaction [37].

Structure-activity studies of human GLP-1 have shown that the amino acids His1, Gly4,

Phe6, Thr7, and Asp9 in the N-terminal domain play an important role in binding and

activation of the GLP-1 receptor and Phe22 and Ile23 in the C-terminal domain are of

critical importance in maintaining the conformation of the molecule that is recognized

by the receptor [38]. In addition, it is known that the Ala18, Ala19, Lys20, and Leu26

residues of GLP-1 interact with N-terminal extracellular domain of the receptor and so

influence binding affinity [39]. The primary structures of the fish peptides are

13

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

compared in Fig. 8 with those of glucagon, oxyntomodulin, GLP-1, and the GLP1R

agonist, exendin-4 that was first isolated from the venom of a reptile, the Gila monster

Heloderma suspectum [40].

It is apparent that the primary structure of glucagon has been better preserved

during the course of evolution than that of GLP-1. The His1, Asp9, Ser11, and Lys12

residues of glucagon have been conserved in all fish peptides and a positively charged

residue (Arg or Lys) is present at position 17. Similarly, the His1, Gly4, Phe5, Asp9,

Phe22, and Leu26 residues of GLP-1 have also been conserved in all fish peptides tested

and all peptides contain a Val residue instead of Ile at position 23. Thus, sequence

differences at these sites are not responsible for the observed differences in activity and

specificity of the fish peptides. The high potency of paddlefish glucagon was

unexpected in view of the fact that it contains the substitutions Ser8 → Asn and Ser16 →

Glu which would be expected to reduce ability to bind to and activate GCGR and Ala18

→ Arg which would be expected to reduce binding affinity to GLP1R. However, the

peptide contains the segments Glu15-Glu16 and Glu24-Trp25-Leu26-Lys27-Asn28-Gly29 in

common with GLP1R agonist, exendin-4 that are not found in the other fish peptides

(Fig. 8).

Exendin-4 is both more potent and has a longer half-life in the circulation than

GLP-1 and is in use in clinical practice as exenatide [41]. Thus, paddlefish may be

regarded as a naturally occurring glucagon/exendin-4 hybrid peptide with dual agonist

activity at the glucagon and GLP-1 receptors. A previous study has shown that twice-

daily administration of the hybrid peptide [S2s]glucagon-exendin-4 (31-39) to high fat-

fed mice for 28 days reduced body weight, energy intake and non-fasting glucose

levels, as well as increasing insulin concentrations and improving glucose tolerance and

insulin sensitivity [42]. Glucagon/exendin-4 hybrid peptides have also been designed

14

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

that show agonist activity at GLP1R and antagonist activity at GCGR [43]. GLP-1 and

glucagon lack defined secondary structure in aqueous solution, but in membrane-

mimetic environments, adopt an α-helical structure in the midsection, with flexible N-

and C-terminal regions and it has been suggested that the helical structure is required

for binding to their respective receptors [43]. The conformations of the fish glucagons

have yet to be determined experimentally but application of the AGADIR program, an

algorithm based on the helix/coil transition theory which predicts the helical behaviour

of monomeric peptides [44] indicates that paddlefish glucagon has a very strong

propensity to adopt a stable α-helical conformation between residues 8-27 whereas the

predicted helices adopted by the dogfish, ratfish, and lamprey peptides are much less

stable. In addition, it is suggested that the relatively weak insulinotropic activity of trout

glucagon may arise from the substitutions Asp15 → Leu decreasing binding affinity at

GCGR and Thr7 → Ser and Lys20 → Gln decreasing affinity at GLP1R. In common

with most teleost fish studied to-date, the trout genome contains two genes encoding

proglucagon that are presumed to have arisen as a result of an ancient whole genome

duplication event [16]. Trout glucagon II contains five amino acid substitutions (His1 →

Gln, Asp9 →Tyr, Gln20 → Arg, Val23 → Leu and Gln24 → His) compared with the

peptide used in this study [24]. The insulinotropic and glucose-lowering activities of

this paralog have yet to be determined but, in light of the substitutions at positions 1

and 9, the peptide is unlikely to display glucagon-like biological activities in

mammalian test systems.

Structure-activity relationships of the oxyntomodulin molecule have been less

well studied than those of glucagon. The insulinotropic activities and receptor

selectivities of ratfish and dogfish oxyntomodulin differ appreciably and it is suggested

that the substitution Ser2 → Thr in the ratfish peptide contributes to its inability to

15

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

activate GCGR and the substitutions Thr7 → Ser and Ala19 → Thr are, at least in part,

responsible for its lower potency and effectiveness in stimulating GLP1R-mediated

insulin release in vitro and weaker glucose-lowering activity in vivo compared with

dogfish oxyntomodulin. Replacement of residues 15-23 in oxyntomodulin (Asp-Ser-

Arg-Arg-Ala-Gln-Asp-Phe-Val) by the corresponding section of exendin-4 (Glu-Glu-

Glu-Ala-Val-Arg-Leu-Phe-Val) resulted in an approximately 10-fold increase in

binding affinity at the rat GLP-1 receptor [45] but neither dogfish not ratfish

oxyntomodulins show structural similarity with exendin-4 in this region.

In conclusion, this study has extended earlier work that identified dogfish

glucagon as a peptide that interacts with both the GLP-1 and glucagon receptors to

stimulate insulin release and lower blood glucose concentrations [13,14] by

demonstrating that glucagon-related peptides from other phylogenetically ancient fish

possess similar properties. In particular, paddlefish glucagon, by incorporating

structural features found in exendin-4, is more potent in stimulating insulin release from

rat clonal β-cells than dogfish glucagon (threshold concentration 30 pmol L-1 versus 100

pmol L-1) and provokes a greater insulin release response than human GLP-1 when

administered to mice. In the light of the global pandemic of obesity-related T2DM,

there is an urgent need for new types of safe and more effective anti-diabetic agents.

The structure-activity properties of fish glucagon-related peptides revealed for the first

time in this study will aid in the design of new therapeutic agents. Paddlefish glucagon

in particular has the potential to act as a template for development of new long-acting,

peptidase-resistant analogues for treatment of patients with the disease that display the

advantages of unimolecular dual agonist peptides [2,3].

16

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Acknowledgements

The work was supported by the Northern Ireland Department of Education and

Learning (DEL) and Ulster University Strategic Funding. The authors wish to thank

Bernard Thorens (University of Lausanne, Switzerland) for GLP-1 receptor-transfected

CHL cells, Cecilia Unson (The Rockefeller University, USA) for glucagon receptor-

transfected HEK293 cells, and Jaqueline Naylor (MedImmune, Cambridge, UK) for

CRISPR/Cas9-engineered INS-1 cells.

Declaration of interest

The authors declare no conflict of interest

Author Contributions

P.R.F., Y.H.A., and J.M.C. conceived and designed the study. GVG performed the in

vitro and animal experiments. GVG and JMC purified the synthetic peptides.and wrote

the manuscript. All authors analyzed and interpreted the data and have approved the

final submission.

17

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

References

1. C.J. Bailey, Glucose-lowering therapies in type 2 diabetes: Opportunities and

challenges for peptides, Peptides 100 (2018) 9-17.

2. N. Khajavi, H. Biebermann, M. Tschöp, R. DiMarchi, Treatment of diabetes and

obesity by rationally designed peptide agonists functioning at multiple metabolic

receptors, Endocr. Dev. 32 (2017) 165-182.

3. S.J. Brandt, A. Götz, M.H. Tschöp, T.D. Müller, Gut hormone polyagonists for

the treatment of type 2 diabetes, Peptides 100 (2018) 190-201.

4. A. Hasib, M.T. Ng, D. Khan, V.A.Gault, P.R.Flatt, N. Irwin N, A novel

GLP-1/xenin hybrid peptide improves glucose homeostasis, circulating lipids and

restores GIP sensitivity in high fat fed mice, Peptides 100 (2018) 202-211.

5. M.A. Sánchez-Garrido, S.J. Brandt, C. Clemmensen, T.D. Müller, R.D. DiMarchi,

M.H. Tschöp, GLP-1/glucagon receptor co-agonism for treatment of obesity,

Diabetologia 60 (2017) 1851-1861.

6. H.S. Tager, D.F. Steiner, Isolation of a glucagon-containing peptide: primary

structure of a possible fragment of proglucagon, Proc. Natl. Acad. Sci. USA 70

(1973) 2321-2325.

7. P. Blache, A. Kervran, D. Bataille, Oxyntomodulin and glicentin: brain-gut

peptides in the rat, Endocrinology 123 (1988) 2782-2787.

8. A. Pocai, Unraveling oxyntomodulin, GLP-1's enigmatic brother, J. Endocrinol.

235 (2012) 335-346.

9. J.J. Holst, N.J.W. Albrechtsen, M.B.N. Gabe, M.M. Rosenkilde, Oxyntomodulin:

actions and role in diabetes, Peptides 100 (2018) 48-53.

18

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

10. M.A. Cohen, S.M. Ellis, C.W. Le Roux, R.L. Batterham, A. Park, M. Patterson, et

al., Oxyntomodulin suppresses appetite and reduces food intake in humans, J.

Clin. Endocrinol. Metab. 88 (2003) 4696-4701.

11. V.K. Bhat, B.D. Kerr, P.R. Flatt, V.A. Gault, A novel GIP-oxyntomodulin hybrid

peptide acting through GIP, glucagon and GLP-1 receptors exhibits weight

reducing and anti-diabetic properties, Biochem. Pharmacol. 85 (2013) 1655-1662.

12. J.M. Conlon, N. Hazon, L. Thim, Primary structures of peptides derived from

proglucagon isolated from the pancreas of the elasmobranch fish, Scyliorhinus

canicula, Peptides 15 (1994) 163-167.

13. F.P.M. O’Harte, M.T. Ng, A.M. Lynch, J.M. Conlon, P.R. Flatt, Novel dual

agonist peptide analogues derived from dogfish glucagon show promising in vitro

insulin releasing actions and anti-hyperglycaemic activity in mice, Mol. Cell.

Endocrinol. 431 (2016) 133-144.

14. F.P.M. O’Harte, M.T. Ng, A.M. Lynch, J.M. Conlon, P.R. Flatt, Dogfish

glucagon analogues counter hyperglycaemia and enhance both insulin secretion

and action in diet-induced obese diabetic mice, Diabetes Obes. Metab. 18 (2016)

1013-1024.

15. G.V. Graham, J.M. Conlon, Y.H. Abdel-Wahab, V.A. Gault, P.R. Flatt,

Evaluation of the insulinotropic and glucose-lowering actions of zebrafish GIP in

mammalian systems: evidence for involvement of the GLP-1 receptor, Peptides

100 (2018) 182-189.

16. D.M. Irwin, S. Mojsov, Diversification of the functions of proglucagon and

glucagon receptor genes in fish, Gen. Comp. Endocrinol. 261 (2018) 148-165.

19

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

17. Y. Wang, P.F. Nielsen, J.H. Youson, I.C. Potter, J.M. Conlon, Multiple forms of

glucagon and somatostatin isolated from the intestine of the southern-hemisphere

lamprey Geotria australis, Gen. Comp. Endocrinol. 113 (1999) 274-82.

18. J.M. Conlon, J.H. Youson, T.P. Mommsen, Structure and biological activity of

glucagon and glucagon-like peptide from a primitive bony fish, the bowfin (Amia

calva), Biochem. J. 295 (1993) 857-861.

19. J.M. Conlon, R. Göke, P.C. Andrews, L. Thim, Multiple molecular forms of

insulin and glucagon-like peptide from the Pacific ratfish (Hydrolagus colliei),

Gen. Comp. Endocrinol. 73 (1989) 136-146.

20. K.D. Crow, C.D. Smith, J.F. Cheng, G.P. Wagner, C.T. Amemiya, An

independent genome duplication inferred from Hox paralogs in the American

paddlefish - a representative basal ray-finned fish and important comparative

reference, Genome. Biol. Evol. 4 (2012) 937-953.

21. E.M. Plisetskaya, T.P. Mommsen, Glucagon and glucagon-like peptides in

fishes, Int. Rev. Cytol. 68 (1996) 187-257.

22. J.M. Conlon, P.F. Nielsen, J.H. Youson, Primary structures of glucagon and

glucagon-like peptide isolated from the intestine of the parasitic phase lamprey

Petromyzon marinus. Gen. Comp. Endocrinol. 91 (1993) 96-104.

23. T.M. Nguyen, T.P. Mommsen, S.M. Mims, J. M. Conlon, Characterization of

insulins and proglucagon-derived peptides from a phylogenetically ancient fish,

the paddlefish (Polyodon spathula), Biochem. J. 300 (1994) 339-345.

24. D.M. Irwin, J. Wong, Trout and chicken proglucagon: alternative splicing

generates mRNA transcripts encoding glucagon-like peptide 2, Mol.

Endocrinol. 9 (1995) 267-277.

20

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

25. J.M. Conlon, R.C. Moffett, J. Leprince, P.R. Flatt, Identification of components

in frog skin secretions with therapeutic potential as antidiabetic agents, Methods

Mol. Biol. 719 (2018) 319-333.

26. N.H. McClenaghan, C.R. Barnett, E. Ah-Sing, Y.H.A.Abdel-Wahab, F.P.

O’Harte, T.W. Yoon et al., Characterization of a novel glucose-responsive

insulin-secreting cell line BRIN-BD11 produced by electrofusion, Diabetes 45

(1996) 1132-1140.

27. J.T. McCluskey, M. Hamid, H. Guo-Parke, N.H. McClenaghan, R. Gomis, P.R.

Flatt, Development and functional characterization of insulin-releasing human

pancreatic beta cell lines produced by electrofusion, J. Biol. Chem. 286 (2011)

21982-21992.

28. B.O. Owolabi, O.O. Ojo, D.K. Srinivasan, J.M. Conlon, P.R. Flatt, Y.H. Abdel-

Wahab, In vitro and in vivo insulinotropic properties of the multifunctional frog

skin peptide hymenochirin-1B: a structure-activity study, Amino Acids 48 (2016)

535-544.

29. B. Thorens, A. Porret, L. Buhler, S.P. Deng, P. Morel, C. Widmann, Cloning and

functional expression of the human islet GLP-1 receptor. Demonstration that

exendin-4 is an agonist and exendin-(9-39) an antagonist of the receptor,

Diabetes 42 (1993) 1678-1682.

30. F.P.M. O'Harte, Z.J. Franklin, E.P. Rafferty, N. Irwin, Characterisation of

structurally modified analogues of glucagon as potential glucagon receptor

antagonists, Mol. Cell. Endocrinol. 381 (2013) 26-34.

21

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

31. V. Pathak, V.A. Gault, P.R. Flatt, N. Irwin, Antagonism of gastric inhibitory

polypeptide (GIP) by palmitoylation of GIP analogues with N- and C-terminal

modifications improves obesity and metabolic control in high fat fed mice, Mol.

Cell. Endocrinol. 401 (2015) 120-129.

32. P.R. Flatt, C.J. Bailey, Development of glucose intolerance and impaired plasma

insulin response to glucose in obese hyperglycaemic (ob/ob) mice, Horm. Metab.

Res. 13 (1981) 556-560.

33. M. Goto, T. Maki, T. Kiyoizumi, S. Satomi, A.P.Monaco, An improved method

for isolation of mouse pancreatic islets, Transplantation 40 (1985) 437-438.

34. T. Ikegami, A.M. Cypess, B. Bouscarel, Modulation of glucagon receptor

expression and response in transfected human embryonic kidney cells, Am. J.

Physiol. Cell. Physiol. 281 (2001) 396–1402.

35. J. Naylor, A.T. Suckow, A. Seth, D.J. Baker, I. Sermadiras, P. Ravn, et al., Use

of CRISPR/Cas9-engineered INS-1 pancreatic β cells to define the pharmacology

of dual GIPR/GLP-1R agonists, Biochem. J. 473 (2016) 2881-2891.

36. C.G. Unson, R.B. Merrifield, Identification of an essential serine residue in

glucagon: implication for an active site triad, Proc. Natl. Acad. Sci. USA 91

(1994) 454-458.

37. C.G. Unson, C.R. Wu, C.P. Cheung, R.B. Merrifield, Positively charged residues

at positions 12, 17, and 18 of glucagon ensure maximum biological potency, J.

Biol. Chem. 273 (1998) 10308-10312.

38. B. Gallwitz, M. Witt, G. Paetzold, C. Morys-Wortmann, B. Zimmermann, K.

Eckart, et al. Structure/activity characterization of glucagon-like peptide-1, Eur.

J. Biochem. 225 (1994) 1151-1156.

22

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

39. C.R. Underwood, P. Garibay, L.B. Knudsen, S. Hastrup, G.H. Peters, R. Rudolph

R, et al., Crystal structure of glucagon-like peptide-1 in complex with the

extracellular domain of the glucagon-like peptide-1 receptor, J. Biol. Chem. 285

(2010) 723-730.

40. J. Eng, W.A. Kleinman, L. Singh, G. Singh, J.P. Raufman, Isolation and

characterization of exendin-4, an exendin-3 analogue, from Heloderma

suspectum venom. Further evidence for an exendin receptor on dispersed acini

from guinea pig pancreas, J. Biol. Chem. 267 (1992) 7402-7405.

41. F.K. Knop, A. Brønden, T. Vilsbøll, Exenatide: pharmacokinetics, clinical use,

and future directions, Expert Opin. Pharmacother. 18 (2017) 555-571.

42. A.M. Lynch, N. Pathak, V. Pathak, F.P. O'Harte, P.R. Flatt, N. Irwin et al., A

novel DPP IV-resistant C-terminally extended glucagon analogue exhibits

weight-lowering and diabetes-protective effects in high-fat-fed mice mediated

through glucagon and GLP-1 receptor activation, Diabetologia 57 (2014) 1727-

1736.

43. C.Q. Pan, J.M. Buxton, S.L. Yung, I. Tom, L. Yang, H. Chen et al., Design of a

long acting peptide functioning as both a glucagon-like peptide-1 receptor agonist

and a glucagon receptor antagonist, J. Biol. Chem. 28 (2006) 12506-12515.

44. V. Muñoz, L. Serrano, Elucidating the folding problem of helical peptides using

empirical parameters. Nat. Struct. Biol. 1 (1994) 399-409.

45. M.R. Druce, J.S. Minnion, B.C. Field, S.R. Patel, J.C. Shillito, M. Tilby M, et al.,

Investigation of structure-activity relationships of oxyntomodulin (Oxm) using

Oxm analogs, Endocrinology 150 (2009) 1712-1722.

23

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Table 1. Primary structures and molecular masses of the glucagon-related fish

peptides used in this study

Peptide Amino acid sequence Calculated Observed molecular molecular mass (Da) mass (Da)

Lamprey glucagon HSEGTFTSDYSKYLENKQAKDFVRWLMNA 3466.9 3466.5

Dogfish oxyntomodulin HSEGTFTSDYSKYMDNRRAKDFVQWLMSTKRNG 3957.4 3958.5

Ratfish oxyntomodulin HTDGIFSSDYSKYLDNRRTKDFVQWLLSTKRNGANT 4235.6 4233.6

Paddlefish glucagon HSQGMFTNDYSKYLEEKRAKEFVEWLKNGKS 3751.1 3750.8

Trout glucagon HSEGTFSNDYSKYQLERMAQDFVQWLMNS 3512.9 3512.4

24

1

2

34567

8

9

10

11

12

13

Table 2. Effects of glucagon-related peptides on the release of insulin from BRIN-

BD11 cells.

The values are mean ± S.E.M. for n = 8. *P < 0.001 compared to 5.6 mmol L-1 glucose

control. The values are derived from the data shown in Figure 1. The threshold

concentration is the minimum concentration of the peptide producing a significant (p <

0.05) increase in the rate of insulin release. The rate of insulin release at 3 µm

concentration represent the maximum or near maximum rate.

25

Peptide Threshold concentration

Rate at 3 µM ng 106 cells-1 20 min-1

Control (glucose alone) 1.02 ± 0.02

Lamprey glucagon 3 nmol L-1 3.3 ± 0.2*

Dogfish oxyntomodulin 1 nmol L-1   2.4 ± 0.1*

Ratfish oxyntomodulin 3 nmol L-1 2.3 ± 0.2*

Paddlefish glucagon 30 pmol L-1 3.6 ± 0.2*

Trout glucagon 30 nmol L-1 1.6 ± 0.1*

Human glucagon 10 nmol L-1 2.7 ± 0.1*

Human GLP-1 10 pmol L-1 3.6 ± 0.2*

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

Legends to figures

Figure 1 Concentration-dependence of the effects of (A) lamprey glucagon, (B) dogfish

oxyntomodulin, (C) ratfish oxyntomodulin, (D) paddlefish glucagon (E) trout glucagon

(F) human glucagon and (G) human GLP-1 on the release of insulin from BRIN-BD11

rat clonal β- cells. Values are mean ± S.E.M., n = 8. *P < 0.05, **P < 0.01, ***P <

0.001 compared to 5.6 mmol L-1 glucose alone.

Figure 2 Effects of glucagon-related peptides (10-8 and 10-6 M) on insulin release from

human 1.1 B4 cells. Values are mean ± S.E.M., n = 8. *P < 0.05, *P < 0.01,

***P < 0.001 compared to 16.7 mmol L-1 glucose alone. ΔΔP < 0.01 and ΔΔΔP < 0.001

compared to GLP-1.

Figure 3 Effects of glucagon-related peptides (10 nmol L-1 and 1 µmol L-1) on insulin

release from pancreatic islets isolated from NIH Swiss mice. The values are mean ±

S.E.M. for n = 4; *P < 0.05, **P < 0.01, ***P < 0.001 compared to 16.7 mmol L-1

glucose alone. ΔP < 0.05. ΔΔP < 0.01 and ΔΔΔP < 0.001 compared to GLP-1.

Figure 4 Effects of (A) the GLP-1 receptor antagonist, exendin-4(9-39), (B) the

glucagon receptor antagonist [desHis1,Pro4,Glu9]glucagon amide and (C) the GIP

receptor antagonist GIP(6-30)Cex-K40[Pal], on the ability of glucagon-related peptides

(0.1 µmol/L) to stimulate insulin release from BRIN-BD11 cells. Values are mean ±

S.E.M., n = 8 **P < 0.01, ***P < 0.001 compared with 5.6 mmol L-1 glucose alone.

ΔΔP < 0.01, ΔΔΔP < 0.001 compared with the effect in the presence of antagonist.

26

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Figure 5. Effects of glucagon-related peptides and GIP (10 nmol L-1 and 1 µmol L-1)

on cAMP production in (A) GLP1R-transfected CHL cells, and (B) GCGR-transfected

HEK293 cells. Values are mean ± S.E.M. for n = 4. *P < 0.05, ** P< 0.01 and ***p <

0.001 compared with 5.6 mmol L-1 glucose. In panel A, ΔP < 0.05, ΔΔΔP < 0.001

compared with the effect of GLP-1. In panel B, ΔΔP < 0.01, ΔΔΔP < 0.001 compared with

the effect of glucagon.

Figure 6. Effects of glucagon-related peptides (10 nmol L-1 and 1 µmol L-1) on the rate

of insulin release from (A) wild-type INS-1 cells, (B) CRISPR/Cas9-engineered GLP-

1R knock-out cells, and (C) CRISPR/Cas9-engineered GIPR knock-out cells. Values

are mean ± S.E.M., n = 8, *P < 0.05, **P < 0.01 and ***P < 0.001 compared with 5.6

mmol L-1 glucose alone. ΔP < 0.05, ΔΔP < 0.01, ΔΔΔP < 0.001 compared with effects in

wild-type INS-1 cells.

Figure 7. Effects of acute administration of glucagon-related peptides (25 nmol kg

body weight-1) on blood glucose (panels A and C) and plasma insulin (panels B and D)

concentrations in mice after intraperitoneal injection of glucose (18 mmol kg body

weight-1). The integrated responses are shown in panels E and F. The values are mean ±

S.E.M., n = 6. *P < 0.05, **P < 0.01 and ***P < 0.001 compared with glucose alone

and Δ P < 0.05, and ΔΔP < 0.01 compared with GLP-1.

27

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Figure 8 A comparison of the primary structures of the glucagon-related fish peptides

with those of human glucagon, oxyntomodulin and GLP-1 and with the GLP1R

agonist, exendin-4. Amino acid residues in the human peptides that have been shown to

be important in receptor binding and signal transduction are shown in red. Differences

in these residues are shown in green. Regions of structural similarity between

paddlefish glucagon and exendin-4 are underlined. The values in parentheses show %

sequence identity with the corresponding human peptides.

28

1

2

3

4

5

6

7

8

Fig. 1

29

1

2

3

4

5

6

Fig.2

30

1

2

3

4

5

6

7

8

9

Fig. 3

31

1

2

3

4

5

6

7

8

9

10

Fig. 4

32

1

2

3

Fig. 5

33

1

2

Fig. 6

34

1

2

3

4

5

Fig. 7

35

1

2

3

4

5

67

Glucagon HSQGTFTSDY10SKYLDSRRAQ20DFVQWLMNTOxyntomodulin HSQGTFTSDY SKYLDSRRAQ DFVQWLMNTKRNKNNIA Lamprey glucagon HSEGTFTSDY SKYLENKQAK DFVRWLMNA (72)Dogfish oxyntomodulin HSEGTFTSDY SKYMDNRRAK DFVQWLMSTKRNG (83) Ratfish oxyntomodulin HTDGIFSSDY SKYLDNRRTK DFVQWLLSTKRNGANT (69) Paddlefish glucagon HSQGMFTNDY SKYLEEKRAK EFVEWLKNGKS (66)Trout glucagon HSEGTFSNDY SKYQLERMAQ DFVQWLMNS (72)

GLP-1 HAEGTFTSDV10SSYLEGQAAK20EFIAWLVKGR30G Exendin-4 HGEGTFTSDL SKQMEEEAVR LFIEWLKNGGPSSGAPPPSLamprey glucagon HSEGTFTSDY SKYLENKQAK DFVRWLMNA (57) Dogfish oxyntomodulin HSEGTFTSDY SKYMDNRRAK DFVQWLMSTKRNG (50)Ratfish oxyntomodulin HTDGIFSSDY SKYLDNRRTK DFVQWLLSTKRNGANT (40) Paddlefish glucagon HSQGMFTNDY SKYLEEKRAK EFVEWLKNGKS (57)Trout glucagon HSEGTFSNDY SKYQLERMAQ DFVQWLMNS (40)

Fig. 8

36

123456789

1011121314151617

18