Purification, conformational analysis, and properties of a family of

tigerinin peptides from skin secretions of the crowned bullfrog

Hoplobatrachus occipitalis (Dicroglossidae)

Christopher M. McLaughlin†#, Sandrina Lampis‡#, Milena Mechkarska†∞, Laurent

Coquet§, Thierry Jouenne§, Jay D. King⊥, Maria Luisa Mangoni║, Miodrag L. LukicΔ,

Mariano A. Scorciapino‡, J. Michael Conlon†*

†SAAD Centre for Pharmacy and Diabetes, School of Biomedical Sciences,

University of Ulster, Coleraine, U.K.

‡Department of Chemical and Geological Sciences and Department of Biomedical

Sciences - Biochemistry Unit, University of Cagliari, Italy

§CNRS UMR 6270, PISSARO, University of Rouen, Institute for Research and

Innovation in Biomedicine (IRIB), Mont-Saint-Aignan, France 

⊥Rare Species Conservatory Foundation, St. Louis, MO, USA

║Instituto Pasteur-Fondazione Cenci Bolognetti, Department of Biochemical Sciences,

Sapienza University of Rome, Rome, Italy

ΔCenter for Molecular Medicine, Faculty of Medicine, University of Kragujevac,

Kragujevac, Serbia

*Corresponding author. J. E-mail address: m.conlon@ulster.ac.uk; Tel: +44-28-

70832917; Fax: +44-28-70124965

1

AUTHOR INFORMATION

Corresponding Author

* Tel: +44-7918526277. Fax: +44-2870124965; E-mail: m.conlon@ulster.ac.uk

# These authors contributed equally to the article

∞ Present address: Department of Life Science, The University of the West Indies, St

Augustine, Trinidad and Tobago

2

ABSTRACT

Four host-defense peptides belonging to the tigerinin family (tigerinin-1O:

RICTPIPFPMCY; tigerinin-2O: RTCIPIPLVMC; tigerinin-3O: RICTAIPLPMCL;

and tigerinin-4O: RTCIPIPPVCF) were isolated from skin secretion of the African

crowned bullfrog Hoplobatrachus occipitalis (Dicroglossidae). In aqueous solution at

pH 4.8, the cyclic domain of tigerinin-2O adopts a rigid amphipathic conformation

that incorporates a flexible N-terminal tail. The tigerinins lacked antimicrobial (MIC

> 100 µM) and hemolytic (LC50 > 500 µM) activities but, at a concentration of 20

µg/mL, significantly (P < 0.05) inhibited production of interferon-γ (IFN-γ) by

peritoneal cells from C57BL/6 mice without affecting production of IL-10 and IL-17.

Tigerinin-2O and -4O inhibited IFN-γ production at concentrations as low as 1

µg/mL. The tigerinins significantly (P ≤ 0.05) stimulated the rate of insulin release

from BRIN-BD11 clonal β-cells without compromising the integrity of the plasma

membrane. Tigerinin-1O was the most potent (threshold concentration 1 nM) and the

most effective (395% increase over basal rate at a concentration of 1 µM). Tigerinin-

4O was the most potent and effective peptide in stimulating the rate of glucagon-like

peptide-1 release from GLUTag enteroendocrine cells (threshold concentration 10

nM; 289% increase over basal rate at 1 µM). Tigerinin peptides have potential for

development into agents for the treatment of patients with Type 2 diabetes.

3

Peptidomic analysis of frog skin secretions has proved to be valuable in identifying

components with therapeutic potential for development into antimicrobial agents for

use against drug-resistant bacteria and fungi as well as those with anti-cancer and

anti-viral properties. In addition, peptides that were first identified on the basis of

their cytotoxic activities have subsequently been shown to display

immunomodulatory and anti-diabetic activities.1,2 Comparisons of the primary

structures of the host-defense peptides in skin secretions has provided insight into the

evolutionary history of species within particular families, such as the Ranidae3,

Leptodactylidae4, and the Pipidae5 and may be used to complement data derived from

the nucleotide sequences of mitochondrial and/or nuclear genes in elucidating

phylogenetic relationships.

Tigerinin-1R, first isolated from an extract of the skin of the Vietnamese

common lowland frog Hoplobatrachus rugulosus6 in the family Dicroglossidae, is one

such amphibian peptide that has excited interest as a lead compound for drug

development in the treatment of patients with Type 2 diabetes. Largely devoid of

antimicrobial and cytotoxic properties, tigerinin-1R6 and its analogs7,8 stimulate

insulin release from BRIN-BD11 clonal β-cells and the release of glucagon-like

peptide-1 (GLP-1) from the GLUTag cell line9 at concentrations that are not toxic to

the cells. Studies in vivo have shown that tigerinin-1R10 and its analogs8,11 enhance

both insulin sensitivity and pancreatic β-cell function and decrease adiposity and

plasma triglycerides when administered to mice given a high fat diet to produce

obesity and insulin resistance. In addition, tigerinin-1R increases production of the

anti-inflammatory cytokine interleukin-10 (IL-10) in both spleen cells from C57BL/6

mice and human peripheral blood mononuclear cells without stimulating production

4

of the pro-inflammatory cytokines interleukin-12 (IL-12) and interleukin-23 (IL-23)12

suggesting a possible role in treatment of sepsis.

The African crowned bullfrog Hoplobatrachus occipitalis (Günther, 1858),

also known as the Eastern groove-crowned bullfrog, occupies broad but disjunctive

ranges in the African savannah zone with a swath in North Africa that includes

portions of Morocco, Algeria and Libya and a much larger range in sub-Saharan

Africa extending from the Atlantic coast of West Africa eastward to Ethiopia, Chad,

Eritrea, Sudan and south to Angola and Mozambique.13,14 It is tolerant of a wide

variety of habitats and is common over much of its range. It is listed as a Species of

Least Concern by the International Union for Conservation of Nature (IUCN) Red

List although populations have been depleted in certain areas by loss of habitat and

human consumption.15 The present study extends the previous work with tigerinin-1R

by describing the isolation, structural characterization, conformational analysis, and

biological activities of peptides belonging to tigerinin family that are present in

norepinephrine-stimulated skin secretions of H. occipitalis.

RESULTS AND DISCUSSION

Purification and characterization of the peptides. The pooled skin

secretions from H. occipitalis, after partial purification on Sep-Pak C-18 cartridges,16

were chromatographed on a Vydac C-18 preparative reversed-phase HPLC column

(Figure 1). The well-resolved peaks designated 1- 4 were purified to near

homogeneity, as assessed by a symmetrical peak shape and mass spectrometry, by

further chromatography on semipreparative Vydac C-4 and Vydac C-8 columns.

Subsequent structural analysis revealed that peak 1 contained tigerinin-1O (125), peak

5

2 tigerinin-2O (60), peak 3 tigerinin-3O (55), and peak 4 tigerinin-4O (10). The

values in parentheses show the approximate yields of the purified peptides in nmol.

Figure 1. Reversed-phase HPLC on a preparative Vydac C-18 column of skin

secretions from H. occipitalis after partial purification on Sep-Pak cartridges. The

peaks designated 1 - 4 contained tigerinins and were purified further. The dashed line

shows the concentration of acetonitrile in the eluting solvent.

The amino acid sequences of the tigerinin peptides were established by

automated Edman degradation and their primary structures are shown in Figure 2. The

molecular masses of the peptides, determined by MALDI-TOF mass spectrometry,

were consistent with the proposed structures and demonstrate that the tigerinins were

isolated in the oxidized form with a cysteine bridge and are not C-terminally α-

amidated.

6

[M+H]obs [M+H]calc pI H

Tigerinin-1O RICTPIPFPMCY 1438.7 1438.7 8.23 7.4

Tigerinin-2O RTCIPIPLVMC 1243.7 1243.6 8.23 15.5

Tigerinin-3O RICTAIPLPMCL 1328.7 1328.7 8.23 16.9

Tigerinin-4O RTCIPIPPVCF 2486.4 2486.3* 8.23 11.0

*calculated as a dimer

Figure 2. Primary structure, observed molecular mass ([M+H]obs), calculated

molecular mass ([M+H]calc), isoelectric point (pI) and hydrophobicity (H) of the

peptides isolated from H. occipitalis skin secretions. H is calculated using the

hydrophobicity scales for amino acids of Kyte and Doolittle.17

The tigerinin family of peptides was first identified in the skin of Indian frog

Hoplobatrachus tigerinus (formerly classified as Rana tigerina) in the family

Dicroglossidae18 and subsequently in the Asian frogs Hoplobatrachus rugulosus6 and

Fejervarya cancrivora19 in the same family. Fejervarya is regarded as sister taxon to a

clade containing Hoplobatrachus + Euphlyctis.20 As shown in Figure 3, small cyclic

peptides with limited structural similarity to the tigerinins have been isolated from

skin secretions of the Costa Rican frog, Lithobates vaillanti in the family Ranidae21

and the East-African Mueller’s clawed frog Xenopus muelleri in the family Pipidae22

but their evolutionary relationship to tigerinins isolated from species belonging to the

7

Dicroglossidae is unknown. This study has led to the purification of four peptides in

skin secretions of H. occipitalis whose primary structures identify them as members

of the tigerinin family. In common with the vast majority of frog skin host-defense

peptides, they are cationic (isolectric point pI = 8.23) and appreciably hydrophobic

(Figure 2). Mass spectrometry indicated that tigerinin-4O was isolated as a dimer

(observed molecular mass 2486.4 Daltons) but the synthetic replicate behaved as a

monomer (observed molecular mass 1243.3 Daltons). The reason for the anomalous

behaviour of the naturally occurring peptide is unclear but, as one possibility, an

intramolecular disulphide bond may have been formed at some stage during the

biosynthetic pathway.

H. occipitalis 1 RICTPIPFPM*CY

H. occipitalis 2 RTCIPIPLVM*C

H. occipitalis 3 RICTAIPLPM*CL

H. occipitalis 4 RTCIPIP*PV*CF

H. tigerinus 1 FCTMIPIPR*CYa

H. tigerinus 2 RVCFAIPLPI*CHa

H. tigerinus 3 RVCYAIPLPI*CYa

H. tigerinus 4 RVCYAIPLPI*Ca

F. cancrivora 1 RVCSAIPLPI*CH

F. cancrivora 2 RVCMAIPLPL*CH

H. rugulosus RVCSAIPLPI*CHa

L. vaillanti RICYAMWIPYPC

X. muelleri WCPPM*IPL*CSRFa

8

Figure 3. A comparison of the primary structures of tigerinin peptides from the

Asian frogs from the Dicroglossidae (H. tigerinus, H. rugulosus, and F. cancrivora )

with those from the African frog H. occipitalis. The primary structure of peptides with

limited sequence similarity to the tigerinins from L. vaillanti (Ranidae) and X.

muelleri (Pipidae) are also shown. a indicates that the peptide is C-terminally -

amidated. Shading is used to indicate conserved amino acid residues. Gaps, denoted

by *, are inserted into some sequences to maximize structural similarity.

The genus Hoplobatrachus within the extensive subfamily Dicroglossinae (168

species) of the family Dicroglossidae comprises five species. H. crassus, H. litoralis,

H. rugulosus and H. tigerinus are distributed in Asia with H. occipitalis being the

only African species.14 Firm molecular evidence for monophyly of the Asian

Hoplobatrachus has been provided but the position of H. occipitalis was not fully

clarified.23. However, the Asian and African species share many morphological and

osteological features both in the adult frogs and the tadpoles.24 Mitochondrial DNA

analysis has suggested the “out of Asia” hypothesis that the genus has an oriental

origin with H. occipitalis migrating to Africa in the Miocene (between 8 and 25

million years ago).24 The presence of tigerinin peptides with appreciable structural

similarity to those from the Asian species in H. occipitalis skin secretions supports the

proposal that all five species within the Hoplobatrachus share a common origin.

Conformational Analysis. 1H and 13C resonance assignments were obtained

through the analysis of a series of 2D spectra (DQF-COSY, TOCSY, NOESY, and 1H

- 13C-HSQC)25 collected from an aqueous solution of tigerinin-2O at pH 4.8 (Table

9

S1). The presence of the disulfide bridge (Cys3–Cys11) was confirmed by the

characteristic chemical shift of the cysteine residues’ C, reflecting their oxidized

state26. Measured chemical shift values of 1HN, 1Hα, 1Hβ, 13Cα and 13Cβ for all the

residues were analyzed with TALOS+ software27, which provides predicted and

backbone torsion angles by comparing the experimental chemical shift values to its

reference protein structures dataset. Only predictions with a high consensus were used

for structure calculations. Given the short sequence of tigerinin-2O, no specific

secondary structure was assigned by TALOS+. Nevertheless, all the amino acids were

predicted to fall within the allowed regions of the Ramachandran plot (Figure 4a)

except for the N- and C-terminal residues. The analysis of NOESY spectra provided

evidence for inter-proton through-space dipolar interactions. On the basis of the

relative intensity observed for the corresponding cross-peak, inter-proton distances

were estimated and used as restraints within peptide structure calculations. The

DYNAMO simulated annealing scheme was applied to finally obtain 300 structures

compatible with both backbone torsions and inter-proton distances derived from the

NMR experiments. The 30 conformers with the lowest potential energy were then

extracted and analyzed. As shown in Figure 4b, a rather rigid and stretched rim-like

structure, whose consistency is confirmed by a root mean square deviation (RMSD)

of just 0.96 ± 0.38 Å is indicated. Figure 4c represents a side view of the same 30

conformers in which side chain residues are shown for completeness and the bended

conformation of the peptide backbone appears clear. Heavy atoms RMSD was 2.309

± 0.705 Å. The orientation of all the residues except Arg1 appear consistent among the

conformers analyzed and a RMSD as low as 1.352 ± 0.310 Å was obtained by

neglecting the first residue. Thus, while the ring portion of the peptide is somewhat

rigid, it is endowed with a flexible, hydrophilic N-terminal tail. Tigerinin-2O is highly

10

amphipathic with complete segregation of the hydrophobic and hydrophilic residues

(Figure 4d). Figure 5 shows schematically the tigerinin-2O sequence with the

disulfide bridge and the experimental NOE connectivities determined in this work.

Figure 4. (a) Ramachandran plot for tigerinin-2O. The values of the backbone

torsional angles determined with TALOS+ are shown together with the corresponding

uncertainty. Only the high-consensus predictions are shown. Shaded areas represent

11

the energetically most favoured regions. (b) Superposition of the backbone trace of

the 30 conformers with the lowest potential energy obtained from the DYANMO

simulated annealing scheme out of a total of 300 structures calculated. (c) Side view

of the same 30 conformers. Residues side chain are shown in addition to the backbone

trace. (d) Representative 3D structure of tigerinin-2O showing the segregation of

hydrophilic (blue) and hydrophobic (yellow) residues. Atoms are represented as

spheres with the radius proportional to the corresponding Van der Waals radius.

Figure 5. Schematic representation of the tigerinin-2O sequence illustrating NOEs

connectivity. In particular, inter-residue HN-HN and HN-H dipolar interactions are

shown.

Cytotoxic activities. No tigerinin peptide showed significant hemolytic activity

against mouse erythrocytes at concentrations up to 500 μM or growth inhibitory

activity against the Gram-positive bacterium Bacillus megaterium and the Gram-

negative bacterium Escherichia coli at concentrations up to 100 μM. The tigerinin

peptides from H. tigerinus28 and F. cancrivora19 are reported to possess broad

12

spectrum antimicrobial activity whereas tigerinin-1R is not active against Gram-

positive Staphylococcus aureus or Gram-negative E. coli at concentrations up to 500

µM.7 The reason for this discrepancy is unclear but may be a consequence of the

different methodologies used to assay for antimicrobial activity. The lack of

cytotoxicity of the tigerinin-O peptides is consistent with the report of Srinivasan et

al.7 demonstrating a similar lack of activity for tigerinin-1R.

Insulin and GLP-1 release. The basal rate of insulin release from BRIN-BD11

glucose-responsive clonal β-cells29 in the presence of 5.6 mM glucose alone was 1.00

± 0.03 ng/106 cells/20 min. The rate of insulin release increased to 3.05 ± 0.05 ng/106

cells/20 min (P < 0.001) by incubation with 10 mM alanine, to 3.14 ± 0.09 ng/106

cells/20 min (P < 0.001) by incubation with 1 µM GLP-1, and to 4.05 ± 0.17 ng/106

cells/20 min (P < 0.001) by incubation with 30 mM KCl. In common with tigerinin-

1R from H. rugulosus and the more potent analogs [I10W]tigerinin-1R and [S4R]

tigerinin-1R7, the tigerinins from H. occipitalis produced a concentration-dependent

stimulation of insulin release from BRIN-BD11 clonal β-cells. As shown in Figure

S1, all tigerinins produced a significant (P ≤ 0.01) increase in the rate of insulin

release at the maximum concentration tested (1 µM). Tigerinin-1O was the most

potent peptide producing an 83% increase (P < 0.001) in the rate of insulin release at a

threshold concentration (minimum concentration producing a significant increase in

the rate of insulin release over the basal rate in the presence of 5.6 mM glucose only)

of 1 nM and was also the most effective producing a 395 % increase (P < 0.001) in

the rate of insulin release at a concentration of 1 µM (Table 1). The potency and

effectiveness of tigerinin-1O was comparable to that of physiologically important

incretin peptide GLP-1 under the same experimental conditions. No peptide, at

concentrations up to 1 µM, produced a significant increase in the rate of release of the

13

cytosolic enzyme LDH indicating that the integrity of the plasma membrane had been

preserved (data not shown).

The mechanism of insulin-releasing action of the H. occipitalis tigerinins was

not addressed in this study but patch-clamp studies have shown that [S4R]tigerinin-

1R blocks KATP channels in BRIN-BD11 cells and the resulting depolarization

indirectly increases the activity of the L-type Ca channels leading to increase Ca2+

influx and consequent increase in the rate of insulin secretion.8

GLUTag cells are a stable and relatively well differentiated murine

enteroendocrine cell line that release glucagon-like peptide-1 (GLP-1) in a regulated

manner in response to a range of physiological and pharmacological stimulatory

agents.30 As shown in Table 1, tigerinin-20, -30, and -40 produced a significant (P ≤

0.05) increase in the rate of GLP release with tigerinin-4O being the most potent and

effective peptide producing an 289% increase in the rate of GLP-1 release at 10 nM.

The effect is similar in magnitude to that produced by 10 mM glutamine (Figure S2).30

No peptide, at concentrations up to 1 µM, produced a significant increase in the rate

of release of LDH (data not shown). The ability of the tigerinins to stimulate GLP-1

release from GLUTag cells suggests that, as well as a direct stimulatory effect on

insulin release by the β-cells of pancreas, administration of the tigerinins in vivo may

stimulate the L-cells of the intestine to release the potent endogenous incretin GLP-1.

Consequently, the data provide further support for the contention that naturally

occurring tigerinin peptides may serve as templates for the design of non-toxic and

long-acting analogs for use in the treatment of patients with Type 2 diabetes.1,8,11

14

Table 1. Effect of tigerinin peptides on the rate of insulin from BRIN-BD11 clonal β-

cells and GLP-1 release from GLUTag enteroendocrine cells

INSULIN RELEASE GLP-1 RELEASEPeptide Threshhold

concentration (M)

% of basal release at 10-6 M

Threshhold concentration (M)

% of basal release at 10-6 M

Tigerinin-1O 10-9 395 ± 12*** ND 159 ± 11 NS

Tigerinin-2O 10-7 201 ± 7*** 10-7 193 ± 17.5*

Tigerinin-3O 10-7 233 ± 17** 10-6 217.5 ± 13.3*

Tigerinin-4O 10-6 163 ± 13** 10-8 289.3 ± 32**

Basal release refers to the rate of insulin and GLP-1 release in the absence of peptide

and is set at 100%. The threshold concentration is the minimum concentration of

peptide producing a significant (P < 0.05) increase in the rate of release. Values show

mean ± SEM (n = 8). * P < 0.05, **P < 0.01, ***P < 0.001. ND: not determined, NS: not

significant.

Cytokine production. In the first series of experiments, all four tigerinins at

a concentration of 20 µg/mL produced a significant (P < 0.05) decrease in the

production and release of interferon-γ (IFN-γ) by peritoneal cells from C57BL/6 mice

(Figure 6A). Effects upon the production of IL-10 and interleukin-17 (IL-17) were

not significant (data not shown). In the second series designed to determine the dose-

dependency of the inhibitory effect, concentrations of tigerinin-2O and -4O as low as

15

1.0 µg/mL produced a significant (P < 0.05) inhibition of IFN-γ production (Figure

6B). Tigerinin-1O and -3O did not produce significant inhibition of IFN-γ production

at concentrations below 20 µg/mL.

Figure 6. Effects on the production of IFN-γ by unstimulated peritoneal cells from

C57BL/6 mice by (A) 20 µg/mL of tigerinin-1O, -2O, -3O, and -4O and by (B) 1, 5,

and 10 µg/mL of tigerinin-2O and -4O. * P < 0.05 compared to production in medium

only.

The effects of naturally occurring frog skin peptides on production of pro-

inflammatory and anti-inflammatory cytokines by mouse cells are complex with both

inhibitory and stimulatory actions being reported.1,12 In addition, the effects of the

peptides are dependent on the genetic background of the inbred strain of the mouse

16

from which the cells are derived.12 C57BL/6 mice are regarded as prototypical Th1-

biased and Th1-type cytokines are important in producing the pro-inflammatory

responses responsible for killing pathogenic bacteria. IFN-γ is considered to be a

major Th1 cytokine which, along with the Th17 cytokine, IL-17 is responsible for the

inflammation underlying conditions such as Crohn's disease.32 A previous study has

shown that the tigerinin-like peptides, tigerinin-1M (5 μg/mL) and tigerinin-1V (5

μg/mL) downregulated the production of IFN-γ in spleen cells from C57BL/6 mice

while the production of pro-inflammatory IL-17 was not affected by either tigerinin.12

Consistent, with these data, tigerinin-2O and -4O are particular potent in inhibiting

IFN-γ in mononuclear cells from these mice but do not affect IL-17 production.

However, in contrast to the effect of tigerinin-1R on IL-10 production by spleen cells,

effects of the four tigerinin-O peptides on IL-10 production were not significant. The

The role of the tigerinin-O peptides in the frog’s host defense strategy and in

mediating skin functions in general and the physiological importance of the inhibitory

effect on IFN-γ production in particular is unclear. Excessive cytokine-mediated pro-

inflammatory responses can lead to uncontrolled tissue damage so that the tigerinins

may act to attenuate the frog’s Th1 cytokine response to invasion by pathogenic

microorganisms in the environment.

EXPERIMENTAL SECTION

General Experimental Procedures. All experiments with live animals were

approved by the Animal Research Ethics committee of U.A.E. University (Protocol

No. A21-09) and were carried out by authorized investigators. H. occipitalis frogs (n

= 2; male 165 g; female 250 g) were collected at an undetermined site in the Republic

of Benin, West Africa by a United States Fish and Wildlife Service-approved

17

importer for use in the pet trade. The animals were purchased at a pet store in the St

Louis area. The frogs were injected via the dorsal lymph sac with norepinephrine

hydrochloride (40 nmol/g body mass) and placed in a solution (100 mL) of distilled

water for 15 min. The frogs were removed and the collection solution was acidified

by addition of trifluoroacetic acid (TFA) (1 mL) and immediately frozen. The

solutions containing the secretions from both frogs were pooled and partially purified

on 6 Sep-Pak C-18 cartridges (Waters Associates, Milford, MA, USA) connected in

series as previously described.3

The pooled skin secretions, after partial purification on Sep-Pak cartridges,

were injected onto a (2.2 cm x 25 cm) Vydac 218TP1022 (C-18) reversed-phase

HPLC column (Grace, Deerfield, IL, USA) equilibrated with 0.1% (v/v) TFA/water at

a flow rate of 6.0 mL/min. The concentration of acetonitrile in the eluting solvent was

raised to 21% (v/v) over 10 min and to 63% (v/v) over 60 min using linear gradients.

Absorbance was monitored at 214 nm and 280 nm, and fractions (1 min) were

collected. The peptides were purified to near homogeneity by successive

chromatographies on a (1.0 cm x 25 cm) Vydac 214TP510 (C-4) column and a (1.0

cm x 25 cm) Vydac 208TP510 (C-8) column. The concentration of acetonitrile in the

eluting solvent was raised from 21% to 56% over 50 min and the flow rate was 2.0

mL/min.

MALDI-TOF mass spectrometry was carried out using a Voyager DE-PRO

instrument (Applied Biosystems, Foster City, CA, USA) as previously described.3

The accuracy of mass determinations was < 0.02%. The primary structures of the

peptides were determined by automated Edman degradation using an Applied

Biosystems model 494 Procise sequenator.

18

The cyclic (disulphide-bridged) forms of each tigerinin peptide were supplied

in crude form by Ontores Biotechnologies Co., Ltd (Zhejiang, China). The peptides

were purified by reversed-phase HPLC using the same conditions of chromatography

used to isolate the naturally occurring peptides. Identities of the peptides were

confirmed by electrospray mass spectrometry and the purity of all peptides tested was

>98%.

Conformational Analysis of tigerinin-2O. NMR spectra were acquired at

300 K with a Varian Unity INOVA 500 high-resolution spectrometer operating at a

1H frequency of 500 MHz as previously described.33 Tigerinin-2O was dissolved in

700 μL of 10% D2O at a final concentration of 1.97 mM. The pH was adjusted to 4.8

by adding small aliquots of HCl 0.1 M. The chemical shift scale of both 1H and 13C

were referred to the methyl signal of deuterated 3-(trimethylsilyl)-2,2′,3,3′-

tetradeuteropropionic acid, added as internal reference at a concentration of 2 mM. 1H

spectra were acquired using a 6.1 s pulse (90°), 1 s delay time, 1.5 s acquisition time,

and a spectral width of 6.0 kHz. 2D experiments (1H - 1H DQF-COSY, 1H - 1H

TOCSY, and 1H - 1H NOESY) were recorded over the same spectral window using

2048 complex points and sampling each of the 512 increments with 64 scans. Mixing

times of 80 and 260 ms were applied for the TOCSY and NOESY experiments,

respectively. The 1H - 13C HSQC spectra were recorded using a spectral window of

6.0 kHz for 1H and 26 kHz for 13C. Apart from the HSQC, in all the other experiments

a WET suppression scheme (uburp shape centered at the water resonance with a width

of 100 Hz) was applied to reduce the strong resonance from water. 34,35. TALOS+

software was applied to analyse 1Hα, 1Hβ, 13Cα, and 13Cβ chemical shift values.27 The

software compares the experimental values with its high-resolution structural database

and provides statistical estimates of both Φ and Ψ backbone angles. Only the

19

predictions ranked as “good” were used as restraints for structure calculation. The 3D

structure of tigerinin-2O was obtained using a simulated annealing protocol using the

Dynamo software (http://spin.niddk.nih.gov/NMRPipe/dynamo/). Unambiguous

NOEs and backbone angles from TALOS+ were used as interproton distances and

torsional angle restraints, respectively. In particular, NOEs were classified as strong,

medium, and weak on the basis of the relative intensity of the cross-peaks in the

NOESY spectra, and upper limits of 0.27, 0.33, and 0.50 nm, respectively, applied.

The potential energy contribution was zero below the upper limit, while a harmonic

potential was applied above. 300 structures were calculated, and the 30 conformers

with the lowest potential energy were selected for the analysis. Solvent molecules

were not included in the calculations.

Insulin and GLP-1 release. BRIN-BD11 cells, maintained in culture as

previously described7were seeded into 24-well plates and allowed to attach during

overnight incubation at 37 °C. Incubations with purified synthetic tigerinins (10-12 to

10−6 M; n = 8) were carried out for 20 min at 37 °C in Krebs-Ringer bicarbonate

(KRB) buffer supplemented with 5.6 mM glucose as previously described.7 After

incubation, aliquots of cell supernatant were removed for insulin radioimmunoassay36.

Control incubations were carried out in parallel with the well-established insulin

stimulatory agents 10mM alanine, 1µM GLP-1, and 30 mM KCl.

GLUTag cells were maintained in culture as previously described (Ojo et al.,

2013) and seeded at a density of 105 cells per well. Peptides (10-10 -10-6 M) were

incubated with cells for 2 h at 37 oC in KRB buffer supplemented with 2 mM glucose.

Control incubations were carried out in parallel with 20 mM glucose and with the

well-established GLP-1 stimulatory agents, glutamine (10mM) and forskolin (5

mM).30 GLP-1 concentrations were measured by ELISA using an assay kit supplied

20

by Abcam (Cambridge, MA, USA) according to the manufacturer’s recommended

protocol.

In order to investigate the effects of the peptides on the integrity of the plasma

membrane, BRIN-BD11 and GLUTag cells were incubated with peptides (1 nM -1

μM); n = 3) for 20 min at 37 °C using KRB buffer supplemented with 5.6 mM

glucose as previously described.7 Lactate dehydrogenase (LDH) concentrations in the

cell supernatants were measured using a CytoTox96 nonradioactive cytotoxicity assay

kit (Promega, Madison, WI, USA) according to the manufacturer’s protocol.

In order to determine hemolytic activities, peptides in the concentration range 31 -

500 μM were incubated for 60 min at 37 oC with washed erythrocytes (2 x 107 cells)

taken from male NIH Swiss mice (Harlan Ltd, Bicester, UK) as previously

described.33 The LC50 value was taken as the mean concentration of peptide

producing 50% hemolysis in three independent incubations. Activities of the

tigerinins (0.39 - 100 μM) against the Gram-positive bacterium B. megaterium BM11

and the Gram-negative bacterium E. coli ATCC 25922 were evaluated by a standard

microbroth dilution method according to the Clinical and Laboratory Standards

Institute recommended protocol.37 Minimum inhibitory concentration (MIC)

producing 100% inhibition of microbial growth was determined as previously

described.33

Cytokine production. All of the animal procedures were subjected to review and

approval by the Ethics Committee of the Medical Faculty, University of

Kragujevac which also complies with the National Institutes of Health (NIH)

guidelines for humane treatment of laboratory animals. Experiments were performed

on cells collected from the peritoneal cavity of unstimulated adult C57BL/6 mice

under sterile conditions using 5 ml of cold phosphate-buffered saline according to the

21

procedure of Ray and Dittel.38 After supplementing with 5% (v/v) fetal bovine serum

(FBS; Invitrogen, Carlsbad, CA, USA), the cell suspension was centrifuged (1500

rpm for 5 min). The supernatant was discarded and the cells were resuspended in

supplemented cell culture medium (RPMI 1640 containing 10% (v/v) FBS) for further

analyses.

Isolated cells, representing a mixed population of macrophages, B-, T-, and NK

cells, were suspended in RPMI 1640 culture medium containing 10% fetal bovine

serum. In the first series of experiments, peptides (20 μg/mL equivalent to

approximately 14 μM) were incubated with unstimulated cells (2 x 105 cells/well) for

24h at 37 °C in three independent incubations with 6 mice per group. After

incubation, cell-free supernatants were collected and kept at -20°C until time of

analysis. In a second series, peritoneal cells were stimulated with 1, 5, and 10 µg/mL

of tigerinin-1O, -2O, -3O, and -4O for 24 h on 37°C in three independent incubations

with 6 mice per group. Concentrations of IFN- γ, IL-10, and IL-17 were determined in

triplicate using ELISA assay kits from R & D Systems (Minneapolis, MN, USA)

according to manufacturer’s recommended protocols.

Statistical analysis. Statistical analyses were performed using commercially

available GraphPad Prism software version 5.01 (La Jolla, CA). Results are expressed

as mean ± standard error of mean (SEM), and values were compared using two-way

analysis of variance followed by Newman-Keuls post-hoc test. Groups of data were

considered to be significantly different if P < 0.05.

ASSOCIATED CONTENT

Supporting Information

Table S1 with 1H and 13C resonance assignments.

22

Figure S1 showing concentration-dependence of insulin release from BRIN-BD11

cells

Figure S2 showing concentration-dependence of GLP-1 release from GLUTag cells.

The Supporting Information is available free of charge at the ACS Publications

website at DOI.

Notes

Conflict of interest statement

The authors declare no competing financial interest.

ACKNOWLEDGEMENTS

The authors thank Manju Prajeep for technical assistance in the project and Prof.

P.R.Flatt, Ulster University for providing laboratory facilities to J.M.C.

23

REFERENCES

(1) Conlon, J.M.; Mechkarska, M.; Lukic, M.L.; Flatt, P.R. Peptides 2014, 57, 67-

77.

(2)Xu, X.; Lai, R. Chem. Rev. 2015, 115, 1760-1846.

(3)Conlon, J.M.; Kolodziejek, J.; Mechkarska, M.; Coquet, L.; Leprince, J.; Jouenne, T.;

Vaudry, H.; Nielsen, P.F.; Nowotny, N.; King, J.D. Comp. Biochem. Physiol.,

Part D: Genomics Proteomics 2014, 9, 49-57.

(4)Marani, M.M.; Dourado, F.S.; Quelemes, P.V.; de Araujo, A.R.; Perfeito, M.L.;

Barbosa, E.A.; Véras, L.M.; Coelho, A.L.; Andrade, E.B.; Eaton, P.; Longo, J.P.;

Azevedo, R.B.; Delerue-Matos, C.; Leite, J.R. J. Nat. Prod. 2015, 78, 1495-

1504.

(5)Conlon, J.M.; Mechkarska, M.; Kolodziejek, J.; Nowotny, N.; Coquet, L.; Leprince J.;

Jouenne, T.; Vaudry, H. Comp. Biochem. Physiol., Part D: Genomics

Proteomics 2014 12, 45-52.

(6)Ojo, O.O.; Abdel-Wahab, Y.H.A.; Flatt, P.R.; Mechkarska, M.; Conlon, J.M. Diabetes Obes.

Metab. 2011, 13, 1114-1122.

(7)Srinivasan, D.; Ojo, O.O.; Abdel-Wahab, Y.H.; Flatt, P.R.; Guilhaudis, L.; Conlon,

J.M. Peptides 2014, 55, 23-31.

(8)Ojo, O.O.; Srinivasan, D.K.; Owolabi, B.O.; McGahon, M.K.; Moffett, R.C.; Curtis,

T.M.; Conlon, J.M.; Flatt, P.R.; Abdel-Wahab, Y.H.A. Biol. Chem. 2016, in

press.

(9)Ojo, O.O.; Conlon, J.M.; Flatt, P.R.; Abdel-Wahab, Y.H.A., Biochem. Biophys.Res.

Commun. 2013, 431, 14-18.

24

(10) Ojo, O.O.; Srinivasan, D.K., Owolabi, B.O., Flatt, P.R., Abdel-Wahab, Y.H.A.,

Biochimie 2015, 109, 18-26.

(11) Srinivasan, D.K., Ojo, O.O.; Owolabi, B.O.; Conlon, J.M.; Flatt, P.R.; Abdel-

Wahab, Y.H. Acta Diabetol. 2016, 53, 303-315.

(12) Pantic, J.M.; Mechkarska, M.; Lukic, M.L.; Conlon, J.M. Biochimie 2014, 101,

83-92.

(13) Rödel, M.-O. Herpetofauna of West Africa: amphibians of the West African

Savannah. Edition Chimaira: Frankfurt, Germany, 2000.

(14) Frost, D.R. Amphibian Species of the World: An Online Reference. Version 6.0.

Electronic Database accessible at http://research.amnh.org/herpetology/

amphibia/index.html. American Museum of Natural History, New York, USA,

2016

(15) IUCN SSC Amphibian Specialist Group. The IUCN Red List of Threatened Species.

Version 2015.2. www.iucnredlist.org, 2014.

(16) Conlon, J.M. Nat. Protoc. 2007, 2, 191-197.

(17) Kyte, J.; Doolittle, D.F. J. Mol. Biol. 1982, 157, 105-132.

(18) Sai, K.P.; Jagannadham, M.V.; Vairamani, M.; Raju, N.P.; Devi, A.S.; Nagaraj,

R.; Sitaram, N. J. Biol. Chem. 2001, 276, 2701-2707.

(19) Song, Y.; Lu, Y.; Wang, L.; Yang, H.; Zhang, K.; Lai, R. Peptides 2009, 30,

1228-1232.

(20) Pyron, R.A.; Wiens, J.J. Mol. Phylogenet. Evol. 2011, 61, 543–583.

(21) Conlon, J.M.; Raza, H.; Coquet, L.; Jouenne, T.; Leprince, J.; Vaudry, H.; King,

J.D. Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol. 2009,150, 150-154.

(22) Mechkarska, M.; Ahmed, E.; Coquet, L.; Leprince, J.; Jouenne, T.; Vaudry, H.;

King, J.D.; Conlon, J.M. Peptides 2011, 32, 1502-1508.

25

(23) Alam, M.S.; Igawa, T.; Khan, M.M.; Islam, M.M.; Kuramoto, M.; Matsui, M.;

Kurabayashi, A.; Sumida, M. Mol. Phylogenet. Evol. 2008, 48, 515-527.

(24) Kosuch, J.; Vences, M.; Dubois, A.; Ohler, A.; Böhme, W. Mol. Phylogenet.

Evol. 2001, 21, 398-407.

(25) Cavanagh, J.; Fairbrother, W. J.; Palmer, A. G., III; Rance, M.; Skelton, N. J.

Protein NMR Spectroscopy - Principles and Practice, 2nd ed.; Elsevier

Academic Press: Oxford, U.K., 2007.

(26) Sharma, D.; Rajarathnam, K. Journal of Biomolecular NMR, 2000, 18, 165-171.

(27) Shen, Y.; Delaglio, F.; Cornilescu, G.; Bax, A. J. Biomol. NMR, 2009, 44, 213-

223.

(28) Sitaram, N.; Sai, K.P.; Singh, S.; Sankaran, K.; Nagaraj, R. Antimicrob. Agents

Chemother. 2002, 46, 2279-2283.

(29) McClenaghan, N.H.; Barnett, C.R.; Ah-Sing, E.; Abdel-Wahab, Y.H.; O'Harte,

F.P.; Yoon, T.W.; Swanston-Flatt, S.K.; Flatt, P.R. Diabetes 1996, 45, 1132-

1140.

(30) Brubaker, P.L.; Schloos, J.; Drucker, D.J. Endocrinology 1998, 139, 4108-4114.

(31) Reimann, F., Williams, L., Xavier, G.D., Rutter, G.A., Gribble, F.M

Diabetologia 2004, 47, 1592-1601.

(32) Strober, W.; Zhang, F.; Kitani, A.; Fuss, I.; Fichtner-Feigl, S. Curr. Opin.

Gastroenterol. 2010, 26, 310-317.

(33) Manzo, G., Scorciapino, M.A., Srinivasan, D., Attoub, S., Mangoni, M.L.,

Rinaldi, A.C., Casu, M., Flatt, P.R., Conlon, J.M. J. Nat. Prod. 2015, 78, 3041-

3048.

(34) Ogg, R., Kingsley, P., and Taylor, J. J. Magn. Reson., Ser. B 1994, 104, 1-10.

(35) Smallcombe, S., Patt, S. L., and Keifer, P. J. Magn. Reson., Ser. A 1995, 117,

295-303.

(36) Flatt, P.R.; Bailey, C.J. Diabetologia 1981, 20, 573–577.

26

(37) Clinical Laboratory and Standards Institute. Approved Standard M07-A8; CLSI:

Wayne PA, 2008.

(38) Ray, A.; Dittel, B.D. J Vis Exp. 2010; 35: 1488.

27

Graphic

28