Peptides 71 (2015) 1–7

Contents lists available at ScienceDirect

Peptides
journal homepage: www.elsevier.com/locate/peptides

Stability, toxicity and intestinal permeation enhancement of two

food-derived antihypertensive tripeptides, Ile-Pro-Pro and
Leu-Lys-Pro
John P. Gleeson, Joanne Heade, Sinéad M. Ryan, David J. Brayden ∗
UCD School of Veterinary Medicine and UCD Conway Institute, University College Dublin, Belfield, Dublin 4, Ireland

a r t i c l e i n f o a b s t r a c t

Article history: Two food-derived ACE inhibitory peptides, Ile-Pro-Pro (IPP) and Leu-Lys-Pro (LKP), may have potential
Received 22 January 2015 as alternative treatments for treatment of mild- or pre-hypertension. Lack of stability to secretory and
Received in revised form 7 May 2015 intracellular peptidases and poor permeability across intestinal epithelia are typical limiting factors of
Accepted 25 May 2015
oral delivery of peptides. The stability of IPP and LKP was confirmed in vitro in rat intestinal washes,
Available online 3 June 2015
and intestinal and liver homogenates over 60 min. A positive protein control for peptidases, insulin, was
significantly digested in each format over the same period. Neither tripeptide showed cytotoxic activity
Keywords:
on Caco-2 and Hep G2 cells using the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-
Oral peptides
Food-derived peptides
sulfophenyl)-2H-tetrazolium (MTS) assay, even after chronic exposure. The basal Papp of fluorescein
Intestinal peptide transport isothiocyanate (FITC)-labeled IPP and FITC-LKP across isolated rat jejunal and colonic mucosae were
Permeation enhancers low, but were significantly increased in each tissue type by the medium chain fatty acids (MCFA) per-
Intestinal peptidases meation enhancers, sodium caprate (C10 ) and the sodium salt of 10-undecylenic acid (uC11 ). IPP and LKP
Anti-hypertensive peptides were therefore stable against intestinal and liver peptidases and were non-cytotoxic; their Papp values
across rat intestinal mucosae were low, but could be increased by MCFA. There is potential to make on
oral dosage form once in vivo pharmacology is confirmed.
© 2015 Elsevier Inc. All rights reserved.

Introduction potentially be used as a nutritional treatment for mild hyperten-

sion or pre-hypertension, when formulated in functional foods or
The body regulates blood pressure in part via the renin- as nutraceuticals [6–9].
angiotensin-aldosterone system (RAAS). Renin is produced in the Proteins from food undergo enzymatic hydrolysis by digestive
kidney and acts on angiotensinogen, producing angiotensin I, which enzymes including stomach pepsin and duodenal serine proteases
then passes through the pulmonary circulation and is converted to release smaller peptides, which may be bioactive. For example,
by angiotensin converting enzyme (ACE) to the potent vaso- Ile-Pro-Pro (IPP) was isolated from milk ␤-casein following fer-
constrictor, angiotensin II [1,2]. ACE inhibitors act competitively, mentation by Lactobacillus helveticus [10]. Leu-Lys-Pro (LKP) was
whereas angiotensin receptor antagonists prevent receptor bind- isolated from chicken and fish muscle after digestion by the bac-
ing of angiotensin II; both cause vasodilation [3]. Food-derived terial enzyme, thermolysin [11]. Both tripeptides are established
ACE inhibitors may provide the opportunity to maintain nor- as ACE inhibitors in vitro, with respective IC50 values of 5 ␮M and
mal blood pressure and may prevent escalation of hypertension. 0.32 ␮M [12]. Hypotensive effects of the tripeptides on systolic
Meta-analyses of randomized control trials of food-derived anti- blood pressure were demonstrated in the Spontaneously Hyperten-
hypertensive peptides concluded that these peptides may lead sive Rat (SHR) following intravenous administration. It was noted
to reduction of blood pressure. This conclusion could be made, however, that a reduction of 18 mm Hg with oral dose of 5 mg/kg
even though these peptides were not presented in optimized IPP and of 18 mmHg with a dose of 60 mg/kg LKP could be detected
oral formulations, although fermented milk matrices may offer in the rats [13,14]. These studies indicate that some oral bioavail-
some benefits as a delivery system [4,5]. Therefore they could ability leading to a pharmacodynamic effect could be achieved with
non-optimized formulations.
Stability of the tripeptides could be an issue, as they may
∗ Corresponding author. Tel.: +353 1 7166013. potentially be digested to single amino acids by intracellular
E-mail address: [email protected] (D.J. Brayden). carboxypeptidases, however IPP and LKP appeared resistant to

http://dx.doi.org/10.1016/j.peptides.2015.05.009
0196-9781/© 2015 Elsevier Inc. All rights reserved.
2 J.P. Gleeson et al. / Peptides 71 (2015) 1–7

metabolism by Caco-2 cell brush border aminopeptidases [15]. Triton X-100TM (0.05%) as a positive control. The time points were
Permeation of the tripeptides likely requires a contribution from selected to mimic acute (1 h) exposure in Caco-2 cells and chronic
transcellular routing across small intestinal enterocytes. If so, they exposure (24 and 72 h) in each cell type. Cells were treated with
are likely candidates for the hPEPT1 carrier, similar to captopril MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-
[16], and further investigation of their stability to intracellular 2-(4-sulfophenyl)-2H-tetrazolium accordingly. Optical density
peptidases is necessary in the event that they exploit this path- (OD) was measured at 490 nm. Each value presented was normal-
way. It is also possible that IPP and LPK can also take advantage ized against untreated control and calculated from three separate
of the paracellular pathway due to their hydrophilicity and rela- experiments, each of which included six replicates.
tively low molecular weights. Most nutrient absorption occurs in
the small intestine due to its larger surface area and leakier epithe- Preparation of ex vivo rat gastrointestinal enzyme fluids and liver
lial tight-junctions compared to the colon [17]. Medium chain fatty homogenates
acids (MCFA) are powerful permeation enhancers acting in part by
re-organising tricellulin and claudin 5 proteins at the tight junc- Male Wistar rats (250–300 g; Charles River, UK) were eutha-
tion [18], as well as through a mild detergent fluidizing effect nized by stunning and cervical dislocation. A small intestine section
on the plasma membrane [19]. Several peptides are in oral clini- of 15 cm was removed, flushed with simulated intestinal fluid san
cal trials using MCFA-based technologies [20,21]. One of the lead pancreatin (SIFsp). SIFsp was prepared as per United States Phar-
MCFA in clinical trials, capric acid, is found in mM concentrations macopeia: a 25 mM KH2 PO4 buffer solution at pH 6.8 [24]. The
in dairy products, and its sodium salt (C10 ) has shown extensive fluid was collected and extracted three times with 5 ml ice-cold
enhancement effects in vitro and in vivo [22]. Recently, an over-the- dichloromethane to remove lipids. The extract was filtered using
counter nutritional supplement and antifungal agent, the sodium 0.45 ␮m syringe filters [25], resulting in a fluid designated as ‘gut
salt of 10-undecylenic acid (uC11 ), demonstrated similar perme- wash’ (GW). The initial tissue was then homogenized in HBSS at
ation enhancement effects to C10 in vitro using isolated rat intestinal 30 Hz for 2 min using a Qiagen TissueLyser® . The homogenate was
mucosae and in vivo in rat intestinal loop instillations [23]. How- centrifuged at 3695 rpm for 5 min and the supernatant was fil-
ever, applying such enhancers to food-derived peptides in oral tered using 0.45 ␮m syringe filters resulting in a fluid designated
formulations has not been attempted before. ‘intestinal homogenate’ (IH). A fresh liver (∼7 g) was harvested from
This study was undertaken to (a) determine the stability of a euthanized Wistar rat, placed in ice-cold saline and homoge-
IPP and LKP in rat intestinal fluid, intestinal homogenates and nized in 10 mM PBS pH 7.4 at 30 Hz for 2 min. The homogenate
liver homogenates; (b) to determine cytotoxicity of the peptides was centrifuged 2000 rpm at 4 ◦ C for 5 min, and the supernatant
in Caco-2 intestinal cells and Hep G2 liver cells; and (c) to test collected was designated as ‘liver homogenate’ (LH). Each fluid and
the capacity of C10 and uC11 to increase the permeability of fluo- homogenate was prepared using three independent replicates.
rescein isothiocyanate FITC-IPP and LKP across isolated rat small
intestinal and colonic tissue mucosae mounted in Ussing cham-
Stability studies in isolated rat intestinal and liver extracts
bers. The data show that that IPP and LKP are stable, non-cytotoxic,
and are amenable to having their permeability increased across the
IPP and LKP (4 mM) were incubated in GW, IH and LH for 60 min
rat intestinal epithelium. This is the first demonstration that uC11
at 37 ◦ C and agitated on a shaker at 150 rpm. Recombinant human
can improve intestinal permeability of a peptide.
insulin (200 ␮M) was incubated in GW, IH and LH as a positive
control to confirm metabolic activity [25]. Samples were taken at
Materials and methods 0, 30 and 60 min, centrifuged at 10,000 rpm at 4 ◦ C; the reaction
was stopped by placing the samples on dry ice. All samples were
Reagents and chemicals analysed for direct stability by reverse phase (RP)-HPLC. Each value
was calculated from three separate experiments.
Synthetic IPP (MW = 325), LKP (MW = 356) and correspond-
ing FITC-labeled versions were obtained from China Peptides Reverse phase HPLC analysis
(Shanghai, China). CellTitre 96® AQueous One Solution Cell Prolif-
eration Assay was supplied by Promega (Madison, USA). C10 was Stability of the tripeptides was measured by RP-HPLC using a
obtained from Fluka (Germany); uC11 was from Chemos (Germany). modified LC-MS method [26]. Samples were analysed with a Varian
Human recombinant insulin USP grade was expressed in Saccha- 920 HPLC using a Luna 5 ␮ C18 [2] column 250 × 4.6 mm (Phen-
romyces cerevisiae from Sigma–Aldrich (UK). All other reagents, omenex, UK). Gradient elution was carried out at a flow rate of
chemicals and solvents were analytical grade from Sigma–Aldrich 0.5 mL/min, with a mobile phase A, containing water and 0.05% TFA,
(UK). Caco-2 cells (passage 38–48) were obtained from European and a mobile phase B, containing acetonitrile and 0.05% TFA. The
Collection of Cell Cultures (Salisbury, UK). HepG2 cells (passage gradient sequence was: 100% A from 0 to 5 min, 5–30% B from 10
6–16) were obtained from American Type Culture Collection. to 25 min, 30–70% B from 25 to 30 min, 70–5% B from 30 to 31 min,
Experiments on post-mortem rat intestinal tissue were carried out 5–0% B from 31 to 35 min, and 100% A from 35 to 37 min. Tray tem-
following approval by the local UCD Animal Research Ethics Com- perature was maintained at 8 ◦ C, the injection volume was 50 ␮L
mittee (Protocol: AREC 14–28-Brayden). and the UV absorbance was 214 nm.

MTS assay Ussing chamber studies: FITC-tripeptide fluxes across isolated rat
intestinal mucosae
Caco-2 and HepG2 at a density of 2 × 104 cells/well were
cultured on 96-well plated in DMEM and EMEM respectively, Male Wistar rats (250–300 g; Charles River, UK) were euth-
supplemented with 10% fetal bovine serum, 1% l-glutamine, 1% anized by stunning and cervical dislocation. Colon and jejunum
penicillin-streptomycin and 1% non-essential amino acids, and were removed, opened along the mesenteric border and rinsed
incubated at 37 ◦ C in a humidified incubator with 5% CO2 and 95% with warm oxygenated Krebs–Henseleit buffer (KH) [27]. Fifteen
O2 . The assay was carried out using 1 and 24 h exposure times for cm of jejunum was excised at a point 10–12 cm proximal to the
the peptides on Caco-2 cells, and for 72 h on HepG2 cells, using stomach and pinned with the mucosal side down and mounted in
 J.P. Gleeson et al. / Peptides 71 (2015) 1–7 3

Ussing chambers with a circular window of 0.63 cm2 , bathed bilat- A

erally with 5 ml KH and continuously gassed with 95% CO2 /5% O2
140
at pH 7.4 maintained at 37 ◦ C. Colon tissue was excised, cleaned of

Percentage of IPP (%)

fecal matter, pinned with the mucosal side down and the mus- 120
cle layers were removed and mounted in Ussing chambers as 100
described. The transepithelial potential difference (PD; mV) and
short circuit current (Isc , ␮A) were measured across jejunal and 80
colonic mucosae using a DVC-4000 voltage clamp apparatus (World 60
Precision Instruments, Hertfordshire, UK). Analog data were digi-
40
tized with a Powerlab® data acquisition unit and analysed with
LabChart® software (AD Instruments, Oxford, UK). After an initial 20
30 min equilibration (jejunum) and 45 min equilibration (colon), 0
PD and Isc were used to calculate the transepithelial electrical 0 30 60
resistance (TEER,  cm2 ) over 120 min [28]. FITC-IPP (500 ␮M) or Time (min)
FITC-LKP (500 ␮M) were added to the apical chamber with or with-
out the addition of C10 (10 mM) or uC11 (10 mM). Apical additions B
for MCFA were in calcium-free KH to prevent precipitation. Baso- 120

Percentage of LKP (%)

lateral samples (100 ␮l) were collected every 20 min over 120 min
100
and transferred to light-protected white 96 well plates. Volumes
were replaced with fresh KH, apical-side samples were collected 80
before MCFA were added and also after 120 min. Fluorescence was
measured in a spectrofluorimeter (MD Spectramax Gemini) with 60
an excitation wavelength of 490 nm and an emission wavelength 40
of 525 nm. The Papp (cm s−1 ) was determined using the following
equation: 20

dQ 1 0
Papp = 0 30 60
dt A · C0
where dQ/dt is the transport rate across the epithelium (slope
Time (min)
of cumulative amount FITC-peptide versus time) (mol/s); A is the
C
surface area (0.63 cm2 ); C0 is the initial concentration of the FITC- 120
Percentage of Insulin (%)

peptide in the apical compartment (mol/ml).

100
Histology 80 *

For experiments with jejunal tissue and colonic mucosae 60 ***

*** **
exposed to FITC-peptides and MCFA, tissues were removed after
40
120 min exposure in Ussing chambers and immersed in 10% (v/v)
buffered formalin for at 48 h. Tissues were stained with hema- 20
toxylin and eosin (H&E), Alcian blue, and neutral red. Slides
were visualized under a light microscope (NanoZoomer 2.0- 0
HT light microscopy, Hamamatsu) and images were taken with 0 30 60
high-resolution camera (Micropublisher 3.3 RTV, QImaging) and Time (min)
Image-Pro® Plus version 6.3 (Media Cybernetics Inc., USA) acquisi-
Fig. 1. Stability of (A) IPP, 4 mM; (B) 4 mM LKP 4 mM; (C) 200 ␮M human insulin
tion software.
in rat gut washes (GW – White), intestinal homogenates (IH – Gray) and liver
homogenates (LH – Black). One-way ANOVA with Dunnett’s multiple comparison.
Statistics Mean ± SEM n = 3; *P < 0.05, **P < 0.01, and ***P < 0.001 respectively, compared with
control (0 min). 3. 2 MCFA increase the Papp of FITC-tripeptides in rat jejunal and
colonic mucosae.
Statistical significance was measured by one-way ANOVA or
two-way ANOVA with either Dunnett and Bonferroni post hoc tests.
P < 0.05 was the required level to denote statistical significance. SIFsp, which is buffered to the optimum pH for enzymatic activity,
whereas, homogenates were prepared in salt solutions (HBSS and
Results PBS). Overall, IPP and LKP appear highly resistant to metabolism by
rat intestinal and liver enzymes.
Tripeptides are stable in rat gastrointestinal and liver extracts Jejunal basal TEER values were 56 ± 17  cm2 (n = 47), within
the published range [29]. Jejunal TEER gradually decreased over
The stability of IPP (4 mM) and LKP (4 mM) against GW, IH 120 min to 80% of the basal value in the presence of apically added
and LH were analysed by RP-HPLC. Incubation of the two pep- FITC-tripeptides (500 ␮M), but this was no different from untreated
tides in the three extracts showed no evidence of metabolism over control (data not shown). Apical addition of 10 mM C10 and uC11
60 min, confirming stability (Fig. 1A and B). Peptidase capacity significantly decreased TEER to 44% and 40% of basal value respec-
of the three systems was confirmed by significant breakdown of tively, showing significant decreases between 15 and 120 min
200 ␮M human recombinant insulin in each system over the same compared to control tissue (Fig. 2). Colonic mucosae basal TEER val-
period (Fig. 1C) showing breakdown similar to a previous study ues were 112 ± 32  cm2 (n = 29), again within the accepted range
[25]. The extract with the greatest capacity for metabolism was the [29] In the presence of FITC-tripeptides, colonic TEER showed a
rat gut wash (GW), as indicated by 65% breakdown of insulin com- small decrease of 9% over 120 min, but this was not different from
pared to that seen in the two homogenates. GW was prepared using controls. Addition of C10 and uC11 significantly decreased TEER, to
4 J.P. Gleeson et al. / Peptides 71 (2015) 1–7

A 120 A
0.8
% Control TEER
100

Papp (cm/s)(x 10-5)

0.6 *** ** ***
80 *** ***
*** 0.4 *
60 ***
*** 0.2
40
0.0
-20 0 20 40 60 80 100 120 No Enhancer + C10 + uC11
Time (min)
B 120 B 1.5
100 ***
% Control TEER

***

Papp (cm/s)(x 10-5)

*** ***
80 *** 1.0
60
***
40 0.5
***
20 *** *** *** *** *** ***
0 0.0
No Enhancer + C10 + uC11
-20 0 20 40 60 80 100 120
Time (min) Fig. 3. Effect of apical addition of 10 mM C10 and uC11 on Papp of apically-added FITC-
IPP (500 ␮M; white column) and FITC-LKP (500 ␮M; black column) in (A) jejunal
Fig. 2. MCFA decrease TEER across (A) isolated rat jejunum; (B) rat colon. Concen- tissue and (B) colonic mucosae. One-way ANOVA with Dunnett’s multiple compari-
trations were 500 ␮M FITC-IPP FITC-LKP and 10 mM C10 and uC11 . FITC-IPP, son n = 5; *, **, and *** indicate P < 0.05, 0.01 and 0.001 respectively, compared with
FITC-IPP + C10 , FITC-IPP + uC11 , FITC-LKP, FITC-LKP + C10 , control. Each value represents the mean ± SEM.
, FITC-LKP + uC11 . Two-way ANOVA with Bonferonni’s post hoc test, n = 5; *P < 0.05,
**P < 0.01, and ***P < 0.001 respectively versus FITC-tripeptide counterpart in the
absence of the MCFA. Each value represents the mean ± SEM. IPP and LKP are not cytotoxic in vitro: MTS and histology

There was no evidence of cytotoxicity for each peptide in two

just 3% and 8% of basal values respectively from between 5 and cell lines at different time points (Table 2). The MTS assay data
120 min compared to tissues exposed to FITC-tripeptides alone indicate that unlabelled IPP and LKP was not cytotoxic in Caco-2
(Fig. 2). Both MCFA caused similar TEER reductions in colonic TEER or HepG2 cells from 0.1 to 10 mM at selected time points. The MTS
values. assay yielded low errors, similar to the findings of others [30]. A pos-
The basal Papp of FITC-IPP showed no differences across itive control, Triton® -X-100 (0.05%), massively reduced viability of
isolated jejunal tissue (1.39 ± 0.73 × 10−6 cm s−1 ) or colonic both cell types at each incubation time point.
mucosae (1.35 ± 0.68 × 10−6 cm s−1 ). However, FITC-LKP had Triton® : Triton® -X-100 (0.05% w/v). Data are expressed as
a significantly increased basal Papp across jejunal mucosae means ± SEM. One-way ANOVA with Dunnett’s multiple compari-
(2.14 ± 0.64 × 10−6 cm s−1 , P < 0.01) compared to colonic mucosae son test. 1 h and 24 incubations of unlabelled peptides on Caco-2
(0.71 ± 0.36 × 10−6 cm s−1 ) (Fig. 3). Apical addition of the two
MCFAs (at the same concentrations as those that reduced TEER)
Table 2
led to an increased Papp of FITC-IPP and FITC-LKP across both jeju-
MTS analysis of IPP and LKP following incubation for 1 h and 24 h on Caco-2 and for
nal and colonic mucosae. More significant FITC-tripeptide Papp 72 h on HepG2 cells.
increases induced by the MCFA were seen in colonic mucosae com-
pared to jejunal (Table 1). 1h 24 h 72 h
Caco-2 Caco-2 HepG2

IPP
10 mM 101.2 ± 1.4 84.3 ± 1.8 99.7 ± 0.8
Table 1
5 mM 100.7 ± 1.0 89.2 ± 0.4 99.3 ± 0.3
Papp values across intestinal mucosae of FITC-tripeptides in the presence of MCFA.
1 mM 100.6 ± 0.9 94.6 ± 2.2 101.3 ± 0.6
Jejunum Colon 100 ␮M 99.5 ± 2.9 94.9 ± 1.0 99.3 ± 0.3
LKP
−6 −1
Papp (10 cm s ) Fold Papp (10−6 cm s−1 ) Fold 10 mM 93.7 ± 2.8 93.7 ± 5.6 82.2 ± 8.9
increase increase 5 mM 102.0 ± 4.0 101.7 ± 2.9 99.7 ± 2.7
1 mM 105.7 ± 2.3 107.0 ± 3.6 107.3 ± 1.7
FITC-IPP 1.3 ± 0.0 – 1.3 ± 0.3 –
100 ␮M 98.3 ± 0.6 108.7 ± 2.6 103.0 ± 2.0
+C10 5.0 ± 0.6*** 3.6 11.6 ± 0.7*** 8.6
+uC11 4.8 ± 0.6** 3.4 9.3 ± 1.2*** 6.9 Triton® 36.8 ± 3.6*** 34.9 ± 4.3*** 18.4 ± 2.0***
FITC-LKP 2.1 ± 0.2 – 0.7 ± 0.1 –
+C10 3.2 ± 0.2* 1.4 10.3 ± 1.0*** 14.4 Triton® : Triton® -X-100 (0.05% w/v). Data are expressed as means ± SEM. One-way
+uC11 4.8 ± 0.4*** 2.2 9.3 ± 0.8*** 13.0 ANOVA with Dunnett’s multiple comparison test. 1 h and 24 incubations of unla-
belled peptides on Caco-2 and 72 h on HepG2 cells.
Data are expressed as means ± SEM. One-way ANOVA with Dunnett’s multiple com- *
P < 0.05.
parison test; concentrations as in Fig. 3; n = 5. **
P < 0.01.
*
P < 0.05 ***
P < 0.001.
**
P < 0.01, and n = 3 independent experiments for each concentration and time point with replicates
***
P < 0.001 compared with control peptides in the absence of MCFA. of six.
 J.P. Gleeson et al. / Peptides 71 (2015) 1–7 5

Fig. 4. Alcian blue and neutral red-strained light micrographs of rat intestinal mucosae mounting in Ussing chambers and exposed to FITC-tripeptides (500 ␮M) and MCFA
(10 mM) for 120 min. (A) Untreated colon, (B) FITC-IPP in colon, (C) 10 mM C10 in colon, (D) 10 mM uC11 in colon, (E) Untreated jejunum, (F) FITC-IPP in jejunum, (G) 10 mM C10
in jejunum, (H) 10 mM uC11 in jejunum. Horizontal bars denote 100 ␮m.

and 72 h on HepG2 cells.*, **, and *** indicate P < 0.05, 0.01 and 0.001 in an un-buffered physiological relevant salt solution. Therefore,
respectively. n = 3 independent experiments for each concentration GW and IH can be used as complementary methods of determin-
and time point with replicates of six. ing oral peptide stability to secretory and/or intracellular enzymes.
The effects of FITC-IPP and FITC-LKP in the presence or absence It was also found that IPP and LKP are resistant to metabolism
of C10 and uC11 were examined by histological analysis of jejunal by liver enzymes, suggesting that the tripeptides will not be
tissue and colonic mucosae following 120 min exposure in Ussing degraded during first pass metabolism and will be transported
chambers. It should be noted that the cytotoxicity assays exam- into systemic circulation intact. The presence of proline moieties in
ined unlabelled peptides whereas the histology study assessed the these tripeptides seem to provide inherent resistance to enzymatic
FITC-labeled peptides. Previous work with these peptides on Ussing degradation [35,36], so the stability results are in keeping with
chambers noted no impact on the isolated tissue [31], and there is that.
no evidence suggesting that FITC is damaging [32]. Control jejunal The tripeptides showed no cytotoxicity using the MTS assays on
tissue after 120 min showed an intact epithelium with active goblet human colonic-derived epithelial enterocyte- and liver hepatocyte
cells; exposure to 0.5 mM FITC-IPP or FITC–LKP showed no damage cell lines. A previous study found no potential for IPP and VPP to
(Fig. 4E and F). The MCFA induced cellular sloughing of jejunal villi, induce cytotoxicity or clastogenicity in Chinese hamster lung cells
however, the overall physiological structure was retained (Fig. 4G [37]. The cytotoxicity data and histology confirm that any impact
and H). Untreated rat colonic mucosae after 120 min also had an in the tissue mounted on the Ussing chambers are most likely from
intact epithelium (Fig. 4A); FITC-IPP and FITC-LKP caused no dam- the MCFA, which were previously shown to have significant cyto-
age to the epithelium (Fig. 4B). Effect of MCFA on colonic mucosae toxicity in the MTT assay at 1 and 24 h exposures in Caco-2 cells
shows perturbation of the epithelium, associated with their known [23]. Mucosa histology also shows a perturbation of the intestinal
mild surfactant actions (Fig. 4C and D). Overall, the histology was and colonic epithelia, comparable to previous studies investigating
complementary to the data indicating absence of cytotoxic effects these MCFA [23]. Repair is possible following transient mild dam-
in the presence of the peptides, suggesting that any mild damage age to the tissue in intact in situ rat intestinal instillations using
caused was due to the MCFA. similar enhancers, which caused similar initial perturbation [38].
Due to the stability of IPP and LKP, the main limiting factor in
Discussion their application must therefore be poor intestinal permeability.
Some oral bioavailability was indicated when IPP was delivered in a
A previous study showed the resistance of IPP and LKP to brush- yogurt matrix, showing 2.1-fold increase compared to placebo con-
border peptidases in Caco-2 monolayers incubated for 60 min in trol in a small randomized human cohort [39], but overall delivery
HBSS [15]. The data presented here confirm these findings, how- was still very low. Others showed that after oral administration of a
ever, HBSS is not very relevant for the physiology of the GI tract. Due [14 C]-IPP analog to rats, peak plasma radioactivity (albeit low level)
to this, simulated intestinal fluids are used as a more physiological was seen after 2 h, indicating some absorption across the intestine
in vitro system incorporating modified pH and surfactants, how- [40]. Accumulation was seen in tissues associated with the renin-
ever, they still underestimate potential breakdown [33], as they do angiotensin system, although the highest was in the duodenum and
not include brush-border and intracellular peptidases. The intesti- jejunum. Further to this, the Papp of IPP and fluorescein across dis-
nal gut wash and homogenate used incorporates these enzymes tal jejunum mounted in Ussing chambers was reported by another
and provides a better prediction of the stability of these peptides group as 5.6 ± 0.7 × 10−8 cm s−1 and 5.9 ± 1.1 × 10−6 cm s−1 respec-
in vivo. The ex vivo gastrointestinal fluids include secretory diges- tively [31], consistent with hydrophilic agents with relatively
tive enzymes in the gut wash (GW) and brush-border peptidases low passive permeability. In sum, rat in vitro and in vivo along
and intracellular enzymes in intestinal homogenate (IH) [25]. Each with human in vivo studies have shown a low permeability of
fluid was metabolically active, demonstrating digestion of insulin; IPP, whereas values for LKP are unreported. Upon conjugation
GW showed the most degradation of insulin comparable to a similar of FITC to IPP and LKP, the resulting Papp values across prox-
study [34]. As the GW was produced by flushing the small intestinal imal jejunum in the current study were 1.39 ± 0.7 × 10−6 and
segment with SIFsp, the resulting fluid is buffered to a pH favorable 2.14 ± 0.6 × 10−6 cm s−1 more comparable to fluorescein, which
for digestive enzymes. Whereas, the tissues were homogenized is used as a marker of paracellular transport with permeation
6 J.P. Gleeson et al. / Peptides 71 (2015) 1–7

enhancers (Table 1). The Papp of FITC-tripeptides was indirectly Acknowledgements

quantified by the FITC moiety, which is covalently bonded to the
tripeptides and evidence suggests integrity of FITC-labeled pep- This study was supported by an Irish Department of Agricul-
tides [41]. Therefore, successful formulation of IPP or LKP in an oral ture Food Institutional Research Measure (FIRM) grant ‘NUTRADEL,’
dosage form will likely require significant permeation enhance- grant number 11F042. We thank Mrs Margot Coady for assistance
ment as a design feature. with the histology.
Addition of C10 and uC11 at 10 mM resulted in significant
increases in FITC-tripeptides permeability across both jejunum
References
and colon tissues. C10 is regarded as a ‘gold standard’ intesti-
nal permeation enhancer, however, recently uC11 was shown [1] Skeggs LT, Dorer FE, Kahn JR, Lentz KE, Levine M. The biochemistry of the
to have similar efficacy in isolated rat colon and in in situ rat renin-angiotensin system and its role in hypertension. Am J Med 1976;60(6):
intestinal instillations, whereby the Papp of FD-4 was increased 737–48.
[2] Laragh JH, Baer L, Brunner HR, Buhler FR, Sealey JE, Vaughan ED. Renin,
several fold by both C10 and uC11 [23]. These enhancers reduced angiotensin and aldosterone system in pathogenesis and management of
colonic TEER values in accordance with previously published hypertensive vascular disease. Am J Med 1972;52(5):633–52.
results [21]. Both MCFA were effective in increasing colonic perme- [3] Bezalel S, Mahlab-Guri K, Asher I, Werner B, Sthoeger ZM. Angiotensin convert-
ing enzyme inhibitor induced angioedema. Am J Med 2015;128(2):120–5.
ability of FITC-IPP and FITC-LKP. Insulin absorption enhancement [4] Sipola M, Finckenberg P, Korpela R, Vapaatalo H, Nurminen M-L. Effect of long-
by C10 using a loop model showed a rank order of efficacy: term intake of milk products on blood pressure in hypertensive rats. J Dairy Res
colon > ileum > jejunum > duodenum [42]. This correlates with the 2002;69(01).
[5] Sipola M, Finckenberg P, Santisteban J, Korpela R, vapaatalo H, Nurminen ML.
increase in Papp of FITC-IPP and FITC-LKP in colon (8.62 and 14.49
Long-term intake of milk peptides attenuates development of hypertension in
respectively) compared to jejunum (3.62 and 1.49 respectively). spontaneously hypertensive rats. J Physiol Pharmacol 2001;52(4):745–54.
However, uC11 had not previously been tested in jejunal tissue, with [6] Xu JY, Qin LQ, Wang PY, Li W, Chang C. Effect of milk tripeptides on
blood pressure: a meta-analysis of randomized controlled trials. Nutrition
increases in Papp 3.4-fold (FITC-IPP) and 2.2-fold (FITC-LKP). This is
2008;24(10):933–40.
in agreement with previous work on other anionic surfactants stat- [7] Pripp AH. Effect of peptides derived from food proteins on blood pressure: a
ing reduced enhancement activity due to reduced decrease in TEER meta-analysis of randomized controlled trials. Food Nutr Res 2008;51:1–9.
in jejunum/ileum compared to colon (Fig. 2). Studies investigating [8] Cicero AF, Rosticci M, Veronesi M, Bacchelli S, Strocchi E, Melegari C,
et al. Hemodynamic effects of lactotripeptides from casein hydrolysate in
intestinal permeation enhancement of tetradecyl maltoside (TDM) Mediterranean normotensive subjects and patients with high-normal blood
and dodecyl maltoside (DDM), and coco-glucosides (CG) showed no pressure: a randomized, double-blind, crossover clinical trial. J Med Food
enhancement of FD-4 permeability across jejunum [27,43]. MCFA, 2010;13(6):1369–78.
[9] Turpeinen AM, Jarvenpaa S, Kautiainen H, Korpela R, Vapaatalo H. Antihyper-
TDM, DDM and CG increase intracellular calcium after 1 h treat- tensive effects of bioactive tripeptides – a random effects meta-analysis. Ann
ment in Caco-2 cells, corresponding to decrease in TEER values in Med 2013;45(1):51–6.
colon [19]. However, jejunal epithelia is less sensitive to the effects [10] Nakamura Y, Yamamoto N, Sakai K, Okubo A, Yamazaki S, Takano T. Purification
and characterization of angiotensin I-converting enzyme inhibitors from sour
of these enhancers. 0.2% CG reduced TEER values in rat jejunum, milk. J Dairy Sci 1995;78(4):777–83.
showing a reduction equivalent to decreases from C10 and uC11 [11] Fujita H, Yokoyama K, Yoshikawa M. Classification and antihypertensive activ-
shown here. Although CG showed a significant decrease in jejunal ity of angiotensin i-converting enzyme inhibitory peptides derived from food
proteins. J Food Sci 2000;65(4):564–9.
TEER, the Papp of FD-4 was not affected. However, this is possi-
[12] Ryan JT, Ross RP, Bolton D, Fitzgerald GF, Stanton C. Bioactive peptides from
bly due to the size of the marker used (FD-4). The data presented muscle sources: meat and fish. Nutrients 2011;3(9):765–91.
here is in agreement with the initial hypothesis that many per- [13] Nakamura Y, Yamamoto N, Sakai K, Takano T. Antihypertensive effect of
sour milk and peptides isolated from it that are inhibitors to angiotensin I-
meation enhancers are more efficacious in colon than the small
converting enzyme. J Dairy Sci 1995;78(6):1253–7.
intestine [44], however, some enhancers including MCFA can still [14] Fujita H, Yoshikawa M. LKPNM: a prodrug-type ACE-inhibitory peptide derived
increase permeability across jejunum, the target site for nutrient from fish protein. Immunopharmacology 1999;44(1):123–7.
and bioactive peptide absorption. [15] Ohsawa K, Satsu H, Ohki K, Enjoh M, Takano T, Shimizu M. Producibility
and digestibility of antihypertensive beta-casein tripeptides, Val-Pro-Pro and
Future work is required on formulating antihypertensive pep- Ile-Pro-Pro, in the gastrointestinal tract: analyses using an in vitro model of
tides such as IPP, VPP (Val-Pro-Pro) and LKP and its prodrug mammalian gastrointestinal digestion. J Agric Food Chem 2008;56(3):854–8.
form LKPNM (Leu-Lys-Pro-Asp-Met). These types of agents have [16] Boll M, Marchovich D, Weber WM, Korte H, Daniel H, Murer H. Expression
cloning of a cDNA from rabbit small intestine related to proton-coupled trans-
been rejected for health claim status by the European Food Safety port of peptides, beta-lactam antibiotics and ACE-inhibitors. Pflugers Arch
Authority (EFSA), due to an insufficient “cause and effect” rela- 1994;429(1):146–9.
tionship [44–47]. Current formulations, often in functional food [17] Smart AL, Gaisford S, Basit AW. Oral peptide and protein delivery:
intestinal obstacles and commerical prospects. Expert Opin Drug Deliv
matrices rely on peptides reaching systemic circulation to affect 2014;11(8):1323–35.
blood pressure. Due to their low oral bioavailability however, such [18] Krug SM, Amasheh M, Dittmann II, Fromm C, Amasheh MS. Sodium caprate as
studies have not shown sufficient reduction or maintenance of an enhancer of macromolecule permeation across tricullular tight junctions of
intestinal cells. Biomaterials 2013;34:275–82.
blood pressure. Further work in SHR using intestinal permeation
[19] Brayden DJ, Gleeson JP, Walsh EG. A head-to-head multi-parametric high con-
enhancers may provide sufficient increased oral bioavailability to tent analysis of a series of medium chain fatty acid intestinal permeation
elicit a proper sustained hypotensive response. If so, this would enhancers in Caco-2 cells. Eur J Pharm Biopharm 2014;88(3):830–9.
[20] Maher S, Duffy B, Ryan A, Brayden DJ. Formulation strategies to improve oral
provide stimulus to allow formulation of these peptides with MCFA
peptide delivery. Pharm Pat Anal 2014;3(3):313–36.
in a solid dosage form. [21] Maher S, Leonard TW, Jacobsen J, Brayden DJ. Safety and efficacy of sodium
caprate in promoting oral drug absorption: from in vitro to the clinic. Adv Drug
Deliv Rev 2009;61(15):1427–49.
[22] Soderholm JD, Oman H, Blomguist L, Veen J, Lindmark T, Olaison G. Reversible
Conclusion increase in tight junction permeability to macromolecules in rat ileal
mucosa in vitro by sodium caprate, a constituent of milk fat. Dig Dis Sci
IPP and LKP were stable against secretory, brush-border, and 1998;1998(43):7.
[23] Brayden DJ, Walsh E. Efficacious intestinal permeation enhancement induced
intracellular intestinal enzymes and non-cytotoxic to intestinal and by the sodium salt of 10-undecylenic acid, a medium chain fatty acid derivative.
liver cells. C10 and uC11 significantly increased the permeability of AAPS J 2014;16(5):1064–76.
FITC-labeled IPP and LKP in isolated rat jejunal and colonic tissue. [24] Klein S. The use of biorelevant dissolution media to forecast the in vivo perfor-
mance of a drug. AAPS J 2010;12(3):397–406.
Combining IPP and LKP in a particulate formulation with selected [25] Hu S, Niu M, Hu F, Lu Y, Qi J, Yin Z, et al. Integrity and stability of oral liposomes
MCFA may have potential as an oral formulation for use in hyper- containing bile salts studied in simulated and ex vivo gastrointestinal media.
tension control. Int J Pharm 2013;441(1–2):693–700.
 J.P. Gleeson et al. / Peptides 71 (2015) 1–7 7

[26] van Platerink CJ, Janssen HG, Horsten R, Haverkamp J. Quantification of [38] Wang X, Maher S, Brayden DJ. Restoration of rat colonic epithelium after in situ
ACE inhibiting peptides in human plasma using high performance liquid intestinal instillations of the absorption promoter, sodium caprate. Therap
chromatography–mass spectrometry. J Chromatogr B Analyt Technol Biomed Deliv 2010;1(1):75–82.
Life Sci 2006;830(1):151–7. [39] Foltz M, Meynen EE, Bianco V, Van Platerink C, Koning TM, Kloek J. Angiotensin
[27] Petersen SB, Nolan G, Maher S, Rahbek UL, Guldbrandt M, Brayden DJ. Eval- converting enzyme inhibitory peptides from a lactotripeptide-enriched milk
uation of alkylmaltosides as intestinal permeation enhancers: comparison beverage are absorbed intact into the circulation. J Nutr 2007;137(4):953–8.
between rat intestinal mucosal sheets and Caco-2 monolayers. Eur J Pharm [40] Jauhiainen T, Wuolle K, Vapaatalo H, Kerojoki O, Nurmela K, Lowrie C, et al.
Sci 2012;47(4):701–12. Oral absorption, tissue distribution and excretion of a radiolabelled analog
[28] Cuthbert AW, Margolius HS. Kinins stimulate net chloride secretion by the rat of a milk-derived antihypertensive peptide, Ile-Pro-Pro, in rats. Int Dairy J
colon. Br J Pharmacol 1982;75(4):587–98. 2007;17(10):1216–23.
[29] Ungell AL, Nylander S, Bergstrand S, Sjoberg A, Lennernas H. Membrane trans- [41] Clausen AE, Bernkop-Schnurch A. Thiolated carboxymethylcellulose: in vitro
port of drugs in different regions of the intestinal tract of the rat. J Pharm Sci evaluation of its permeation enhancing effect on peptide drugs. Eur J Pharm
1998;87(3):360–6. Biopharm 2001;51:25–32.
[30] Wang P, Henning SM, Heber D. Limitations of MTT and MTS-based assays [42] Morishita M, Morishita I, Takayama K, Machida Y, Nagai T. Site-dependent effect
for measure of antiproliferative activity of green tea polyphenols. PLoS ONE of aprotinin, sodium caprate, Na2EDTA and sodium glycocholate on intestinal
2010;5. absorption of insulin. Biol Pharm Bull 1993;16(1):68–72.
[31] Foltz M, Cerstiaens A, van Meensel A, Mols R, van der Pijl PC, Duchateau GS, [43] Aguirre TAS, Rosa M, Guterres SS, Pohlmann AR, Coulter I, Brayden DJ. Investiga-
et al. The angiotensin converting enzyme inhibitory tripeptides Ile-Pro-Pro tion of coco-glucoside as a novel intestinal permeation enhancer in rat models.
and Val-Pro-Pro show increasing permeabilities with increasing physiological Eur J Pharm Biopharm 2014;88(3):856–65.
relevance of absorption models. Peptides 2008;29(8):1312–20. [44] EFSA, Panel on Dietetic Products, Nutrition, Allergies. Scientific Opinion on
[32] Dittmann I, Amasheh M, Krug SM, Markov AG, Fromm M, Amasheh S. Lau- the substantiation of health claims related to bonito protein peptide and
rate permeates the paracellular pathway for small molecules in the intestinal maintenance of normal blood pressure (ID 1716) pursuant to Article 13(1) of
epithelial cell model HT-29/B6 via opening the tight junctions by reversible Regulation (EC) No 1924/2006. EFSA J 2010;8(10):1–14.
relocation of Claudin-5. Pharm Res 2014;31(9):2539–48. [45] EFSA, Panel on Dietetic Products Nutrition and Allergies. Scientific opinion on
[33] Dressman JB, Reppas C. In vitro-in vivo correlations for lipophilic, poorly water- the substantiation of health claims related to isoleucine–proline–proline (IPP)
soluble drugs. Eur J Pharm Sci 2000;11(2):73–80. and valine–proline–proline (VPP) and maintenance of normal blood pressure
[34] Nielsen EJ, Yoshida S, Kamei N, Iwamae R, Khafagy el S, Olsen J, et al. In vivo proof (ID 615, 661, 1831, 1832, 2891), and maintenance of the elastic properties of the
of concept of oral insulin delivery based on a co-administration strategy with arteries (ID 1832) pursuant to Article 13(1) of Regulation (EC) No 1924/2006.
the cell-penetrating peptide penetratin. J Control Release 2014;189:19–24. EFSA J 2009;7(9):1–18.
[35] Cunningham DF, O’Connor B. Proline specific peptidases. Biochim biophysica [46] EFSA, Panel on Dietetic Products Nutrition and Allergies. Scientific Opinion on
acta 1997;1343:160–86. the substantiation of a health claim related to isoleucyl-prolyl-proline (IPP) and
[36] Yaron A, Naider F. Proline-dependent structural and biological properties of valyl-prolyl-proline (VPP) and maintenance of normal blood pressure pursuant
peptides and proteins. Crit Rev Biochem Mol Biol 1993;28(1):31–81. to Article 13(5) of Regulation (EC) No 1924/20061. EFSA J 2011;9(9):1–18.
[37] Maeno M, Mizuno S, Mennear JH, Bernard BK. Studies of the toxicological [47] EFSA, Panel on Dietetic Products Nutrition and Allergies. Scientific Opinion on
potential of tripeptides (l-valyl-l-prolyl-l-proline and l-isoleucyl-l-prolyl-l- the substantiation of health claims related to isoleucine–proline–proline (IPP)
proline): VIII Assessment of cytotoxicity and clastogenicity of tripeptides- and valine–proline–proline (VPP) and maintenance of normal blood pressure
containing casein hydrolysate and Lactobacillus helveticus-fermented milk (ID 661, 1831, 1832, 2891, further assessment) pursuant to Article 13(1) of
powders in Chinese hamster lung cells. Int J Toxicol 2005;24(Supp. 4):97–105. Regulation (EC) No 1924/2006. EFSA J 2012;10(6):1–22.