Role of the serotonergic pathway in uterotonic activity of Ananas comosus (L.) Merr. – An in vitro and in vivo study

Accepted Manuscript

Role of the serotonergic pathway in uterotonic activity of Ananas comosus (L.) Merr. - an in vitro and in vivo study Faezeh Monji , P Ganesan Adaikan , Lang Chu Lau , Abrar Al-Mahmood Siddiquee , Baharudin Bin Said , Lay-Kien Yang , Yoganathan K. , Mahesh A Choolani PII: DOI: Reference:

S0944-7113(18)30167-3 10.1016/j.phymed.2018.05.004 PHYMED 52506

To appear in:

Phytomedicine

Received date: Revised date: Accepted date:

1 November 2017 17 March 2018 7 May 2018

Please cite this article as: Faezeh Monji , P Ganesan Adaikan , Lang Chu Lau , Abrar Al-Mahmood Siddiquee , Baharudin Bin Said , Lay-Kien Yang , Yoganathan K. , Mahesh A Choolani , Role of the serotonergic pathway in uterotonic activity of Ananas comosus (L.) Merr. - an in vitro and in vivo study, Phytomedicine (2018), doi: 10.1016/j.phymed.2018.05.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Role of the serotonergic pathway in uterotonic activity of Ananas comosus (L.) Merr. - an in vitro and in vivo study

CR IP T

Running title: Uterotonic activity of A. comosus

Faezeh Monjia, P Ganesan Adaikana,*, Lang Chu Laua, Abrar Al-Mahmood Siddiqueea,

a

AN US

Baharudin Bin Saida, Lay-Kien Yangb, Yoganathan K.b, Mahesh A Choolania

Department of Obstetrics and Gynaecology, Yong Loo Lin School of Medicine, National

Bioinformatics Institute, Agency for Science, Technology and Research (A*STAR), 30 Biopolis

ED

b

M

University of Singapore, NUHS Tower Block, Level 12, 1E Kent Ridge Road, Singapore 119228

Street, #07-01 Matrix, Singapore 138671

PT

*Corresponding author

CE

Department of Obstetrics and Gynaecology, Yong Loo Lin School of Medicine, National University of Singapore, NUHS Tower Block, Level 12, 1E Kent Ridge Road 119228, Singapore

AC

Tel: +6566015453

1

ACCEPTED MANUSCRIPT

Fax: +6567794753

CR IP T

E-mail address: [email protected]

ABSTRACT

AN US

Background: Ananas comosus (L.) Merr. has been used as a traditional medicine in inducing abortion in many countries. Our previous in vitro experiments showed that the aqueous fraction (F4) of A. comosus extract stimulated the rat and human uterine contractions. Purpose: The aim of this study was to identify the bioactive compound and further investigate

M

the molecular mechanism of F4 induced contraction and the in vivo uterotonic effect of F4.

ED

Materials and Methods: Organ bath studies were employed to compare the stimulatory effect of F4 in non and late pregnant uterine tissue followed by isolation of protein from late pregnant

PT

uterine tissue for the western blot analysis. The PhysioTel transmitter was implanted in pregnant

CE

SD rats to measure the changes in intrauterine pressure (IUP). Analyses of the crude extract and

AC

active principle in F4 was performed using LC-HRMS.

2

ACCEPTED MANUSCRIPT

Results: Ripe F4 in a similar manner as serotonin produced a greater stimulatory response in late pregnant than non-pregnant uterine tissue without significant change in potency; ripe F4 also increased ERK phosphorylation which eventually led to a significant increase of the final product, MMP-13. In pregnant rats (E18), oral ripe F4 (1.5g.100g−1 body weight) and

CR IP T

ergometrine (1mg) did not stimulate the uterine contraction probably due to the low level of estradiol and as a consequence low 5-HT receptors at the time of administration. In contrast, in postpartum rats, oral administration of F4 and ergometrine produced a significant increase in maximal IUP to 4.3 and 4.9 folds of basal IUP respectively. Contrary to the folklore use, unripe

AN US

F4 did not stimulate the uterine activity during pregnancy and postpartum. Bioassay guided fractionation identified serotonin as a major bioactive compound in ripe F4. Conclusions: Our data clearly indicate that the uterotonic effect of ripe F4 is mediated via the

M

serotonergic pathway and suggest that serotonin rich diet may increase the peripheral serotonin

ED

and implicate in diverse physiological functions, including uterine motility.

PT

Keywords: Ananas comosus, Uterine contraction, MMP-13, Intrauterine pressure, Serotonin

CE

Abbreviations: F4, fraction 4; p-ERK1/2, phospho-extracellular signal–regulated kinase1/2;

AC

MMP-13, matrix metallopeptidase 13, 5-HT, 5-hydroxytryptamine; E, embryonic age; IUP, intrauterine pressure

Introduction

3

ACCEPTED MANUSCRIPT

Ananas comosus (L.) Merr., popularly known as pineapple is a plant of bromeliaceae family native to the tropical countries (Lim, 2012). This plant is used traditionally in several developing countries, specifically for women’s healthcare. The unripe pineapple is prescribed by traditional healer in Fiji for this purpose for the pregnant women up to three months to terminate the

CR IP T

pregnancy (Singh, 1986). The use of unripe fruit either juice or aqueous extract to produce abortion is extended to Philippines (Quisumbing, 1951), Indonesia (Burkill, 1966), Malaysia (De Laszlo and Henshaw, 1954) and Trinidad (Simpson, 1962). However, the ripe fruit is also used to prepare hot aqueous extract as an emmenagogue and abortifacient in India (Kamboj, 1988).

AN US

All these documented practices are not based on any scientific studies or merits.

Our previous in vitro experiments showed that the ripe and unripe aqueous fraction (F4) of A. comosus extract stimulated the uterine contractions through 5-HT receptors in rat and human

M

myometrium (Monji et al., 2016). Serotonin has been shown to stimulate uterine motility through

ED

activation of 5-HT2a receptors in different species (Cordeaux et al., 2009; Minosyan et al., 2007). To the best of our knowledge, the molecular mechanism of Ananas comosus induced contraction

PT

have not been examined. Hence, in this study, we investigated the involvement of 5-HT2a receptor in F4 induced contraction by comparing the stimulatory response to serotonin and ripe

CE

F4 in non and late pregnant rat in vitro.

AC

The 5-HT2a receptor activation induces several signaling pathways through Gαq protein (Phospholipase C/diacyl glycerol /Protein Kinase C) leading to muscular contraction (Masson et al., 2012). It has been demonstrated that stimulation of 5-HT2a receptors can induce the upregulation of matrix metallopeptidase 13 (MMP-13) via activation of protein kinase C and extracellular signal regulated kinase1/2 (ERK) phosphorylation (Shum et al., 2002). To establish

4

ACCEPTED MANUSCRIPT

the role of the serotonergic pathway in uterotonic activity of F4, the expression of phosphoextracellular signal regulated kinase1/2 (p-ERK1/2) and MMP-13 was measured to elucidate whether F4 and serotonin initiate similar post-receptor signaling cascades.

CR IP T

Studies carried out in the past on uterine activity were limited to in vitro measurements of tension. These methods evaluate the uterine muscle strips and are not collective. In vivo approaches are critical for the evaluation of whole uterine tissue response as an organ to understand physiological responses. For this purpose, we optimized a rat model to evaluate

AN US

uterine motility following oral administration of F4 in late pregnant and postpartum rats. Taken together, this study explores the molecular mechanism as well as the uterotonic effect of A. comosus extracts in vitro and in vivo. A further chemical analysis in this study identified and quantified a major bioactive compound in F4.

ED

Plant material collection

M

Materials and methods

PT

Unripe (2 months) and ripe (100 days) Ananas comosus (L.) Merr. were collected from the Johor region located in the southern part of Peninsular Malaysia (103°45′28″ E, 1°27′55″ N altitude 32

CE

m) in August 2015. The plant was identified and a voucher specimen was deposited in the SING

AC

herbarium (herbarium number: SING0215652). Preparation of crude extracts and fractions The extract and fractions were prepared by adapting the procedure used by Monji et al. (2016). Briefly, the freeze dried powder of edible part was extracted with 70% ethanol using sonication method. The crude extract was concentrated by a rotatory evaporator at 42 ± 1 oC and

5

ACCEPTED MANUSCRIPT

fractionated through a series of liquid-liquid partitions using hexane (F1), ethyl acetate (F2) and 1-butanol (F3). The aqueous fraction was labeled F4 which consistently demonstrated uterotonic activity.

CR IP T

Animals All the experimental procedures were executed in accordance with the international guideline for animal research under due approval from institutional animal care and committee, National University of Singapore (R12-0216). Virgin (~225g) and pregnant SD female rats (Day 21) were

AN US

acquired for the in vitro study. In non-pregnant rats, the stage of estrous cycle was examined by vaginal smear and those in the dioestrus stage were sacrificed at the time of the experiment. Pregnant SD rats (Day 16) were used for the in vivo telemetry study. Animals were allowed free

M

access to standard laboratory diet and tap water

ED

In vitro functional studies

Rats either pregnant or non-pregnant were euthanized in CO2 chambers. Uterine strips were

PT

prepared as described in a previous study (Monji et al., 2016) . Rat uterine strips were mounted in 10mL thermostatically controlled organ baths (37 ± 1 oC) containing Krebs-bicarbonate,

CE

aerated constantly with 95% O2 / 5% CO2 at pH 7.4. Following equilibration and contracting the strips by KCl (120mM) for normalization of data, cumulative concentrations of serotonin and F4

AC

were added into the bath. Immunoblot analysis To determine if F4 induces the uterine contractility through binding to the 5-HT2a receptor, late pregnant rat uterine smooth muscle was challenged with single dose serotonin (5 µM) or F4 (10

6

ACCEPTED MANUSCRIPT

mg.ml−1) for 5, 10 and 20 min in the presence and absence of 5-HT2a inhibitor, ketanserin (10 µM). Total protein was isolated from independent sets of uterine muscle following the functional studies. A disposable grinding pestle was used to grind and homogenize the sample in RIPA lysis buffer (25 mM Tris-HCl pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1%

CR IP T

SDS) containing Halt Protease & Phosphatase Inhibitor Cocktail (100X) (#1861281, Thermo Scientific, USA). Samples were centrifuged at 15000 g at 4 oC for 30 min, and the supernatants were collected. Protein concentrations were determined by using the Bradford assay (Bradford,

AN US

1976) with Pierce Coomassie Protein Assay Kit (#23200, Thermo Scientific, USA).

Protein samples (10 µg/lane) were separated under non-reducing conditions by using polyacrylamide gel electrophoresis in 10% resolving gels (Laemmli, 1970) and the gels were electroblotted to 0.2 µm PVDF membrane (#1620177, BIO-RAD, USA). Membranes were

M

rinsed in Phosphate Buffered Saline (pH 7.3-7.5) solution containing 137 mM NaCl, 2.7mM KCL and 10mM Phosphate buffer with 0.1% Tween-20 (PBST) for 5 min. Blots were blocked in

ED

5% BSA/PBST for 1h. The following antibodies were used: anti-Phospho-p44/42 ERK (#4376;

PT

Cell Signaling Technology,USA), anti-p44/42 ERK (#4695; Cell Signaling Technology, USA) and anti- MMP-13 (#SC-30073; Santa Cruz Biotechnology, USA) at dilutions of 1:1000 and

CE

were incubated with blots for overnight at 4oC. Blots were rinsed 3 times for 15 min in PBST the following day. Horseradish peroxidase (HRP)-linked goat anti-rabbit immunoglobulin G (IgG)

AC

(#7074; Cell Signaling Technology, USA) was used at dilutions of 1:2500 and incubated with blots for 1 h at room temperature. Blots were washed 3 times for 15 min in PBST. Proteins were detected using the Pierce ECL Western Blotting Substrate (#32209, Thermo Scientific, USA) and ECL Prime western blotting detection reagent (#PRN2232, GE Healthcare, UK).

7

ACCEPTED MANUSCRIPT

As per antibody data sheets, HeLa cell lysates were used as positive control for anti-ERK and anti-phospho ERK. For the anti-MMP-13, protein from postpartum uterine tissue was used as a positive control. HeLa cells were cultivated in monolayers in a 6-well cell culture plate with a seeding density of 0.5 ˟ 106 cells per well. Following 4 h serum starvation, the HeLa cells were

CR IP T

treated with 1 µg.ml−1 Epidermal Growth Factor (EGF) in serum free medium for 10 min to induce the ERK signaling pathway. Following the EGF stimulation, the reaction was stopped by replacing the medium with ice cold Dulbecco’s phosphate-buffered saline (DPBS) and protein was extracted using RIPA buffer as mentioned above. Following immunoblot analysis of the

AN US

abovementioned proteins, analysis of α-actin protein expression was performed. α-actin was constitutively expressed in pregnant rat myometrial tissue under our protein extraction

ED

Determination of gestational age

M

conditions.

To have accurate time mated pregnant rats for surgery, embryonic size based on the crown to

PT

rump length (CRL) was measured using ultrasound scanning and compared to a developed scale

CE

reported previously (Torres et al., 2008). Fig. 1 shows a representative measurement of CRL

AC

(9.925 mm) which confirmed the gestational age to be E13.

8

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN US

CR IP T

Fig.1. Ultrasound image of CRL measurement in millimeters for one fetus

9

ACCEPTED MANUSCRIPT

Surgical procedure The PhysioTel PA series transmitter model PA-C40 [Data Sciences International (DSI), St. Paul, MN] was implanted to measure intrauterine pressure. Before surgery, transmitters were cleaned

CR IP T

and sterilized using a solution of 1% enzyme detergent (Tergazyme TM, Alconox, USA) followed by 2% glutaraldehyde (Sigma-Aldrich, St. Louis, Missouri, U.S.A.). Rats were anesthetized using isoflurane 5% for induction and 1-2% for maintenance, with an oxygen flow rate of 1-2 L/min. The rat’s abdomen was shaved and cleaned. Using aseptic techniques, small

AN US

vertical midline incision (2-3cm) was made on the skin and underlying abdominal muscle. The uterus was taken gently out of the body cavity. A small hole was made using a 18G needle on the uterus (close to the first pup toward ovary side) and the catheter was guided gently between the uterine wall and fetal sac. Tissue adhesive was applied to the catheter insertion site. The uterus

M

was then placed back inside the body cavity. Transmitter body was sutured to the abdominal muscle and then skin sutured or closed using a staple. The animal was monitored after the

ED

surgery. Rats were injected subcutaneously with enrofloxacin (10 mg.kg−1) and carprofen (5

PT

mg.kg−1) post-surgery for 3 and 5 days respectively. The steps of the surgery in the pregnant

AC

CE

animal are shown in Fig. 2.

10

ACCEPTED MANUSCRIPT

Fig. 2. Surgical steps of intrauterine telemetry implantation. a and b: small vertical midline incision on the skin and abdominal muscle. c: pulling the uterus out from the abdominal activity. d: making a hole using a 18G needle. e and f: insertion of the catheter between the uterine wall and fetal sac. g: applying a tissue adhesive at the site of catheter insertion. h: placing the uterus

CR IP T

back into the body activity. i and j: returning the transmitter into the abdominal cavity and

AC

CE

PT

ED

M

AN US

suturing it to the abdominal muscle. k and l: suturing the skin or closing by a staple.

11

ACCEPTED MANUSCRIPT

Experimental protocol For in vitro studies, pregnant (E21) and non-pregnant rats at diestrous were euthanized in CO2 chambers and uterine tissue prepared using the same method that was detailed in our previous study (Monji et al., 2016). For in vivo study in a pregnant animal, the time of implantation was

CR IP T

optimized to be on E16 as implantation at this gestational age did not cause any damage to the pups and all the pups were alive after delivery. The A. comosus extracts as well as controls were administered in the late pregnant rat at E18 and also in the postpartum (Day 1) rats.

AN US

LC-HRMS analysis of the crude extract

The crude extract was analyzed using the liquid chromatography/high-resolution mass spectrometry (LC-HRMS). Five standards (citric acid, leucine, phenylalanine, serotonin and tyrosine) were weighed and reconstituted in ultrapure H2O to form the stock solution of 1.0

M

mg.ml−1. Working standards were prepared from the stock solutions by dilution with ultrapure H2O. Calibration curves were constructed for each standard at the following concentration range:

ED

phenylalanine and serotonin (0.125 to 2 µg.ml−1); leucine and tyrosine (0.625 to 10 µg.ml−1);

PT

citric acid (25 to 200 µg.ml−1). Ripe pineapple crude extract was weighed (3 samples) and dissolved in ultrapure H2O at 2 mg.ml−1. The samples were sonicated for 15 min, and centrifuged

CE

at 14000 for 5 min, prior to injection. The gradient was started at an initial composition of 1% B (0.1% formic acid in acetonitrile) for 0.5 min, to 2% B in 3 min, then 1 min linear gradient to

AC

100% B, held for 4 min, and return to the initial conditions. Peak area of the precursor ion was used for quantitation. Isolation and quantification of active principle in F4

12

ACCEPTED MANUSCRIPT

The F4 extract (300 mg) was dissolved in water and separated by preparative reversed-phase high-performance liquid chromatography (RP-HPLC) on a C18 column (19 ×f 100 mm, Waters Atlantis T3). After 6 min of isocratic elution at 100% A (0.1% formic acid in H2O), a linear gradient was run over the next 6 min to 15% B; then to 40% B in 8 min. Sub-fractions collected

CR IP T

were submitted for biological testing. Active fraction (43A) was analyzed by LC-HRMS. The HRESIMS and MS2 spectra were acquired on Agilent UHPLC 1290 Infinity coupled to Agilent 6540 accurate-mass quadrupole time-of-flight (QTOF) mass spectrometer equipped with a splitter and an ESI source. A linear gradient from 100% A to 50% B was run over a 15 min

AN US

period, then 1 min linear gradient to 100% B, held for 4 min, and return to the initial conditions. Quantification of serotonin was performed using the LC-HRMS. The purity of fraction 43A was approximately 65%.

M

Drugs and chemicals used

The following compounds were used in this study: ergometrine (1mg), a partial antagonist of

ED

serotonin and adrenergic receptors; mifepristone (15 mg.kg-1), an antiprogesterone and anti-

PT

glucocorticosteroid agent; serotonin (0.05-5µM), a neurotransmitter that binds to 5-HT receptors leading to increased contractility; ripe and unripe aqueous fraction (F4) (1.5 g.100 g-1 body

CE

weight, orally).

AC

Data acquisition and analysis In vitro uterine activity was quantified and analyzed as previously reported by Monji et al (2016). Densitometric analysis of immunoblots was performed using ImageJ software (National Institutes of Health, USA). Densitometric measurements of ERK, p-ERK and MMP-13 proteins on immunoblots were normalized to those of α-actin. Data from immunoblot analyses of at least

13

ACCEPTED MANUSCRIPT

five experiments using uterine muscle strips from different rats were plotted using GraphPad Prism (version 4.03; GraphPad Software Inc., San Diego, CA). The intrauterine pressure was measured by PowerLab software Chart (version 8.0.5). Continuous

CR IP T

measurement of pressure was initiated at E17, one day after surgery and continued through the time of delivery and post-delivery. The changes in intrauterine pressure were compared 12 and 24 h before and after delivery, by measuring the average of all data points in 1 h, calculated by Powerlab software Chart. The effect of uterotonic fraction and controls on uterine activity were

AN US

compared by measuring the intrauterine pressure 1 h before and after the intervention, which was the average of all the data points of every 5 min acquired within an h.

All data are expressed as means ± S.E. Means were compared using Student’s t test and ANOVA between two and three or more groups respectively. *p < 0.05, **p < 0.01, ***p <

M

0.001 and ****p < 0.0001 indicates statistically significant difference between the control and

ED

tested groups.

PT

Results

CE

Comparison of ripe F4 and serotonin induced contraction in non and late pregnant rat in vitro Uterine strips of non-pregnant and late-pregnant rats were exposed to cumulative addition of ripe

AC

F4 (0.1-10mg.ml-1) and serotonin (5-HT: 0.05-5 µM). The maximal contractile response of 5-HT and F4 increased significantly in a concentration-dependent manner at the end of pregnancy (p < 0.0001, and p < 0.001, Student’s t test, n = 6, respectively). However, the uterine sensitivity represented by pEC50 to 5-HT (Non-pregnant = -6.04 ± 0.24, Late-pregnant = -5.93 ± 0.16 µM) and ripe F4 (Non-pregnant = 3.85 ± 0.72, Late-pregnant = 3.21 ± 0.19 µgml-1) did not change significantly (p = 0.73 and p = 0.32, t test, respectively) (Fig. 3A,B). 14

ACCEPTED MANUSCRIPT

Fig. 3. Stimulation of contractions by 5-HT (0.05-5 µM) and F4 (0.1-10 mg.ml-1) in nonpregnant and late-pregnant rat uterine strips. The mean uterine activity following 5-HT (A) and

AN US

CR IP T

ripe F4 (B) increased significantly in late pregnant rat uterine strips.

ERK phosphorylation and MMP-13 expression following stimulation by serotonin and F4

M

Immunoblot analysis demonstrated that 5-HT, as well as F4, mediated ERK phosphorylation and consequently resulted in MMP-13 upregulation (Fig. 4A,B). 5-HT and F4 significantly elevated

ED

(ρ < 0.05, Student’s t test, n = 4) the expression of the above-mentioned proteins in the late pregnant uterine tissue in the absence of ketanserin. Increase in the expression of p-ERK and

PT

MMP-13 was detected following 5 min exposure to 5-HT and F4. Maximal phosphorylation and

CE

as a consequence MMP-13 was achieved within 10 min. Expression of proteins diminished at 20 min (Fig. 4C,D). A significant increase in p-ERK level was also observed after inducing the

AC

HeLa cells (positive control) with EGF, compared to the untreated HeLa cell lysates (Fig. 4A). Similarly, a significant increase in MMP-13 level was observed in the untreated postpartum (pp) tissue lysate (positive control for anti-MMP-13) (Fig. 4C).

15

ACCEPTED MANUSCRIPT

Fig. 4. Immunoblot analysis of p-ERK and MMP-13 protein expression in late pregnant tissue following the addition of 5-HT and F4 in the absence and the presence of ketanserin (10uM).

CR IP T

(A,B) Representative immunoblots p-ERK, MMP-13 and α-actin protein expression. (C,D) Densitometric analysis illustrating protein expression following administration of 5-HT and F4

AC

CE

PT

ED

M

AN US

suppressed in the presence of ketanserin, a 5-HT2a receptor specific antagonist.

16

ACCEPTED MANUSCRIPT

Evaluation of changes of intrauterine pressure during pregnancy and delivery Fig. 5A is a representative tracing of continuous 4-days recording of the intrauterine pressure during pregnancy, delivery and 1 day postpartum. This figure depicts the increasing trend of

CR IP T

intrauterine pressure which is an indicator of uterine contractility. The amplitude and frequency of IUP gradually increased over the course of pregnancy and delivery. A magnified tracing of the rat intrauterine pressure for 0.5 h between 11:00 and 11:30 am from E18 to E22 and during the delivery time showed a gradual increase in magnitude and number of contraction as well as the

AC

CE

PT

ED

M

AN US

basal tone as uterus underwent pregnancy and delivery (Fig. 5B).

17

ACCEPTED MANUSCRIPT

Fig. 5. Representative tracings of continuous recording of the intrauterine pressure during pregnancy and delivery. An increase in frequency and amplitude of intrauterine pressure peaks was observed over the course of pregnancy and delivery. a. A sample of recording from E20 to 1 day postpartum. White and black bars on top of recordings refer to 12 h light and dark cycles,

CR IP T

respectively. b. Magnified tracing of the rat intrauterine pressure for 6 consecutive days from

AC

CE

PT

ED

M

AN US

E18 to E22 and during delivery.

18

ACCEPTED MANUSCRIPT

To quantify the changes in intrauterine pressure during pregnancy and delivery, the hly averages of pressure were calculated at 12 h time point when delivery occurred and compared with 12 h before this cycle. The overall hly average of the IUP during the cycle of delivery increased gradually from 3.7 ± 0.1 to 10.1 ± 1.0 mmHg and reached the highest level (17.2 mmHg) at

CR IP T

the time of delivery which is around 4 folds of basal IUP (ρ < 0.05, Student’s t test, n = 10), as shown in Fig. 6A. Extending the measurement of IUP to 24 h before and after the delivery also showed a gradual increasing trend in 24h. In contrast, 24 h continuous recording of IUP after the delivery showed a significant drop in IUP after delivery that continued from 6 to 12 h. After 12 h

AN US

postpartum, the intrauterine pressure gradually rose and spiked close to that of the delivery time

AC

CE

PT

ED

M

(Fig. 6B).

19

ACCEPTED MANUSCRIPT

Fig. 6. Overview of intrauterine pressure (hly average) changes during pregnancy, delivery and post-delivery. (A) A gradual increase of IUP in a 12 h delivery period in comparison with 12 h before this cycle. A significant increase was observed at the time of delivery (ρ < 0.05, Student’s

CR IP T

t test, n = 10). (B) Changes of IUP 24 before and after the delivery time showing the gentle rise

AC

CE

PT

ED

M

AN US

in IUP before delivery followed by a significant drop after delivery.

20

ACCEPTED MANUSCRIPT

Evaluation of IUP in response to F4 (ripe and unripe), vehicle, ergometrine and mifepristone in pregnant rats at E18 As shown in Fig. 7A,B oral administration of ripe F4 (1.5g.100g−1 body weight) did not stimulate the uterine contraction when compared to the basal activity and vehicle control.

CR IP T

Although there was a slight increase in uterine motility following the oral administration of ergometrine (1mg), no significant changes were detected (Fig. 7C). Similar to ripe F4, unripe F4 did not affect the IUP following oral administration during pregnancy at E18 (data not presented here). Mifepristone (15mg.kg−1) significantly increased the frequency of intrauterine pressure

AN US

peaks at about 20 h post subcutaneous injection, which caused premature labor in SD rats (Fig.

AC

CE

PT

ED

M

7D,E).

21

ACCEPTED MANUSCRIPT

Fig. 7. Intrauterine pressure (time average) in response to oral F4, vehicle and positive controls. (A,B) F4 (1.5g.100g−1 body weight) did not increase the IUP over the course of pregnancy, which was similar to the vehicle (5 ml). (C) Ergometrine (1mg) increased the IUP slightly (ρ > 0.05, Student’s t test, n = 6). (D) Mifepristone (15 mg.kg−1) increased the IUP at around 20 h

CR IP T

after SC injection which resulted in premature labor at E19. (E) Representative tracing of continuous recording of IUP during preterm birth following injection of mifepristone showed

AC

CE

PT

ED

M

AN US

low amplitude and high frequency of IUP spikes.

22

ACCEPTED MANUSCRIPT

Evaluation of IUP in response to oral F4 (ripe and unripe), vehicle and ergometrine in postpartum rats Ripe F4 (1.5g.100g−1 body weight) and the positive control, ergometrine (1mg) showed significant increase in maximal pressure (15 min after oral administration) to 4.3 and 4.9 folds of

CR IP T

the basal IUP, respectively which depicts uterotonic effect of ripe F4 and ergometrine in postpartum rats, while oral administration of vehicle (water) did not affect IUP in comparison to the baseline activity. Unripe F4 (1.5g.100g−1 body weight) did not cause a significant change in

AC

CE

PT

ED

M

AN US

IUP compared to the baseline activity in postpartum rats (Fig. 8A-E).

23

ACCEPTED MANUSCRIPT

Fig. 8. Intrauterine pressure (5 min average) in response to oral ripe and unripe F4, vehicle, ergometrine and (A-D) IUP rose remarkably after oral administration of F4 and ergometrine (ρ < 0.05, Student’s t test, n = 8) while vehicle did not contribute to the uterotonic effect. (E)

AC

CE

PT

ED

M

AN US

CR IP T

Frequency and amplitude of IUP peaks increased in response to ripe 4 and ergometrine.

24

ACCEPTED MANUSCRIPT

LC-HRMS analysis Citric acid, serotonin and three major amino acids in the crude extract were identified by comparing the retention time and HRMS data with those of reference standard compounds.

CR IP T

Representative extracted ion chromatograms (EIC) of m/z of standards in reference solution and crude extract were shown in Fig. 9A-D.

Fig. 9. Illustration of EIC overlay of citric acid, tyrosine, serotonin, phenylalanine and leucine in a reference solution and crude extract. The EIC overlay of all tested standard compounds was

PT

ED

M

AN US

plotted as retention time in min against a function of intensity in counts per seconds.

Isolation and identification of serotonin in F4 by bioassay guided fractionation

CE

Twenty one fractions were generated from F4. An active fraction was identified by its ability to produce significant uterotonic effects in late pregnant rat uterine tissue in vitro. Active fraction

AC

(43A) eluted at 14.4 min after the injection (Fig. 10). It was analyzed by LC-HRMS. The active compound was identified by comparing its retention time and mass data with those of a serotonin standard.

Fig. 10. Prep HPLC chromatogram (254 nm) of the aqueous extract F4 of ripe A. comosus

25

ACCEPTED MANUSCRIPT

Fig. 11A showed a representative total ion chromatogram (TIC) of the active fraction, suggesting that there was only one major compound in the active fraction. The same retention times and shapes indicated that the compound in active fraction was similar to serotonin. Positive HRMS data showed an m/z of 177.1030 which was the same as serotonin (Fig. 11B). MS/MS spectra of

CR IP T

the active fraction and reference serotonin showed that fragmentation pattern of the active fraction was in excellent agreement with pure serotonin and the most abundant ion was at m/z 160, which involved the neutral loss of NH3 (Fig. 11C). The concentration of 5-HT obtained from the ripe and unripe F4 was estimated to be about 0.56 µmole.mg−1 and 0.08 µmole.mg−1,

M

AN US

respectively.

ED

Fig. 11. Total ion current (TIC) plot (A), MS (B) and MS2 (C) spectra of the reference serotonin

AC

CE

PT

and isolated compound from the active fraction.

26

ACCEPTED MANUSCRIPT

Discussion Our previous in vitro studies established the role of 5-HT receptors in stimulating uterine motility in late pregnant SD rats by aqueous fraction of Ananas comosus (L.) Merr. (Monji et

CR IP T

al., 2016). To further investigate the involvement of the 5-HT receptors in uterotonic activity of A. comosus, the effect of F4 and serotonin was compared in non-pregnant and late pregnant rats in vitro. A similar increasing trend in contractile response by F4 and serotonin in late pregnant rat uterine tissue without significant difference in potency was observed; this could be attributed

AN US

to the upregulation of 5-HT receptors as the pEC50 to 5-HT and F4 did not change significantly. These results confirm the association between the stage of pregnancy and density of 5-HT2a receptors which is in agreement with the earlier study (Minosyan et al., 2007) and also suggest that A. comosus may help in inducing labor.

M

It has been shown that serotonin can initiate a signal transduction pathway via 5HT2a receptor,

ED

which results in increased expression of MMP-13 through ERK phosphorylation. MMP-13 expression occurs during postpartum to contribute in tissue remodeling post-delivery while it has

PT

a low expression in pregnant uterine tissue (Shum et al., 2002). Immunoblot analysis demonstrated that F4, similar to serotonin, increased ERK phosphorylation and as a consequence

CE

increased MMP-13 expression in late pregnant rat uterine tissue. Furthermore, pretreatment of

AC

uterine tissue with ketanserin, a selective 5-HT2a receptor antagonist, suppressed the expression of p-ERK and final product MMP-13 significantly. Thus, this study confirmed that F4 could functionally replicate 5-HT property in uterine tissue. To validate the in vitro findings, telemetry study was designed and optimized to determine the uterotonic effect of aqueous fraction (F4) of A. comosus extract in freely moving rats. This

27

ACCEPTED MANUSCRIPT

model successfully tracked the changes in intrauterine pressure over the course of pregnancy and postpartum in SD rats. The hly average of the intrauterine pressure was elevated about four folds at the time of delivery,

CR IP T

which is in line with the earlier study in mice, showing a threefold increase in maximal IUP at the time of delivery (Pierce et al., 2010). Subcutaneous injection of mifepristone resulted in prolonged delivery after around 20 h. This finding is consistent with the previous report showing vaginal bleeding as well as prolonged delivery of fetuses following 24 h subcutaneous injection

AN US

of mifepristone (Garfield et al., 1987). However, uterine motility during premature labor following mifepristone was quite different from natural delivery, being associated with prolonged low amplitude and high frequency of IUP spikes.

In pregnant rats (E18), ripe and unripe F4, ergometrine and vehicle did not stimulate the uterine

M

contraction following oral administration. Our observation of the lack of efficacy of ergometrine

ED

to produce a significant increase in IUP in late pregnant rats is in line with the previous studies that oral route is not effective in the active management of the third stage of labor (Liabsuetrakul

PT

et al., 2007).

CE

A strong relationship between the level of estradiol and sensitivity to 5-HT has been reported; it was shown that 5-HT receptor was up-regulated either during estrus phase or by administration

AC

of estradiol in rats (Ichida et al., 1984). A possible explanation for the lack of efficacy of the aqueous fraction in pregnant rat model is probably due to the low level of estradiol and as a consequence low 5-HT receptors at the time of administration. The level of estradiol in pregnant rats began to increase towards the end of pregnancy after Day 20 of pregnancy with a significant rise on Day 21. It increased further at the time of parturition (Puri and Garfield, 1982). However,

28

ACCEPTED MANUSCRIPT

in human estradiol increases gradually during pregnancy (Hendrick et al., 1998). Hence, oral consumption of A. comosus and in general serotonin-enriched foods and supplements may elevate the peripheral serotonin and produce responses through the receptors present in various tissues such as uterus in view of recent reports on rapid and prolonged increase in serum 5-HT

CR IP T

following oral administration of 5-HT in mice (Islam et al., 2016) and rabbits (Lancellotti et al., 2015). The levels of 5-hydroxyindoleacetic acid (the metabolite of serotonin) and serotonin have also been reported to be higher in women with a history of recurrent miscarriage compared to the

AN US

normal pregnant women (Sadovsky et al., 1972).

Oral administration of ripe F4 as well as ergometrine produced significant uterotonic response during 24 h postpartum. This result may be explained by the recurring estrous postpartum and high level of estradiol during the first 24 h following parturition (Carrillo-Martínez et al., 2011)

M

suggesting that the active principle in F4 is a 5-HT like compound. The contribution of steroid

ED

hormones in the modulation of uterine motility has been established in vitro (Dodds et al., 2015) and in vivo (Downing et al., 1981) in mice and rats.

PT

Folklore beliefs highlight the abortifacient properties of unripe A. comosus rather than the ripe ones (Burkill, 1966; Rahman, 2014; Simpson, 1962). However, our study demonstrated that the

AC

in vivo.

CE

unripe F4 did not produce any significant change in IUP either during pregnancy or postpartum

Citric acid, serotonin and three major amino acids, including tyrosine, phenylalanine and leucine were identified using LC- HRMS analysis in the crude extract. Further chemical analysis through bioassay guided fractionation in this study identified serotonin as a major bioactive compound in F4 which contributed to the uterotonic activity of A. comosus extracts in vitro and in vivo.

29

ACCEPTED MANUSCRIPT

Quantification of the higher amount of serotonin in ripe F4 compared to the unripe one supported the in vivo data that ripe F4 was more potent than unripe F4 in stimulating uterine contractions. The purpose of the current study was to investigate the in vitro molecular mechanism as well as

CR IP T

the in vivo uterotonic effect of the aqueous fraction of Ananas Comosus extract in SD rat model. Taken together, the findings of this investigation complement those of the earlier study of the contractile activity of A. comosus on the uterus in vitro (Monji et al., 2016). The strengths of the present study were optimizing the in vivo model to evaluate the effect of A. comosus extract and

AN US

identifying molecular mechanism and the active compound mediating the bioactivity of the uterotonic fraction in vitro and in vivo. This study offers some scientific merits into the folkloric beliefs regarding the abortifacient effect of A. comosus.

Since the study was limited to evaluation of pharmacological properties, safety of consumption

M

of A. comosus during pregnancy was not addressed in this study as it is beyond the scope and

ED

logistics of this study. Being limited to SD rats, it was not possible to mimic the exact hormonal changes in larger animal or human during pregnancy. Further studies are required to clarify how

PT

sex hormones during pregnancy and estrous cycle could affect the A. comosus induced

CE

contraction. Our optimized model in SD rat would be ideal for further work. Author contributions

AC

All authors contributed extensively to the work presented in this paper P Ganesan Adaikan: supervised the project, contributed to conception and design Faezeh Monji: designed and performed the experiments, analyzed and interpreted the data Lang Chu Lau: designed the experiments 30

ACCEPTED MANUSCRIPT

Abrar Al-Mahmood Siddiquee and Baharudin Bin Said: performed the experiments Yoganathan S/O Kanagasundaram and Yang Lay Kien: standardized the extract and isolate the active compounds

CR IP T

Mahesh Choolani: Advised the clinical perspective Acknowledgments

The authors wish to thank neuroscience phenotyping core (NUS) for sharing the facility of

AN US

telemetry recording Conflict of interest

AC

CE

PT

ED

M

The authors have declared that there is no conflict of interest.

31

ACCEPTED MANUSCRIPT

References

AC

CE

PT

ED

M

AN US

CR IP T

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72(1-2), 248-254. Burkill, I.H., 1966. A dictionary of the economic products of the Malay Peninsula. A Dictionary of the Economic Products of the Malay Peninsula. 2(2nd edition). Carrillo-Martínez, G.E., Gómora-Arrati, P., González-Arenas, A., Morimoto, S., Camacho-Arroyo, I., González-Flores, O., 2011. Role of progesterone receptors during postpartum estrus in rats. Horm. Behav. 59(1), 37-43. Cordeaux, Y., Pasupathy, D., Bacon, J., Charnock-Jones, D.S., Smith, G.C., 2009. Characterization of serotonin receptors in pregnant human myometrium. J. Pharmacol. Exp. Ther. 328(3), 682-691. De Laszlo, H., Henshaw, P.S., 1954. Plant materials used by primitive peoples to affect fertility. Science 119, 626-631. Dodds, K.N., Staikopoulos, V., Beckett, E.A., 2015. Uterine contractility in the nonpregnant mouse: changes during the estrous cycle and effects of chloride channel blockade. Biol. Reprod. 92(6), 141, 141115. Downing, S.J., Porter, D., Redstone, C., 1981. Myometrial activity in rats during the oestrous cycle and pseudopregnancy: interaction of oestradiol and progesterone. The Journal of physiology 317(1), 425433. Garfield, R.E., Gasc, J.M., Baulieu, E.E., 1987. Effects of the antiprogesterone RU 486 on preterm birth in the rat. Am. J. Obstet. Gynecol. 157(5), 1281-1285. Hendrick, V., Altshuler, L.L., Suri, R., 1998. Hormonal changes in the postpartum and implications for postpartum depression. Psychosomatics 39(2), 93-101. Ichida, S., Oda, Y., Tokunaga, H., Hayashi, T., Murakami, T., Kita, T., 1984. Mechanisms of specific change by estradiol in sensitivity of rat uterus to serotonin. J. Pharmacol. Exp. Ther. 229(1), 244-249. Islam, J., Shirakawa, H., Aso, H., Komai, M., 2016. Measurement of serotonin distribution and 5hydroxyindoleacetic acid excretion after oral administration of serotonin using hplc fluorescence detection. Food Sci. Nutr. Technol. 1, 1–7. Kamboj, V., 1988. A review of Indian medicinal plants with interceptive activity. The Indian journal of medical research 87, 336. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Lancellotti, P., Nchimi, A., Hego, A., Dulgheru, R., Delvenne, P., Drion, P., Oury, C., 2015. High-dose oral intake of serotonin induces valvular heart disease in rabbits. Int. J. Cardiol. 197, 72-75. Liabsuetrakul, T., Choobun, T., Peeyananjarassri, K., Islam, Q.M., 2007. Prophylactic use of ergot alkaloids in the third stage of labour. Cochrane Database Syst. Rev.(2), CD005456. Lim, T., 2012. Ananas comosus, Edible Medicinal and Non-Medicinal Plants. Springer, pp. 593-615. Masson, J., Emerit, M.B., Hamon, M., Darmon, M., 2012. Serotonergic signaling: multiple effectors and pleiotropic effects. Wiley Interdisciplinary Reviews: Membrane Transport and Signaling 1(6), 685-713. Minosyan, T.Y., Lu, R., Eghbali, M., Toro, L., Stefani, E., 2007. Increased 5-HT contractile response in late pregnant rat myometrium is associated with a higher density of 5-HT2A receptors. J. Physiol. 581(Pt 1), 91-97. Monji, F., Adaikan, P.G., Lau, L.C., Said, B.B., Gong, Y., Tan, H.M., Choolani, M., 2016. Investigation of uterotonic properties of Ananas comosus extracts. J. Ethnopharmacol. 193, 21-29. Pierce, S.L., Kutschke, W., Cabeza, R., England, S.K., 2010. In vivo measurement of intrauterine pressure by telemetry: a new approach for studying parturition in mouse models. Physiol. Genomics 42(2), 310316. 32

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN US

CR IP T

Puri, C., Garfield, R.E., 1982. Changes in hormone levels and gap junctions in the rat uterus during pregnancy and parturition. Biol. Reprod. 27(4), 967-975. Quisumbing, E., 1951. Medicinal plants of the Philippines. Department of Agriculture and Commerce, Philippine Islands Technical Bulletin.(16). Rahman, A.M., 2014. Ethno-gynecological study of traditional medicinal plants used by Santals of Joypurhat district, Bangladesh. Biomedicine and Biotechnology 2(1), 10-13. Sadovsky, E., Pfeifer, Y., Polishuk, W., Sulman, F., 1972. The use of antiserotonin-cyproheptadine HCl in pregnancy: An experimental and clinical study, Drugs and fetal development. Springer, pp. 399-405. Shum, J.K., Melendez, J.A., Jeffrey, J.J., 2002. Serotonin-induced MMP-13 production is mediated via phospholipase C, protein kinase C, and ERK1/2 in rat uterine smooth muscle cells. J. Biol. Chem. 277(45), 42830-42840. Simpson, G.E., 1962. Folk medicine in Trinidad. The Journal of American Folklore 75(298), 326-340. Singh, Y.N., 1986. Traditional medicine in Fiji: some herbal folk cures used by Fiji Indians. J. Ethnopharmacol. 15(1), 57-88. Torres, E., Weyrauch, U., Sutcliffe, R., Dunnett, S., 2008. A rat embryo staging scale for the generation of donor tissue for neural transplantation. Cell Transplant. 17(5), 535-542.

33

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN US

CR IP T

Graphical Abstract

34