hum. reprod.-2014-green-humrep-deu309.pdf

7/25/2019 Hum. Reprod.-2014-Green-humrep-deu309.pdf

1/15

ORIGINAL ARTICLEEmbryology

Insulin-like growth factor 1 increases

apical fibronectin in blastocysts

to increase blastocyst attachment

to endometrial epithelial cells in vitro

Charmaine J. Green1, Stuart T. Fraser1,2, and Margot L. Day1,*1Disciplineof Physiology, BoschInstitute, SydneyMedicalSchool, Universityof Sydney, K25 Medical FoundationBuilding, Sydney2006, Australia2Discipline of Anatomy and Histology, SydneyMedical School, University of Sydney, K25 Medical Foundation Building, Sydney 2006, Australia

*Correspondence address. Tel: +61-2-9036-3312; Fax: +61-2-9036-3316; E-mail: [email protected]

Submitted on June 23, 2014; resubmitted on October 14, 2014; accepted on October 28, 2014

study question: Does insulin-like growth factor 1 (IGF1) increase adhesion competency of blastocysts to increase attachment to uterineepithelial cellsin vitro?

summaryanswer: IGF1increases apical fibronectin on blastocysts to increaseattachmentand invasion in an in vitro modelof implantation.

what is known already:Fibronectin integrin interactions are important in attachment of blastocysts to uterine epithelial cells atimplantation.

study design, size, duration: Mouse blastocysts (hatched or near completion of hatching) were cultured in serum starved (SS)medium with varying treatments for 24, 48 or 72 h. Treatments included 10 ng/ml IGF1 in the presence or absence of the PI3 kinase inhibitor

LY294002, an IGF1 receptor (IGF1R) neutralizing antibody or fibronectin. Effects of treatments on blastocysts were measured by attachment

of blastocysts to Ishikawa cells, blastocyst outgrowth and fibronectin and focal adhesion kinase (FAK) localization and expression. Blastocysts

were randomly allocated into control and treatment groups and experiments were repeated a minimum of three times with varying numbers

of blastocystsused in each experiment. FAK and integrin protein expression on Ishikawacells wasquantified in the presenceor absence of IGF1.

participants/materials, setting, methods: Fibronectin expression and localization in blastocysts was studied using im-munofluorescence and confocal microscopy. Global surface expression of integrin avb3, b3 and b1 was measured in Ishikawa cells using flow

cytometry. Expression levels of phosphorylated FAK and total FAK were measured in Ishikawa cells and blastocysts by western blot and

image J analysis. Blastocyst outgrowth was quantified using image J analysis.

main results and the role of chance:The presence of IGF1 significantly increased mouse blastocyst attachment to Ishikawacells compared with SS conditions (P, 0.01). IGF1treatment resulted in distinct apical fibronectin staining on blastocysts, which was reduced by

the PI3 kinase inhibitor LY294002. This coincided with a significant increase in blastocyst outgrowth in the presence of IGF1 (P, 0.01) or fibro-

nectin(P, 0.001), which was abolished by LY294002 (P, 0.001). Apical expression of integrinavb3,b3andb1 in Ishikawacellswas unaltered

by IGF1. However, IGF1 increased phosphorylated FAK (P, 0.05) and total FAKexpression in Ishikawa cells. FAK signalling is linked to integrin

activationand can affect the integrins ability to bind and recognizeextracellular matrix proteins suchas fibronectin. Treatment of blastocysts with

IGF1 before co-culture with Ishikawa cells increased their attachment ( P, 0.05). This effect was abolished in the presence of LY294002 (P,

0.001) or an IGF1R neutralizing antibody (P,

0.05).limitations, reasons for caution: This studyuses an invitro modelof attachment thatuses mouseblastocystsand human endo-metrial cells. This involves a species crossover and although this use has been well documented as a model for attachment (as human embryo

numbers are limited) the results should be interpreted carefully.

wider implications of the findings: This study presents mechanisms by which IGF1improves attachment of blastocysts to Ishi-kawa cells and documents forthe first time howIGF1 canincrease adhesion competency in blastocysts. Failure of the blastocyst to implant is the

major cause of human assisted reproductive technology (ART) failure. As growth factors are absent during embryo culture, their addition to

embryo culture medium is a potential avenue to improve IVF success. In particular, IGF1 could prove to be a potential treatment for blastocysts

before transfer to the uterus in an ART setting.

& The Author 2014. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.

For Permissions, please email: [email protected]

Human Reproduction, Vol.0, No.0 pp. 1 15, 2014

doi:10.1093/humrep/deu309

Hum. Reprod. Advance Access published November 28, 2014


2/15

study funding/competing interest(s): This work was supported by internal funds from the University of Sydney and theBosch Institute. None of the authors has any conflict of interest to declare.

trial registration number: N/A.

Key words: PI3 kinase / fibronectin / implantation / insulin-like growth factor 1 / integrin

Introduction

Unsuccessful embryo implantation is the major cause of human assisted

reproductive technology (ART) failure (Milleret al., 2012). Implantation

is a highly coordinated event in which the receptive endometrium is

primed to receiveadhesion-competentblastocysts.Growthfactorspro-

duced by boththe embryo and the female reproductivetract are thought

to support development of the blastocyst to an adhesion-competent

state, which is synchronizedwith uterine receptivity,to ensure blastocyst

implantation ability (Wang et al., 2000;Armant, 2005). Therefore, an

understanding of how growth factors influence implantation is essential

to improve ART.

Insulin-likegrowthfactor1(IGF1)isoneparticulargrowthfactorthatispresent at the maternalembryo interface in a number of mammalian

species (Murphyet al., 1987;Kapuret al., 1992;Lighten et al., 1998;

Slater and Murphy, 1999). At the time of implantation in the rat IGF1 is

strongly expressed in the basal lamina and the apical surface of uterine

epithelial cells, which are the sites of trophoblast invasion and attach-

ment, respectively (Slater and Murphy, 1999). Additionally IGF1 is

secreted by mouse embryos (Inzunzaet al., 2010). The IGF1 receptor

(IGF1R)is expressedthroughoutmouse preimplantationstagesof devel-

opment (Inzunza et al., 2010) and is expressed apically on trophoblast

cells (Bedzhovet al., 2012), which are the cells that make first contact

with the uterine epithelium. High levels of IGF1, which cause IGF1R

down-regulation, decrease normal blastocyst implantation sites and in-

crease resorption sites (Pintoet al., 2002). Additionally, beads soakedinIGF1elicita decidualresponsein theuterusthat mimicsthatofa blasto-

cyst (Pariaet al., 2001). Despite this knowledge of IGF1/IGF1R expres-

sion in the embryo and uterine tissue, it is not known whether IGF1

affects blastocyst adhesion competency or the mechanisms by which

IGF1 influences early implantation.

Acquisition of adhesion competence by the blastocyst involves an

accumulation of integrins on the apical surface of trophoblast cells. Im-

plantation depends, in part, on the bridging between integrins on the

endometrium and the blastocyst by extracellular matrix (ECM) proteins

suchas fibronectin (Kaneko etal.,2013). Integrina5b1 subunitsincrease

on theapical surface of adhesion competentmouseblastocysts, andthis

correlates withan increasedability to bindfibronectin (Schultz andArmant,

1995;Schultzet al., 1997;Wang et al., 2002). Most ECM proteins bind to

multiple integrins, for example fibronectin, which binds the a5b1,avb3

and aIIb3 integrins in mouse blastocysts (Sutherlandet al., 1993;Schultz

and Armant 1995;Yelianet al., 1995;Schultzet al., 1997).

The process of integrin activation, which occurs via inside-out signal-

ling, regulates integrin affinity for ECM proteins. This inside-out signalling

is mediated by a number of signalling pathways, such as the PI3 kinase

(PI3K)/Akt pathway (Somanathet al., 2007). Akt is activated by IGF1

in mouse blastocysts (Green and Day, 2013) and treatment of tropho-

blast cells with IGF1 increases fibronectin binding mediated by a5b1

and avb3(Kabir-Salmaniet al., 2002,2003,2004).

The current study investigated the role of IGF1 in attachment, using a

well-characterizedin vitromodel of attachment that involves culture of

blastocysts on Ishikawacells.Ishikawacellsare a welldifferentiated endo-

metrial adenocarcinoma cell line (Nishida et al., 1985;Hannan et al.,

2010) that displays apical adhesiveness (Heneweeret al., 2005) and a

similar integrin expression profile to a receptive endometrium, under

the control of estrogen and progesterone (Lesseyet al., 1996;Castel-

baumet al., 1997). Embryos of different species including mouse, rat

and human (Singhet al., 2010;Kanekoet al., 2011a,2012,2013;Kang

et al., 2014) are known to attach to Ishikawa cells. These characteristics

make Ishikawa cells an invaluable model for the study of implantationin

vitro, especially as the use of human embryos is limited by availability

(Kang et al., 2014). In the present study we demonstrate that IGF1increased apical fibronectin expression on blastocysts and attachment

of blastocysts to Ishikawa cells, as well as blastocyst invasiveness.

These actions of IGF1 were all prevented by inhibition of the PI3K/Akt

pathway. Thesedata suggest an importantrole for IGF1in the acquisition

of blastocyst adhesion competence due to regulation of fibronectin ex-

pression. Furthermore, our results indicate that addition of IGF1 to the

culture system may improve the success of human assisted reproduction

by improving the adhesion competence of blastocysts.

Materials and Methods

Use of animals and ethical approval, ovulationinduction and embryo collection and culture

Procedures involving the use of animals were conducted in accordance with

the Australian Code of Practice for Use of Animals in Research and were

approved by the University of Sydney Animal Care and Ethics Committee.

The Quackenbush Swiss (QS) strain of mice was used (Animal Resource

Centre, Perth, Australia). Mice were housed under a 12-h light: 12-h dark

cycle (light; 06:0018:00), and housed in a conventional holding facility

where the temperature was maintained at 20228C and the humidity

kept at 55%. Mice had free access to water and commercially prepared

pellets. Female mice were caged in groups of 10 whilst male mice were

caged singly. Ovulation induction of female mice (410 weeks old) was

achieved by intraperitoneal injection of 10 I.U. of pregnant mare serum

gonadotrophin (PMSG; Intervet, Sydney, Australia) followed 48 h later by

intraperitoneal injection of 10 I.U. of hCG (Intervet) as previously described

(Nasr-Esfahani et al., 1990). Superovulated female mice were then paired

with a stud male QS mouse (1030 weeks old) overnight. The presence of

a vaginal plug the following day indicated successful mating and this was con-

sidered to be Day 1 of pregnancy.

Female mice were sacrificed on Day 4 of pregnancy, 90 94 h after hCG

administration, in order to recover early blastocysts. The oviducts and

uterine horns were flushed with Hepes-buffered modified synthetic human

tubal fluid medium (Hepes mod-HTF; 300 mosM/l, pH 7.4) containing

0.3 mg/ml bovine serum albumin (BSA; Sigma-Aldrich; St Louis, MO,

USA). Blastocysts were then collected and cultured for 2428 h, to enable

hatching from the zona pellucida (Day 5 Blastocysts), in Potassium simplex

2 Greenet al.


3/15

optimized medium (KSOM; 260 mosM/l, pH 7.4) containing 0.3 mg/ml

BSA. All media was made up from stock solutions prepared from tissue

culture grade reagents (Sigma-Aldrich) as previously described (Green and

Day, 2013). All blastocysts used in this study were collected on Day 4 and

allowed to hatch overnight before treatment (all blastocysts used were

hatched or near completion of hatching).

Recombinant mouse IGF1 (R&D Systems; Minneapolis, MN USA) was

reconstituted at 100mg/ml in sterile phosphate-buffered saline (PBS;

AMRESCO; Solon, OH, USA). Mature mouse IGF1 shares 94% amino acidsequence homology with human IGF1 and exhibits species cross reactivity

(Bell et al., 1986). Specifically, R&D Systems test their mouse IGF1 activity

on human MCF-7 breast cancer cell proliferation. Thus mouse IGF1 was

used in allof theexperiments.IGF1at 10 ng/ml hasbeenusedto seepositive

effects on embryo development (Green and Day, 2013). Higher concentra-

tions, such as 30, 100 and 1000 ng/ml IGF1, are known to have detrimental

effects on blastocysts and cause IGF1R down-regulation (Chi et al., 2000;

Green and Day, 2013), and this range of IGF1 concentrations has been

used in this study to investigate the mechanisms of IGF1 in attachment of

the embryo to uterine epithelial cells.

To determinea directeffect of IGF1 through theIGF1R, an IGF1R neutral-

izing antibody (IGF1RnAb; AF-305-NA; R&DSystems)was reconstituted at

0.2 mg/ml in PBS and used at a final concentration of 2 mg/ml.

LY294002 is a specific inhibitor of PI3K with an IC50 of 1.40 mM (Vlahoset al., 1994). Previously LY294002 has been used at concentrations of

500mM to induce apoptosis in blastocysts and although 250 mM did not in-

crease apoptosis in blastocysts (Riley et al., 2006), this concentration was

shown to decrease blastocyst hatching (Riley et al., 2005). In the present

study we chose to use a 5 mM LY294002, a concentration much closer to

the IC50 for the drug.

Ishikawa cell culture

The Ishikawa cell line was a gift from Professor Chris Murphy, University of

Sydney. Cells were thawed and plated on 10 mm glass coverslips (Menzel

Glaser; Braunschweig, Germany) in 24-well plates (Corning, NY, USA) for

co-culture experiments or directly on to 6 well pates (Corning, NY, USA)

for western blot analysis with Dulbeccos modified Eagles medium

(DMEM; GIBCO, Grand Island, NY, USA) containing 10% (v/v) fetal

bovine serum (FBS; Bovogen Biologicals Pty Ltd, Essendon, VIC, Australia)

and 5000 I.U. of penicillin and 5000 mg of streptomycin per ml (Invitrogen,

Camarillo, CA, USA).

Co-culture of mouse blastocysts with Ishikawa

cells in the presence and absence of IGF1

Attachmentexperiment 1 (Fig. 1) wasperformedwithco-culture ofIshikawa

cells and blastocysts in the presence and absence of IGF1. Ishikawa cells that

werenear confluencewere culturedin eitherserum medium(containing 10%FBS) or in serum starved (SS) medium (0.01% FBS) for 2 h prior to receiving

Figure 1 Timeline of experiments used to study attachment of mouse blastocysts to human Ishikawa cellsin vitro. BL: blastocyst.

IGF1 Increases mouse blastocyst attachmentin vitro 3


4/15

blastocysts.After2 h, SS Ishikawa cells were then cultured in thepresenceor

absence of 10 ng/ml IGF1. Blastocysts were isolated from six mice in each

experiment. Between 4 and 12 Day 5 blastocysts were transferred per well

onto theIshikawa cellsin each ofthe treatmentgroups.Blastocystswereran-

domly allocated to treatment groupsand treatment groupswere performed

in duplicate. Initially, co-cultures were incubated undisturbed for 24 h and

checked for attachment. However as no blastocysts had adhered at 24 h

(up to Day 6), co-cultures were incubated undisturbed for 48 h (up to Day 7).

Blastocyst attachment to Ishikawa cells was determined at 48 h by blowingmedia with a glass mouth pipette at the blastocysts whilst examining their

position under a dissecting microscope (Wild M3 microscope; Leica Micro-

systems, Wetlar, Germany). Embryos that did not float away were consid-

ered to have attached (Singh et al., 2010). Co-culture experiments were

performed in duplicate and repeated three separate times. Data were

pooled from thethree separate experiments and counted as theproportion

of blastocyststhat had attached to the Ishikawacells outof the total number

ofblastocystsadded tothe Ishikawacells. Between 14and 19 mice were used

for each experimental group.

Pretreatment of blastocysts before co-culture

with Ishikawa cells

Attachmentexperiment 2 (Fig. 1) was performed by pre-treating blastocystsbefore their co-culture with Ishikawa cells. Blastocysts were isolated from

between 5 and10 mice in each experiment. Day 5 blastocystswere cultured

in the presence of serum, with or without 2 mg/ml IGF1R nAb or in SS

medium in the presence or absence of 10, 100 or 1000 ng/ml IGF1 or

with 10 ng/ml IGF1 in the presence of 2mg/ml IGF1R nAb or 5mM

LY294002 (PI3 kinaseinhibitor; Selleck, Houston,TX, USA) for24 h. Blasto-

cysts (now day 6 blastocysts) were then washed and cultured on Ishikawa

cells for 24 h in SS medium and assessed for attachment as above. Experi-

ments were repeated three to nine times and the data pooled to calculate

the proportion of attached embryos for each treatment.

Immunofluorescence

Co-cultures, from attachment experiment 1 were examined for total focal

adhesion kinase (FAK) protein expression. Blastocysts were isolated from

between three andfive mice in each experiment. Day5 blastocystswerecul-

tured in the absence of Ishikawa cells at lowembryodensity in SS medium in

the presence or absence of 10 ng/ml IGF1 or 5 mM LY294002 and were

examined for fibronectin expression at 24, 48 or 72 h. Co-cultures or separ-

ately cultured blastocysts were fixed in 4% paraformaldehyde (PFA) for

15 min, washed in PBS + 1 mg/ml Polyvinyl alcohol (PBS + PVA) and then

permeabilised with PBS + PVA+ 0.3% Triton X 100 for 30 min and then

washed with PBS + PVA. Blocking was carried out in PBS + PVA+ 0.1%

Tween-20 + 0.7% BSA (PPTB) for 30 min. Primary and secondary anti-

bodies were diluted in PPTB. Samples were incubated with primary anti-

bodies [rabbit anti-integrin b3; Santa Cruz Biotechnology; #sc-14009,

rabbit anti-FAK; Sigma; #F-2918 or rabbit anti-fibronectin (unattached blas-

tocysts); Novus Biologicals, Littleton, CO, USA; #NBP1-91258, or rabbit

anti-fibronectin antibody (for blastocysts attached to coverslips); Calbio-

chem, EMD Chemicals, Gibbstown, NJ, USA] overnight at 48C and then

washed in PPTB before incubation with the secondary antibody (anti-rabbit

Alexa 488; LifeTechnologies, Carlsbad, CA,USA) for 2 h at room tempera-

ture. Samples were then washed three times in PPTB, with the final wash

lasting 10 min and then mounted in 5 ml Vectashield containing 1.5 mg/ml

4,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Burlingame,

CA, USA). Isotype controls were performed, in which cells were incubated

with purified rabbit IgG (Sigma) in place of the primary antibody. Fluores-

cence was visualized using the LSM 510 Meta confocal microscope (Carl

Zeiss, Germany) using a 405 nm laser and Argon laser (458, 477, 488 and

514 nm lines) at40 objective. A Z-stack was taken through the co-cultures

or the blastocysts, at 2.5mm intervals. This provides us with sections

throughout the whole blastocyst so that we can visualize the inner cell

mass (ICM) in relation to the rest of the embryo. Images were analysed

using LSM Image Browser software (Carl Zeiss) and the localization of pro-

teins is described in relation to the ICM. An unbiased third party performed

a blinded analysis of the immunofluorescence images.

Western blot analysis

All western blot data on Ishikawa cells are from Ishikawa cells cultured in the

absence ofblastocysts.Near confluentIshikawacells wereallocatedto serum

mediumor SS medium inthe presenceor absence of10 ng/ml IGF1 for48 h.

Similarly all western blots on blastocysts were performed in the absence of

Ishikawa cells. Blastocysts were isolated from at least eight mice in each ex-

periment. Day 5 blastocysts were allocated randomly to serum medium or

SS medium in the presence or absence of 10 ng/ml IGF1 and cultured for

48 h on non-adherent culture dishes (Corning). After 48 h, cells or blasto-

cysts were washed in cold PBS and lysed with lysis buffer (Cell Signalling,

Beverly, MA, USA) + 1 mM PMSF (Sigma). Collections were repeated

until there were 120 embryos to run per lane, per experiment. Protein

assays were carried out on Ishikawa cells according to the manufacturers

instructions (DC protein assay kit, Bio-Rad). Protein samples were diluted

with 6Laemmli buffer (35 mM Tris HCl, pH 6.8, 10.28% (w/v) sodium

dodecyl sulphate (SDS), 36% (v/v) glycerol, 0.05% (w/v) bromophenol

blue; Sigma) and heated for 5 min at 1008C. Proteins were electrophoresed

on an 8% SDS polyacrylamide gel, transferred to a nitrocellulose transfer

membrane (Hy-BLOT, Australia) and then incubated in blocking buffer

(Odyssey; Li-cor Biosciences, Lincoln, NE, USA) overnight at 48C with

gentle shaking. Membranes were probed with primary antibodies (rabbit

anti-E-cadherin; Novus Biologicals; #NB110-56937, rabbit anti-pFAK; Invi-

trogen; #44-624G, or rabbit anti-integrin b3; Santa Cruz Biotechnology;

#sc-14009 or mouse anti-a-tubulin; Sigma; #T9026) overnight at 48C in

blocking buffer+ 0.1% Tween-20. Membranes were washed in Tris-

buffered saline+ Tween 20 (TBST; 10 mM TrisHCl, pH 7.6, 150 mM

NaCl and 0.1% Tween 20) and subsequently incubated for 2 h with either

1:4000 Donkey anti-Rabbit IRDye 800CW (Odyssey) or 1:4000 Donkey

anti-Mouse IRDye 680LT (Odyssey). Proteins were visualized using the

Odyssey infrared imager and Odyssey application software, version 3

(Odyssey). Densitometry was performed using Image-J software (National

Institutes of Health, Bethesda, MD, USA) or Odyssey program.

Flow cytometry

Surface expression of integrin avb3, b3 andb1 on Ishikawa cells,cultured in

theabsence of blastocysts,was analysedby flowcytometry after 48 h culture

inserummedium orSS mediumin thepresenceor absence of10 ng/mlIGF1.

Cells were removed from the culture plate with 0.5% trypsin (Life Technol-

ogies). Cells were resuspended in 600 ml of PBS + 0.3% BSA and incubated

with primary antibodies pre-conjugated (1:200 anti-human integrin avb3;

eBioscience; San Diego, CA, USA; #11-0519-41, anti-human integrinb3;

eBioscience; #11-0619-41 or anti-human integrin b1; eBioscience;

#11-0299-41) or isotype control antibodies (Armenian hamster IgG/RatIgG 2a/2b; Biolegend; San Diego, CA, USA; #78023) for 30 min on ice.

Samples were then washed and resuspended in 200 ml PBS + 0.1% (v/v)

propidium iodide (Sigma). Analyses were performed using the FACS

Calibur (Becton Dickinson, San Jose, CA, USA) and Cell Quest (Becton

Dickinson) and Flow Jo (Treestar, San Carlos, CA, USA) software packages.

A minimum of 10 000 cells were analysed per sample.

Blastocyst outgrowth

Blastocysts were isolated from at least five mice in each experiment. Day 5

blastocysts were cultured on 24-well plates (Corning) that were either un-

coated, or coated with fibronectin. To coat wells, fibronectin was made to

4 Greenet al.


5/15

a concentration of 25mg/ml in PBS, andenough solution wasaddedto cover

eachwellandleftat48C overnight. Blastocysts wereallowed to outgrow,un-

disturbed for 72 h in serum medium or in SS medium in the presence or

absence of 10 ng/ml IGF1 with or without 5 mM LY294002 (Selleck). At

72 h blastocyst outgrowths were micrographed on the Axiovert 35 micro-

scope (Carl Zeiss) using a Axiocam ICc5 camera (Carl Zeiss) and Zen

image software (Carl Zeiss). For those blastocysts that were attached to

the well, the area of outgrowth (arbitrary units), along with blastocyst area

was measured using Image-J software (National Institute of Health) bytracing around the outgrown cells using the freehand area selection tool

andthenthe wand tool toobtainan areavalue in arbitraryunits. Experiments

were repeated three to eight times and the data pooled to calculate the

average area of outgrowth.

Statistical analysis

Embryos collected from multiple mice were randomly allocated between

treatments for all experiments. Chi-squared analysis of the overall propor-

tion of blastocysts attached to Ishikawa cell co-cultures was performed on

pooled proportion data from three separate experiments (Microsoft Office

Excel,Berkshire, UK).Westernblot analysis was performed three times.The

optical density of protein bands was normalized to a-tubulin and expressed

relative to the control, followed by statistical analysis using unpaired t-tests(Microsoft Office Excel). Flow cytometry analysis was performed three

times, followed by statistical analysis of the average fluorescence intensity,

using unpairedt-tests (Microsoft Office Excel). Embryo outgrowth experi-

ments were performed at least three times, followed by statistical analysis

using KruskalWallis test followed by Dunn multiple comparison test for

non-parametric data (GraphPad Prismversion 6.00 for Windows, GraphPad

Software, La Jolla, CA, USA, www.graphpad.com). A P, 0.05 was

considered statistically significant.

Results

Effect of IGF1 on attachment of mouseblastocysts to Ishikawa cells in vitro

Blastocyst attachmentwas assessedin thepresenceor absenceof 10 ng/ml

IGF1 (attachment experiment 1) after 24 and 48 h, before fixation with

PFA (Singhet al., 2010). At 24 h no blastocysts had attached to the Ishi-

kawa cells in any of the treatment groups (data not shown). Therefore

attachment was assessed at 48 h. The proportion of blastocysts that

attached to Ishikawa cells under SS conditions (27 attached/50) was

reduced compared with attachment in the presence of serum (39

attached/47; Fig. 2). The presence of IGF1 increased the proportion

of blastocyst attachment (IGF1: 39 attached/46) compared with attach-

ment in SS conditions (Fig. 2), tothe level seen inthe presenceof serum.

Effect of IGF1 on E-cadherin and integrin b3

protein levels and integrin avb3, b3 and b1

surface expression in Ishikawa cells

As IGF1 increased attachment of blastocysts to Ishikawa cells, the effect

of IGF1 on several adhesion proteins was investigated in Ishikawa cells

by western blot. There was no difference in E-cadherin expression in

Ishikawa cells between the serum, SS or SS plus IGF1 treatment groups

(Fig.3A). Furthermore, IGF1 had no effect on integrin b3 expression

in Ishikawa cells; however, expression of integrin b3 was increased in

the presence of serum (Fig.3B).

Although IGF1 had no effect on total expression of integrinb3 in Ishi-

kawa cells, it is the integrin expression on the apical surface that is

required for interaction with ECM proteins. Therefore, surface expres-

sion ofavb3, b3 and b1 integrins was investigated in Ishikawa cells

using flow cytometry. Integrins avb3, b3 and b1 were all expressed

on the cell surface in Ishikawa cells, as indicated by the greater fluores-

cence seen with the presence of each antibody compared with the

isotype control (Fig.3C). However, there was no difference in surface

expression of integrins avb3,b3 orb1 between any of the treatment

groups (Fig.3C).

Effect of IGF1 on phosphorylation of FAK and

total FAK expression levels in Ishikawa cells

and mouse blastocysts

FAK signalling is linked to integrin activation (Kornberget al., 1992).

Therefore, activationof FAK signallingpathways, indicatedby phosphor-

ylation of FAK (p-FAK) was investigated by western blot analysis in

Ishikawa cells and blastocysts. Phosphorylation of FAK was increased

in Ishikawa cells after 48 h treatment with 10 ng/ml IGF1 (Fig. 4A).Immunofluorescencewas usedto showthat totalFAK expression in Ishi-

kawa cells was increased on the apical surface in co-cultures after treat-

ment with IGF1 for48 h. Similarly, total FAKexpressionwas increasedin

blastocysts that were attached to Ishikawa cells in co-cultures in the

presence of IGF1 in comparison to co-cultures in the absence of IGF1

(Fig. 4B). Therabbit IgG isotype control had no detectableimmunofluor-

escence staining in either the blastocyst or Ishikawa cells (data not

shown). Western blots were performed on unattached blastocysts

and showed no effect on phosphorylation of FAK or total FAK levels in

blastocysts after IGF1 treatment for 48 h (Fig. 4C and D).

Figure 2 Insulin-like growth factor 1 (IGF1) improves attachment of

mouseblastocysts to Ishikawa cells in vitro. Attachmentof blastocysts to

Ishikawa cells at 48 h after culture in serum medium or serum starved

(SS) medium in the presence or absence of 10 ng/ml IGF1. Results

are displayed as the percentage attachment of blastocysts to the Ishi-

kawa cells.Nvalues in parentheses represent the total number of blas-

tocysts added to the Ishikawa cells, pooled from at least threeexperiments. Chi-square analysis was used to compare the treatment

groups. ** indicatesP, 0.01.

http://www.graphpad.com/http://www.graphpad.com/http://www.graphpad.com/http://www.graphpad.com/


6/15

Figure 3 IGF1 has no effect on E-cadherin or integrin b3 protein levels or integrin avb3, b3 and b1 surface expression in Ishikawa cells. Ishikawa cells

wereculturedfor 48 h inserum medium orSS mediumin the presenceor absence of10 ng/ml IGF1. (A) Representativewesternblot showing E-cadherin

expression. Western blots were probedfor anti-E-cadherin (green)and anti-a-tubulin (red).Graph showingaverage E-cadherinband intensities (+SEM,

n 3) normalized to a-tubulin and expressed relative to the SS control. (B) Representative western blot showing integrin b3 expression (green) and

a-tubulin (red).Graph showsaverageintegrinb3 band intensities(+SEM, n 3) normalizedto a-tubulin and expressed relativeto theSS control. Analysis

of variance with Tukeyspost hocanalysis was used to compare the treatment groups. * indicates P, 0.05 (C) Representative flow cytometry histogram

showing surface expression of integrin avb3, b3 and b1 in Ishikawa cells. MW: molecular weight.

6 Greenet al.


7/15

Figure 4 IGF1 increases focaladhesion kinase(FAK) phosphorylationand totalFAK expressionlevels in Ishikawa cells.Ishikawacells or blastocysts were

cultured for 48 h in SS medium in thepresence or absence of 10 ng/ml IGF1. (A) Representative western blot showing phosphorylayed FAK (p-FAK) ex-

pression in Ishikawa cells. Western blots were probed for anti-p-FAK (green) and anti-a-tubulin (red). Graph shows average p-FAK band intensities

(+SEM, n 3) in Ishikawa cells, normalized to a-tubulin and expressed relative to the SS control. (B) Representative confocal image showing total

FAK expression and localization (green) in blastocysts attached to Ishikawa cells. Plain arrows indicate FAK staining on Ishikawa apical cell surface.

Dottedarrow indicatesincreasedFAK staining on IGF1 treated blastocyst. Scalebars represent20 mm. Nine attachment siteswere stained across 3 experi-

mentsin theSS groupand 10 attachment siteswerestained across3 experiments inthe SSplusIGF1 group.(C) Representativewesternblot showing p-FAK

expressionin blastocysts.Westernblotswere probedfor anti-p-FAK (green) andanti-a-tubulin (lowerpanel).One hundredand twentyblastocystswere

run in each lane. Graph shows average p-FAK band intensities ( +SEM,n 3) in blastocysts, normalized to a-tubulin and expressed relative to the SS

control. (D) Representative western blot showing total FAK expression in blastocysts. Western blots were probed for anti-total-FAK (green) and

anti-a-tubulin (lower panel). One hundred and twenty blastocysts were run in each lane. Graph shows average total FAK band intensities (+SEM,

n 3) in blastocysts, normalizedto a-tubulin and expressedrelative to theSS control. Students T-test wasused to compare thecontrol to thetreatment

groups.



8/15

Fibronectin localization in early and late

stage blastocysts

Fibronectin is an important ECM protein that is secreted by blastocysts

butnot by Ishikawacells(Kaneko etal., 2013). Theexpressionand local-

ization of fibronectin in blastocysts was investigated by immunofluores-

cence over 72 h of culture, in the presence or absence of IGF1. Here

we describe the change in fibronectin expression over time as well as

an IGF-induced increase in fibronectin expression over time. Day5 blas-

tocysts wereculturedin SSmedium inthe presenceor absence of 10 ng/

ml IGF1 and collected at 24, 48 and 72 h and then immunostained for

fibronectin. At 24 h (now day 6 blastocyst), fibronectin was located in

the cytoplasm, nucleus, cellcell junctions and on the basal and apical

surface of the cells in the trophectoderm (Fig.5A and B). The ICM also

expressed fibronectin but this was restricted to the cells facing the

blastocoel. IGF1 treatment had no effect on fibronectin localization or

staining intensity at 24 h. IGF1 treatment increased fibronectin staining

intensity at 48 h compared with the SS control (Fig. 5C versus D). At

48 h (Day 7 blastocyst), fibronectin was located in the cytoplasm,

nucleus, cellcell junctions and on the basal and apical surface of the

trophectoderm (Fig.5C and D) and this staining was polarized to one

side of the blastocyst. At 48 h the majority of ICM cells had lost

fibronectin expression. The localization at 48 h was comparable

between the SS control and IGF1 group. At 72 h (Day 8 blastocyst),

there was a distinct polarization of fibronectin in the trophectoderm

on one side of the blastocyst and the localization was cytoplasmic and

punctate. There was no staining in the ICM (Fig. 5E and F). At 72 h in

the IGF1 treated group, fibronectin staining was of a much higher inten-

sity, compared with the SS group (Fig.5E versus F). Additionally, IGF1

treatment resulted in a distinct apical localization of fibronectin. The

polarized fibronectin staining seen at 48 and 72 h in blastocysts partially

coversthe trophoblastcells at theembryonic pole andextends from the

embryonic pole to the abembryonic pole on one side of the blastocyst.

This is the side that attached to Ishikawa cells in vitro.

The addition of the PI3 kinase inhibitor, 5 mM LY294002, to IGF1

treated blastocysts affected the overall polarization of fibronectin and

compared with the intense apical staining seen in IGF1 treated blasto-

cysts the localization of fibronectin in LY294002 treated blastocysts in

the presence of IGF1, was more cytoplasmic and strong staining in the

cellcell junctions was present (Fig.5F versus G).

To ensure the polarizedfibronectin staining wasfunctionallyrelatedto

attachment, blastocysts treated with IGF1that had adhered to glasscov-

erslips (72 h) were imaged for fibronectin staining. The attached surface

Figure 4 (Continued).

8 Greenet al.


9/15



10/15

of the blastocyst expressed fibronectin. The surface of the blastocyst

facing away from the coverslip did not express fibronectin (Fig. 5H).

Effect of IGF1 on blastocyst outgrowth

Fibronectin is known to increase blastocyst outgrowth in serum free

medium to the level seen in the presence of serum ( Armant et al.,

1986). Therefore blastocyst outgrowth was measured after 72 h (now

day 8 blastocysts) of culture on either uncoated wells or wells coated

with fibronectin, in serum medium or SS medium in the presence or

absence of 10 ng/ml IGF1 or 5 mM LY294002. Culture of blastocysts

in serum medium significantly increased the average area of blastocyst

outgrowth compared with SS blastocysts and blastocysts cultured in

thepresenceof IGF1 (Fig. 6H) and this was seen as an increase in troph-

ectoderm spreading across the culture dish (Fig.6A C). Under SS con-

ditions blastocyst outgrowth was increased by the presence of IGF1

when blastocysts were cultured on uncoated wells (Fig. 6B, C and H).

Blastocyst outgrowth in SS media was increased when blastocysts

were cultured on fibronectin coated wells (Fig.6B, D and H). The out-

growthof blastocystsin thepresence of IGF1 wasnot improved by fibro-

nectin (Fig.6C versus E and H). Outgrowth in the presence of IGF1 and

LY294002 on either uncoatedor fibronectin coated wellswas decreased

comparedwith outgrowth in IGF1alone on either uncoatedor fibronec-

tin coated wells, respectively (Fig.6FH).

Effect of pretreatment with IGF1, an IGF1R

nAb or LY294002 on attachment of mouse

blastocysts to Ishikawa cells

In attachment experiment 2, Day 5 blastocysts were cultured in serum

medium containing2 mg/ml IGF1R nAb or in SS medium in thepresence

or absence of 10, 100 or 1000 ng/ml IGF1 or with 10 ng/ml IGF1 in the

presenceof 2mg/mlIGF1RnAbor 5mM LY294002for 24 h.At theend

of thispretreatment periodthe healthof blastocysts cultured in LY294002was assessed and blastocysts (Day 6 blastocysts) were of healthy appear-

ance (Fig.7AD). Blastocysts were washed in SS medium and then cul-

tured on Ishikawa cells for 24 h, under SS conditions and assessed for

attachment (on Day 7). All blastocysts in all treatment groups remained

of normal appearance after the co-culture period (data not shown).

Pretreatment of blastocysts with serum for 24 h did not improve at-

tachmentform the serumstavedgroup (Fig.7E). Pretreatmentof blasto-

cysts with an IGF1R nAb in the presence of serum had no effect on

blastocyst attachment compared with pretreatment with serum

(Fig. 7E). Pretreatment with10 ng/ml IGF1for 24 h increased blastocyst

attachment to Ishikawa cells compared with pretreatment in SS medium

or medium containing serum (Fig. 7E). The positive effect of IGF1 pre-

treatment on attachment was reduced by the IGF1R nAb, to the levelseen in SS conditions (Fig.7E). Additionally, pretreatment of blastocysts

with the PI3K inhibitor LY294002 in the presence or absence of IGF1

reduced blastocyst attachment to Ishikawa cells compared with SS or

IGF1 pretreatment conditions, respectively (Fig. 7E). Pretreatment of

blastocysts with higher concentrations of IGF1 (100 and 1000 ng/ml)

did not increase blastocyst attachment compared with the SS control.

DiscussionA number of growth factors and their receptors are expressed by both

the preimplantation embryo and the reproductive tract and these influ-

ence preimplantation embryo development and implantation (Hardy

and Spanos, 2002;Armant, 2005). In particular, growth factors such as

IGF1, which are produced by the embryo and at the implantation site,

are thought to play a role in the co-ordination of blastocyst implantation

ability with uterine receptivity (Wanget al., 2000;Armant, 2005). In the

present study we used an in vitro model for embryo attachment to inves-

tigate the effect of IGF1 on mouse blastocyst adhesion competency.

Co-culture of blastocysts with Ishikawa cells for 48 h, under SS condi-

tions, decreased blastocyst attachment compared with attachment in

the presence of serum, which is well known to contain a large number

of growth factors. Addition of IGF1 to the co-cultures, in the absence

of serum, increased attachment to a level similar to that observed in

the presence of serum, suggesting that IGF1 alone can support attach-

ment of blastocysts to Ishikawa cellsin vitro. Furthermore, pretreatment

of Day5 blastocysts with IGF1 for 24 h wassufficient to improve blasto-

cyst attachment to Ishikawa cells compared with pretreatment in SS

medium. This suggests that thein vitrocultured embryo can be manipu-

lated by growth factors, normally found in vivo, to increase the blasto-

cysts adhesive ability.

The effects of IGF1 on blastocyst attachment were mediated by the

IGF1R as an IGF1R nAb negated the positive effects of IGF1 pretreat-

ment on blastocyst attachment.In theblastocyst IGF1 is known to phos-

phorylate Akt (Greenand Day, 2013), a signalling molecule downstreamof the PI3K pathway. In the present study the PI3K inhibitor LY294002

prevented the increase in attachment mediated by IGF1. LY294002

also decreased attachment in SS blastocysts, suggesting that even in SS

conditionsthere maybe autocrinegrowth factor stimulation or constitu-

tive activation of the PI3K pathway to influence attachment. In the

present study, pretreatment with serum did not improve blastocyst at-

tachment, in fact attachment in theseconditions was reduced compared

withattachment after IGF1pretreatment. High levels of insulin and other

growth factors that are present in serum may have negative effects on

blastocyst attachment, although serum appears to positively influence

Ishikawa cells to mediate attachment.

High concentrations of IGF1 are known to be detrimental to the

blastocyst by decreasing blastocyst hatching (Green and Day, 2013)and increasing apoptosis (Chi et al., 2000). High concentrations of

Figure 5 IGF1 increases apical fibronectin in late stage blastocysts. Representative confocal images of blastocysts stained for fibronectin (Green) and

DAPI (Blue). Day 5 blastocysts were cultured in SS medium for 24 (Aand B), 48 (Cand D) or 72 h (EH) in the absence (A, C and E) or presence

(B, D and F) of 10 ng/ml IGF1 or in the presence of a PI3 Kinase inhibitor (5 mM LY294002 + 10 ng/ml IGF1) (G) or in the presence of 10 ng/ml

IGF1, attached to a coverslip (H). Scale bars represent 20 mm. Red line indicates location of coverslip. Arrow indicates location of the inner cell mass

(ICM). The following total number of blastocysts were analysed, from at least 3 experimental repeats in each treatment; 24 in the SS group (24 h), 21 in

SS plus IGF1 (24 h), 26 in SS (48 h), 30 in SS plus IGF1 (48 h), 40 in SS (72 h), 59 in SS plus IGF1 (72 h), 13 in SS plus IGF1 + LY294002 (48 h) and 15 blas-

tocysts were analysed that were attached to a coverslip (72 h).

10 Greenet al.


11/15

IGF1 have also been shown to decrease normal blastocyst implantation

sitesand increase resorptioncompared withthe levelseen in blastocysts

exposed to physiological IGF1 concentrations (Pinto et al., 2002). In the

presentstudy,pretreatmentof blastocystswithhigh concentrationsof IGF1

did not improve blastocyst development unlike 10 ng/ml IGF1. These

effects could be due to the IGF1R down-regulation over time, in response

to high IGF1 concentrations (Pinto et al., 2002). Taken together these data

highlight our previous finding that the total concentration of IGF1 must be

tightly controlled in order to provide for optimal development.

Both pretreatment of blastocysts with IGF1 as well as co-culture of

blastocysts and Ishikawacells withIGF1 was sufficient to increase attach-

ment. IGF1, therefore, actson the blastocyst to mediate attachment and

may also increase attachment via actions on the Ishikawa cells. Thus the

molecular mechanisms by which IGF1 improved attachment were inves-

tigated in Ishikawa cells and blastocysts. Integrins are present in the api-

cal membraneof endometrial epithelialcells at thetime of implantation in

human,mouseand rat (Sutherland et al., 1993; Aplin et al., 1996; Schultz

etal.,1997; Kaneko etal., 2011a,b). Treatmentof Ishikawa cells withIGF1

for48 h hadno effecton total expression ofeither E-cadherin,or integrin

b3 expression levels. Integrins avb3, b3 and b1 were expressed at the

cell surface in Ishikawa cells; however, IGF1 treatment had no effect onthis surface expression. Although there was no change in integrin local-

izationor expressionin Ishikawacellsby IGF1treatment,total expression

of integrin b3 expression was increased by serum.

Fibronectinisproducedbytheembryo( Wartiovaara etal.,1979; Yoh-

kaichiyaet al., 1988;Thorsteinsdottir, 1992;Kaneko et al., 2013), and is

oneof theimportant bridging ligands, providingthe RGDsitefor therec-

ognition by integrins expressed on the apical surface of both the endo-

metrium and the embryo (Kaneko et al., 2013). IGF1 increased apical

fibronectin expressionin blastocysts after48 and 72 h of treatment, cor-

relating with the time of blastocyst attachment and outgrowth in vitro.

This increase in fibronectin at the apical surface may be responsible for

the increase in attachment of blastocysts to Ishikawa cells as the side of

the blastocyst that expressed fibronectin was the side that attached invitro. The overall polarity of fibronectin appeared to be developmentally

regulated, as fibronectin relocalised from trophectoderm cell cell

contacts (at 24 h culture) to the apical membrane (after 48 h culture)

at the time of attachment in vitro. Together these data suggests that

fibronectin producedby themouseblastocyst is an importantECM pro-

tein for attachment, acting as a possible bridging ligand for the integr-

ins expressed at the maternalembryo interface and that fibronectin

polarization may control the orientation of attachment to an underlying

cell layer.

Furthermore the acquisition of fibronectin on blastocysts from Day 7

onwards may regulate adhesion competency of the blastocyst. This was

demonstrated by the inability of Day 5 blastocysts to attach to Ishikawa

cells after 24 h (Day 6), whereas pretreatment of Day6 blastocystswith

IGF1 before placing them on Ishikawa cells enabled attachment after only

24 h (Day 7).ThisIGF1 inducedincreasein blastocystattachment correlated

with the increase in apical fibronectin on blastocysts by IGF1 and suggests

that attachmentis dependent on theadhesive competencyof theblastocyst,

which is developmentally regulated and can be increased by IGF1.

Fibronectin is also known to increase blastocyst outgrowth and there-

fore, as IGF1 increases fibronectin expression, we hypothesized that

IGF1 would also increase blastocyst outgrowth. We observed an in-

crease in blastocyst outgrowth on wells coated with fibronectin, in the

absence of IGF1, as well as an increase in blastocyst outgrowth in the

Figure 6 IGF1 increases mouse blastocyst outgrowth. Representa-

tivemicrographs showing blastocyst outgrowthafter 72 h. Day 5 blasto-

cysts were cultured on fibronectin (Fn) coated wells (D,E and G), or

uncoated wells (A, B , C and F) in the serum medium (A) or in SS

medium in the absence (B and D) or presence (C, E, F and G) of

10 ng/ml IGF1 and/or in the presence of a PI3 Kinase inhibitor (5 mM

LY294002 + 10 ng/ml IGF1) (F and G) in SS conditions (SS). Scale

bars represent 40 mm. (H) Average outgrowth area of blastocysts +

SEM cultured in serum medium (black bar) or SS medium (white bars)

in the presence or absence of 10 ng/ml IGF1 and/or in the presence

of a PI3 Kinase inhibitor (5 mM LY294002 + 10 ng/ml IGF1), pooled

from at least three experiments. Nvalues are given in parentheses and

represent the total number of blastocysts analysed across three to

eight experiments). Kruskal Wallis test followed by Dunns multiple

comparison test for non-parametric data was used to compare pre-

selected treatment groups. Lines between bars show compared

groups with significance. * indicates P, 0.05, ** indicates P, 0.01

and *** indicatesP, 0.001. AU: arbitrary units.



12/15


13/15

shown that attachment of rat blastocysts to Ishikawa cells in vitro is

reduced by pre-incubation of blastocysts and or Ishikawa cells with the

RGD-blocking peptide (Kanekoet al., 2013). In outgrowth studies, the

FN-120 fragment in fibronectin, which contains the RGD sequence,

promotes blastocyst outgrowth, whereas the FN-50 fragment, which is

lacking the RGD sequence, impairs blastocyst outgrowth (Yelianet al.,

1995). Furthermore, function blocking antibodies to integrin b1 and

b3 also reduced fibronectin binding on trophoblast cells and inhibited

blastocyst outgrowth in mice (Schultz and Armant, 1995;Yelianet al.,

1995). In human extravillous trophoblast cells IGF1 treatment increasesfibronectin binding and this is also blocked by an RGD-blocking peptide

(Kabir-Salmaniet al., 2002).

Integrin signallingis a complex process thatinvolvesboth alterations in

integrin affinity for ECM through pathwaysdownstream of growth factor

signalling (inside-out signalling), and signalling downstream of integrin-

ECM interactions (outside-insignalling) (Sastry and Horwitz, 1993). FAK

is a major componentof theoutside-in integrin signal transduction path-

way to regulate a variety of cellular events(reviewed in (Schaller, 2001)).

FAK is notonlyactivated by integrin-ECMligation, butalso is phosphory-

latedby hormonesand growth factors,including insulin and IGF-1 (Baron

et al., 1998,Casamassima and Rozengurt, 1998). In the present study

treatment of Ishikawacells withIGF1 for 48 h increasedphosphorylation

of FAK at Tyr397. Total FAK was also increased by IGF1 and its ex-

pression was apical. In the endometrium FAK is up-regulated in the

early proliferative to mid secretory stage, suggesting a role of FAK in im-

plantation (Orazizadeh et al., 2009). As FAKcan regulateintegrinactiva-

tion to promote integrin binding (Michael et al., 2009), the up-regulation

of FAK phosphorylation at Tyr397 in Ishikawa cells by IGF1 may play

a role in integrin recognition of ECMproteins secretedby theblastocyst,

and can explain the increase in attachment seen on Ishikawa cells,

although IGF1 did not increase surface expression or total expression

of integrins.

IGF1 treatment for 2 h has been shown to phosphorylate FAK in

trophoblast cells obtained from human 610 week placental tissue

(Kabir-Salmani et al., 2002). In the present study total FAK staining

after IGF1 treatment in blastocysts attached to Ishikawa cells appeared

to be stronger,although this wasnot quantified. However in unattached

blastocysts no change in phosphorylated or total FAK after IGF1 treat-

ment was observed, suggesting that FAK phosphorylation is dependent

on attachment status (Baronet al., 1998).

Akt is another signalling mediator that is necessary for integrin acti-

vation and has been shown to regulate fibronectin assembly through ac-tivation of integrin a5b1(Somanathet al., 2007). Akt is known to be

activated by IGF1 in mouse blastocysts (Green and Day, 2013). In the

present study, inhibition of PI3K prevented the polarized distribution

of fibronectin in blastocysts treated with IGF1. PI3K inhibition also

decreasedblastocystoutgrowthin thepresenceof IGF1as wellas attach-

ment of blastocysts to Ishikawa cells, possibly due to the decrease in

apical fibronectin. However, the overall intensity of fibronectin staining

remained high in thepresence of thePI3Kinhibitor, suggesting that fibro-

nectin synthesis in the blastocyst is via a signalling pathway other than

PI3K.

In summary, the present study demonstrated that IGF1 improved the

adhesion competency of the mouse blastocyst. IGF1 up-regulated fibro-

nectin expression on the apical surface of the trophoblasts, via activation

of the PI3K/Akt pathway and we propose that this increases blastocyst

adhesion competency (Fig.8). IGF1 also activated FAK expression and

signalling in Ishikawa cells, potentially regulating integrin activation in

orderto primethe cells forinteraction with thefibronectinon theblasto-

cyst (Fig.8). These findings provide evidence for the important role that

IGF1 plays in vivo, in the co-ordination of blastocyst adhesion compe-

tency and uterine receptivity. This study used a heterologous yet well

documented model to study adhesion of mouse blastocysts to the

human Ishikawa cell line in response to IGF1. This system would be

Figure 8 Schematic diagram proposing mechanisms by which IGF1 increases blastocyst adhesion competency and mediates attachment to Ishikawa

cells.IGF1 increases attachmentof mouseblastocysts to Ishikawa cells in vitro andhas actions on both theIshikawacellsand theblastocysts. IGF1 increases

apicalfibronectin on blastocystsand thismay bridgeintegrinspresent on theapical surface of theblastocyst and Ishikawacells. ThePI3 kinase(PI3K) inhibitor

LY294002 reduces apical fibronectin and pretreatment of blastocysts with LY294002 reduces attachment. IGF1 increases focal adhesion kinase (FAK)expression and phosphorylation and this may increase the integrins ability to bind and recognize extracellular matrix proteins such as fibronectin.



14/15

useful to studyIGF1 adhesion regulationof human blastocysts or a spher-

oidal human trophoblast cell line such as JArcells. At thetime of implant-

ation in the human there is a sufficient amount of IGF1 present in the

uterineand luminal secretions (Lighten etal.,1998); however, thismater-

nal IGF1 is absent during thein vitro culture of embryos. As failure of the

embryo to implant successfullyis a majorcauseof IVFfailure(Milleret al.,

2012),ourresults,alongwithothers(Lighten etal.,1998; GreenandDay,

2013) indicate that pretreatment of human blastocystswith IGF1 before

their introduction back into the uterus may be a possible avenue to

improve implantation rates in assisted reproduction.

Acknowledgements

We would like to thank Professor Chris Murphy (University of Sydney,

Australia) for the Ishikawa cell line and integrin b3 and total FAK anti-

bodies; Dr Steve Assinder (University of Sydney, Australia) for the

E-cadherin antibody; Dr Matthew Naylor (University of Sydney, Austra-

lia)for the p-FAK antibody;Professor Frank Lovicu (Universityof Sydney,

Australia) for the fibronectin antibody. We also thank Dr Louise Cole

and Dr Yingying Su (Advanced Microscopy Facility, Bosch Institute, Uni-

versity of Sydney) for help with confocal microscopy and Dr AngelesSanchez-Perez (Live Cell Analysis facility, Bosch Institute, University of

Sydney) for help with flow cytometry.

Authors roles

C.J.G designed the study, carried out experimental work, analysed data

and prepared manuscript. S.T.F gave technical assistance in running and

analysingflow cytometry experiments as wellas experimental design and

editing of manuscript. M.L.D designed the study, gave technical advice

and critical editing of manuscript.

FundingThis work was supportedby internal funds from theUniversity of Sydney

and the Bosch Institute.

Conflict of interest

The authors declare that there is no conflict of interest that could be

perceived as prejudicing the impartiality of the research reported.

ReferencesAplinJD, SpanswickC, BehzadF, KimberSJ, Vicovac L.Integrins beta 5,beta 3

and alpha v are apically distributed in endometrial epithelium.Mol Hum

Reprod1996;2:527534.

Armant DR. Blastocysts dont go it alone. Extrinsic signals fine-tune the

intrinsic developmental program of trophoblast cells.Dev Biol2005;280:

260280.

ArmantDR, KaplanHA, Lennarz WJ. Fibronectin andlaminin promote invitro

attachment and outgrowth of mouse blastocysts. Dev Biol 1986;116:

519523.

Baron V, Calleja V, Ferrari P, Alengrin F, Van Obberghen E. p125Fak focal

adhesion kinase is a substrate for the insulin and insulin-like growth

factor-I tyrosine kinase receptors.J Biol Chem1998;273:71627168.

Bedzhov I, Liszewska E, Kanzler B, Stemmler MP. Igf1r signaling is

indispensable for preimplantation development and is activated via a

novel function of E-cadherin.PLoS Genet2012;8:e1002609.

Bell GI,StempienMM, Fong NM,RallLB. Sequencesof livercDNAsencoding

two different mouse insulin-like growth factor I precursors.Nucleic Acids

Res1986;14:78737882.

Casamassima A, Rozengurt E. Insulin-like growth factor I stimulates tyrosine

phosphorylation of p130(Cas), focal adhesion kinase, and paxillin. Role of

phosphatidylinositol 3

-kinase and formation of a p130(Cas).Crkcomplex.J Biol Chem 1998;273:2614926156.

Castelbaum AJ, Ying L, Somkuti SG, Sun J, Ilesanmi AO, Lessey BA.

Characterization of integrin expression in a well differentiated endometrial

adenocarcinoma cell line (Ishikawa). J Clin Endocrinol Metab 1997;82:

136142.

ChiMMY,SchleinAL, MoleyKH. High insulin-likegrowth factor1 (IGF-1) and

insulin concentrations trigger apoptosis in the mouse blastocystvia down-

regulation of the IGF-1 receptor. Endocrinology2000;141:47844792.

Green CJ, Day ML. Insulin-like growth factor 1 acts as an autocrine factor to

improve early embryogenesisin vitro.Int J Dev Biol2013;57:837844.

HannanNJ,PaivaP, Dimitriadis E,Salamonsen LA.Modelsfor studyof human

embryo implantation: choice of cell lines?Biol Reprod2010;82:235245.

Hardy K, SpanosS. Growthfactorexpression and function in thehuman and

mouse preimplantation embryo.J Endocrinol2002;172:221236.Heneweer C, Schmidt M, Denker HW, Thie M. Molecular mechanisms in

uterine epithelium during trophoblast binding: the role of small GTPase

RhoA in human uterine Ishikawa cells. J Exp Clin Assist Reprod2005;2:4.

Inzunza J, Danielsson O, Lalitkumar PG, Larsson O, Axelson M, Tohonen V,

Danielsson KG, Stavreus-Evers A. Selective insulin-like growth factor-I

antagonist inhibits mouse embryo development in a dose-dependent

manner.Fertil Steril2010;93:26212626.

Kabir-Salmani M, Shiokawa S, Akimoto Y, Hasan-Nejad H, Sakai K,

Nagamatsu S, Nakamura Y, Hosseini A, Iwashita M. Characterization of

morphological and cytoskeletal changes in trophoblast cells induced by

insulin-like growth factor-I.J Clin Endocrinol Metab 2002;87:57515759.

Kabir-Salmani M, Shiokawa S, Akimoto Y, Sakai K, Nagamatsu S, Sakai K,

Nakamura Y, Lotfi A, Kawakami H, Iwashita M. Alphavbeta3 integrin

signaling pathway is involved in insulin-like growth factor I-stimulatedhuman extravillous trophoblast cell migration. Endocrinology 2003;

144:16201630.

Kabir-Salmani M, Shiokawa S, Akimoto Y, Sakai K, Iwashita M. The role of

alpha(5)beta(1)-integrin in the IGF-I-induced migration of extravillous

trophoblast cells during the process of implantation. Mol Hum Reprod

2004;10:9197.

KanekoY,Day ML,Murphy CR.Integrin beta3 inrat blastocystsand epithelial

cells is essential for implantationin vitro: studies with Ishikawa cells and

small interfering RNA transfection.Hum Reprod2011a;26:16651674.

Kaneko Y, Lecce L, Day ML, Murphy CR. beta(1) and beta(3) integrins

disassemble from basal focal adhesions and beta(3) integrin is later

localised to the apical plasma membrane of rat uterine luminal epithelial

cells at the time of implantation.Reprod Fertil Dev2011b;23:481495.

Kaneko Y, Lecce L, Day ML, Murphy CR. Focal adhesion kinase localizes to

sites of cell-to-cell contact in vivo and increases apically in rat uterine

luminal epithelium and the blastocyst at the time of implantation.

J Morphol2012;273:639650.

Kaneko Y, Murphy CR, Day ML. Extracellular matrix proteinssecreted from

both the endometrium and the embryo are required for attachment: a

study using a co-culture model of rat blastocysts and Ishikawa cells.

J Morphol2013;274:6372.

Kang YJ, Forbes K, Carver J, Aplin JD. The role of the osteopontin-integrin

alphavbeta3 interaction at implantation: functional analysis using three

differentin vitromodels.Hum Reprod2014;29:739749.

14 Greenet al.


15/15

Kapur S, Tamada H, Dey SK, Andrews GK. Expression of insulin-like growth

factor-I (IGF-I) and its receptor in the peri-implantationmouse uterus, and

cell-specific regulation of IGF-I gene expression by estradiol and

progesterone.Biol Reprod1992;46:208219.

Kornberg L, Earp HS, Parsons JT, Schaller M, Juliano RL. Cell adhesion or

integrin clustering increases phosphorylation of a focal adhesion-

associated tyrosine kinase.J Biol Chem1992;267:2343923442.

Lessey BA, Ilesanmi AO, Castelbaum AJ, Yuan L, Somkuti SG, Chwalisz K,

Satyaswaroop PG. Characterization of the functional progesteronereceptor in an endometrial adenocarcinoma cell line (Ishikawa):

progesterone-induced expression of the alpha1 integrin. J Steroid

Biochem Mol Biol1996;59:3139.

Lighten AD, Moore GE, Winston RM, Hardy K. Routine addition of human

insulin-like growth factor-I ligand could benefit clinical in-vitrofertilization

culture.Hum Reprod1998;13:31443150.

Michael KE, Dumbauld DW, Burns KL, Hanks SK, Garcia AJ. Focal adhesion

kinase modulates cell adhesion strengthening via integrin activation. Mol

Biol Cell2009;20:25082519.

Miller PB, Parnell BA, Bushnell G, Tallman N, Forstein DA, Higdon HL 3rd,

Kitawaki J, Lessey BA. Endometrial receptivity defects during IVF cycles

with and without letrozole.Hum Reprod2012;27:881888.

Murphy LJ, Murphy LC, Friesen HG. Estrogen induces insulin-like growth

factor-I expression in the rat uterus.Mol Endocrinol1987;1:445450.Nasr-Esfahani M,JohnsonMH, AitkenRJ.The effectof ironand iron chelators

on the in-vitro block to development of the mouse preimplantation

embryo: BAT6 a new medium for improved culture of mouse embryos

in vitro.Hum Reprod1990;5:9971003.

Nishida M, Kasahara K, Kaneko M, Iwasaki H, Hayashi K. [Establishment of a

new human endometrial adenocarcinoma cell line, Ishikawa cells,

containing estrogen and progesterone receptors]. Nihon Sanka Fujinka

Gakkai zasshi1985;37:11031111.

Orazizadeh M, Rashidi I, Saremi J, Latifi M. Focal adhesion kinase (FAK)

involvement in human endometrial remodeling during the menstrual

cycle.Iran Biomed J2009;13:95101.

PariaBC, Ma W,TanJ, Raja S,Das SK,Dey SK,Hogan BL.Cellular andmolecular

responses of the uterus to embryo implantation can be elicited by locally

applied growth factors.Proc Natl Acad Sci USA 2001;98:10471052.PintoAB, Schlein AL,Moley KH. Preimplantationexposureto highinsulin-like

growth factor I concentrations results in increased resorption rates in vivo.

Hum Reprod2002;17:457462.

Riley JK, Carayannopoulos MO, Wyman AH, Chi M, Ratajczak CK,

Moley KH. The PI3K/Akt pathway is present and functional in the

preimplantation mouse embryo.Dev Biol2005;284:377386.

Riley JK, Carayannopoulos MO, Wyman AH, Chi M, Moley KH.

Phosphatidylinositol 3-kinase activity is critical for glucose metabolism and

embryo survival in murine blastocysts.J Biol Chem2006;281:60106019.

SastrySK, HorwitzAF. Integrincytoplasmic domains:mediators of cytoskeletal

linkages and extra- and intracellular initiated transmembrane signaling.Curr

Opin Cell Biol1993;5:819831.

Schaller MD. Biochemicalsignals and biological responseselicitedby thefocal

adhesion kinase.Biochim Biophys Acta 2001;1540:121.

Schultz JF, ArmantDR. Beta 1- andbeta 3-class integrins mediatefibronectin

binding activity at the surface of developing mouse peri-implantation

blastocysts. Regulation by ligand-induced mobilization of stored

receptor.J Biol Chem1995;270:1152211531.Schultz JF, Mayernik L, Rout UK, Armant DR. Integrin trafficking regulates

adhesion to fibronectin during differentiation of mouse peri-implantation

blastocysts.Dev Genet1997;21:3143.

SinghH, Nardo L,KimberSJ, AplinJD. Earlystages ofimplantationas revealed

by anin vitromodel.Reproduction2010;139:905914.

Slater M, Murphy CR.Differential expression of insulin-like growth factors in

the uterine epithelium and extracellular matrix during early pregnancy.

Matrix Biol1999;18:579584.

Somanath PR,Kandel ES,Hay N, ByzovaTV. Akt1 signaling regulatesintegrin

activation, matrix recognition, and fibronectin assembly.J Biol Chem 2007;

282:2296422976.

Sutherland AE, Calarco PG, Damsky CH. Developmental regulation of

integrin expression at the time of implantation in the mouse embryo.

Development1993;119:11751186.Thorsteinsdottir S. Basement membrane and fibronectin matrix are

distinct entities in the developing mouse blastocyst. Anat Rec 1992;

232:141149.

Vlahos CJ, Matter WF, Hui KY, Brown RF. A specific inhibitor

of phosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-

benzopyran-4-one (LY294002).J Biol Chem 1994;269:52415248.

Wang J, Mayernik L, Schultz JF, Armant DR. Acceleration of trophoblast

differentiation by heparin-binding EGF-like growth factor is dependent

on the stage-specific activation of calcium influx by ErbB receptors in

developing mouse blastocysts.Development2000;127:3344.

Wang J, Mayernik L, Armant DR. Integrin signaling regulates blastocyst

adhesion to fibronectin at implantation: intracellular calcium transients

and vesicle trafficking in primary trophoblast cells. Dev Biol 2002;

245:270279.Wartiovaara J, Leivo I, Vaheri A. Expression of the cell surface-associated

glycoprotein, fibronectin, in the early mouse embryo. Dev Biol 1979;

69:247257.

Yelian FD, Yang Y, Hirata JD, Schultz JF, Armant DR. Molecular interactions

between fibronectinand integrins during mouseblastocystoutgrowth. Mol

Reprod Dev1995;41:435448.

Yohkaichiya T, Hoshiai H, Yajima A. Fibronectin localization in the mouse

embryo from the blastocyst stage to the egg cylinder stage in vitro.

Tohoku J Exp Med1988;154:261269.


hum. reprod.-2014-green-humrep-deu309.pdf

Documents