www.elsevier.com/locate/yexcr

Experimental Cell Research 296 (2004) 317–326

Lysophosphatidic acid regulates murine blastocyst development by

transactivation of receptors for heparin-binding EGF-like growth factor

Zitao Liu and D. Randall Armant*

C.S. Mott Center for Human Growth and Development, Departments of Obstetrics and Gynecology, and Anatomy and Cell Biology,

Wayne State University School of Medicine, Detroit, MI 48201, USA

Received 16 December 2003; received in revised form 3 February 2004

Available online 14 March 2004

Abstract

Transient elevation of intracellular calcium (Cai2+) by various means accelerates murine preimplantation development and trophoblast

differentiation. Several G-protein-coupled receptors (GPCRs), including the lysophosphatidic acid (LPA) receptor (LPAR), induce Cai2+

transients and transactivate the EGF receptor (ErbB1) through mobilization of EGF family members, including heparin-binding EGF-like

growth factor (HB-EGF). Because HB-EGF accelerates blastocyst differentiation in vitro, we examined whether crosstalk between LPA and

HB-EGF regulates peri-implantation development. During mouse blastocyst differentiation, embryos expressed LPAR1 mRNA

constitutively, LPAR2 only in late stage blastocysts and no LPAR3. Consistent with a mechanism based on Cai2+ signaling, LPA rapidly

accelerated the rate of trophoblast outgrowth, an index of blastocyst differentiation, and chelation of Cai2+ with BAPTA-AM blocked LPA

stimulation. Interfering with HB-EGF signaling through ErbB1 or ErbB4 also attenuated LPA stimulation. We established that mouse

blastocysts indeed express HB-EGF and that LPA induces the transient accumulation of HB-EGF on the embryo surface, which was blocked

by treatment with either BAPTA-AM or the protein trafficking inhibitor, brefeldin A. We conclude that LPA accelerates blastocyst

differentiation through its ability to induce Cai2+ transients and HB-EGF autocrine signaling. Transactivation of ErbB1 or ErbB4 by HB-EGF

could represent a convergent signaling pathway accessed in the trophoblast by stimuli that mobilize Cai2+.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Transactivation; Growth factors; Calcium signaling; Lysophosphatidic acid; EGF receptor; HB-EGF; Blastocyst; Trophoblast; Implantation; erbB4

Introduction dently of maternal systemic control to form an adhesion

Blastocyst implantation in the uterine endometrium, re-

quired for successful pregnancy in placental mammals, is

mediated by the trophoblast, which differentiates from out-

side blastomeres at the morula stage [1]. After formation of

the blastocyst, the trophoblast transforms from a transporting

epithelium with nonadhesive apical surfaces (trophectoderm)

into invasive cells capable of adhering to extracellular matrix

(ECM) and remodeling the endometrium (reviewed in Refs.

[2,3]). The activities of the trophoblast and endometrium are

coordinated during development, in part, through uterine

signaling to the embryo that regulates trophoblast differenti-

ation. A fertilized ovum can be cultured in vitro indepen-

0014-4827/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.yexcr.2004.02.006

* Corresponding author. C.S. Mott Center for Human Growth and

Development, Wayne State University School of Medicine, 275 E. Hancock

Avenue, Detroit, MI 48201-1415. Fax: +1-313-577-8554.

E-mail address: d.armant@wayne.edu (D.R. Armant).

competent blastocyst, demonstrating that an endogenous

embryonic program orchestrates its preimplantation devel-

opment. However, embryonic development progresses more

slowly in vitro than in utero, perhaps because paracrine and

juxtacrine signaling from the uterus stimulates embryonic

development. For example, calcitonin and heparin-binding

EGF-like growth factor (HB-EGF), which are expressed in

the rodent endometrium at the time of implantation [4,5],

both accelerate blastocyst development [6,7]. In vitro, calci-

tonin and HB-EGF advance blastocyst differentiation to the

adhesive stage, assessed by accelerated trophoblast out-

growth and precocious trafficking of the integrin a5h1 from

the cytosol to the apical surface of the trophoblast in associ-

ation with heightened fibronectin binding.

Previous work from our laboratory demonstrates that

intracellular Ca2+ (Cai2+) signaling initiated through various

means regulates trophoblast differentiation to an invasive

phenotype. Both calcitonin and HB-EGF induce Cai2+

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326318

transients in blastocysts, and chelation of Cai2+ attenuates

their acceleration of trophoblast differentiation [6,7]. Tro-

phoblast differentiation is also accelerated when a Cai2+

transient is induced using pharmacological agents such as

ethanol or Ca2+ ionophore [8]. Therefore, we have proposed

that Ca2+ is a key second messenger regulating trophoblast

differentiation [2]. Although the contribution of downstream

signaling by protein kinase C (PKC) and calmodulin has

been explored, it remains unclear exactly how Cai2+ per-

forms this function and whether it activates a convergent

pathway in trophoblasts downstream of several external

stimuli.

Lysophosphatidic acid (LPA; 1-acyl-2-sn-glycerol-3-

phosphate) is a lipid-based signaling molecule produced

in stimulated cell membranes through the hydrolysis of

phospholipids [9]. It is released from activated platelets,

leukocytes, epithelial cells, neuronal cells, and tumor cells

by lysophospholipase D, and can be identified in cell

culture medium, plasma, human ascites fluid, and human

follicular fluid [10,11]. Plasma membrane receptors for

LPA are seven-transmembrane G-protein-coupled recep-

tors (GPCRs) that are express in most mammalian organs

[12]. LPA is a potent signaling molecule involved in a

variety of physiological and pathological processes, in-

cluding cell differentiation and proliferation, cytoskeletal

rearrangement, cell–cell communication, cell invasion,

and survival [13]. In mammalian developmental models,

LPA stimulates oocyte maturation [14], preimplantation

development of two- or four-cell embryos to the blasto-

cyst stage [15], and embryo transport in the oviduct [16].

An increase in serum lysophospholipase D activity is

observed during human pregnancy [17], indicative of

LPA accumulation within the reproductive tract. In ovar-

ian carcinoma, LPA stimulates cell motility and invasive-

ness, which are also key properties of differentiated

trophoblasts [18].

HB-EGF is initially expressed as a membrane-anchored

protein (proHB-EGF) that is processed to a secreted form

(sHB-EGF) by metalloproteinases that cleave the extracel-

lular domain near the transmembrane region [19]. Ecto-

domain shedding of HB-EGF is a key step in the

transactivation of the EGF receptor (ErbB1) by a variety

of cell interactions, including the activation of GPCRs

[20]. The LPA receptor (LPAR) can transactivate ErbB1

through crosstalk with HB-EGF [21]. Both Ca2+-depen-

dent and protein kinase C-dependent mechanisms are

involved in signaling upstream to HB-EGF shedding

[22,23]. All three LPA receptors induce Cai2+ transients

through inositol phosphate signaling, which is accompa-

nied by the generation of diacylglycerol to activate protein

kinase C [12]. Therefore, we hypothesize that LPA accel-

erates trophoblast differentiation through a similar mech-

anism involving ErbB1 or the other HB-EGF receptor,

ErbB4 [24]. In addition to its expression in murine uterine

epithelial cells at the site of implantation [5], HB-EGF is

expressed by the hamster blastocyst [25], suggesting that

it is indeed capable of regulating trophoblast differentia-

tion through autocrine signaling. This study was under-

taken to delineate signaling mechanisms activated by LPA

in blastocysts that regulate trophoblast differentiation in

preparation for implantation.

Materials and methods

Production and culture of mouse embryos

Mouse NSA � B6SJL morulae and blastocysts were

generated as previously described [27] collecting embryos

on gestation day (GD) 3 and 4, respectively (66 and 90

h post-hCG). All embryo cultures were carried out at 37jCin a 5% CO2 incubator using Ham’s F-10 medium (Sigma,

St. Louis, MO) containing 4 mg/ml BSA. Embryos were not

cultured in any serum-supplemented medium.

Blastocyst outgrowth culture

Morulae (GD 3) were cultured in Ham’s F-10 for 48

h to the mid-blastocyst stage. Blastocyst outgrowths were

produced by culture on plates pre-coated with 50 Ag/ml of

the fibronectin cell-binding fragment FN-120 (Life Tech-

nologies, Inc., Gaithersburg, MD) beginning at noon of

GD 5, as previously detailed [26,27]. The outgrowths

were observed using a Leica (Wetzlar, Germany) DM

IRB microscope and recorded regularly for 72 h using a

Spot Jr. digital camera (Diagnostic Instrument, Inc., Ster-

ling Heights, MI). Embryos failing to outgrow were

excluded from the analysis. The percentage of blastocysts

outgrowing at each time point is reported, beginning at

noon on GD 5 (time 0). The outgrowth rate was evaluated

quantitatively by determining the time (in hours of out-

growth culture) at which 50% of the embryos outgrew

(T50) using Probit analysis [8]. In some experiments, the

embryos were pretreated for 30 min in medium containing

5 AM BAPTA-AM (Calbiochem), 10 nM AG1478 (Cal-

biochem), 10 Ag/ml brefeldin A (Sigma), 10 Ag/ml

cycloheximide (Sigma), or 0.1 U/ml heparitinase I (EC

4.2.2.8; Seikagaku, Japan) before treatment for 1 h with 1

AM LPA. During treatment with 1 AM LPA, some

embryos were co-treated with 10 Ag/ml of nonimmune

IgG (Jackson ImmunoResearch Laboratories, West Grove,

PA) or neutralizing antibodies against HB-EGF (goat

polyclonal, R&D Systems, Minneapolis, MN), ErbB1

(Ab-2, Lab Vision Corporation, Fremont, CA), or ErbB4

(Ab-3, Lab Vision).

Identification of mRNAs by RT-PCR

RNA was isolated with the RNeasy mini kit (Qiagene)

according to the manufacturer’s instructions. Complemen-

tary DNA (cDNA) was prepared with total RNA from 100

embryos in a 20-ml mixture containing 200 units of

Fig. 1. Acceleration of trophoblast outgrowth by LPA. Morulae collected on

GD 3 were treated with 0 (squares) or 1 (diamonds) AM LPA from GD 3 to

GD 8 (A) or during the indicated time periods (B). The percentage of

blastocysts outgrowing beginning at noon on GD 5 was determined (A) and

used to calculate an outgrowth rate (T50, B) as described in Materials and

methods. Open bar indicates treatment with LPA was for only 1 h. Each

treatment was performed with at least 40 embryos in three independent

experiments. Error bars represent the 95% confidence interval. *P < 0.05

compared to 0 AM LPA.

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326 319

Superscript II RNase H-reverse transcriptase (Invitrogen),

0.5 Ag oligo (dT)12–18 (Invitrogen), 50 mM Tris–HCl, pH

8.3, 72 mM KCl, 3 mM MgCl2, 10 mM dithiothreitol, 40

units RNaseOUT (Invitrogen), and 0.5 mM of a deoxy-

nucleotide mixture containing dATP, dCTP, dGTP, and

dTTP (dNTP; Invitrogen). The reaction was terminated

by heating at 95jC for 5 min. Ten embryo-equivalents of

cDNA were assayed in a 25-Al reaction mixture containing

0.2 mM dNTP, 1.5 mM MgCl2, 0.5 AM of each primer,

and 1 unit Taq DNA polymerase in 1� PCR buffer

(Invitrogen). PCR primer sequences were based on pub-

lished reports [28–30]: LPAR1 (endothelial differentiation

gene-2, EGD-2), 5V-ATCTTTGGCTATGTTCGCCA-3V(sense) and 5V-TTGCTGTGAACTCCAGCCA-3V (anti-

sense); LPAR2 (EDG-4), 5V-TGGCCTACCTCTTCCT-

CATGTTCCA-3V(sense) and 5V-GGGTCCAGCACAC-

CACAAATGCC-3 V(ant isense) ; LPAR3 (EDG-7) :

5V-GAATTGCCTCTGCAACATCTC-3V(sense) and 5V-GAGTAGATGATGGGGTTCA-3V(antisense); HB-EGF, 5V-ATGAAGCTGCTGCCGTCG-3 V( s e n s e ) a nd 5 V-TCAGTGGGAGCTAGCCAC-3V(antisense). The thermal

cycling program consisted of denaturation at 94jC for 1

min, annealing at 56jC (LPAR1, LPAR2, LPAR3) or

63.5jC (HB-EGF) for 1 min, and extension at 72jC for

1 min, with a final extension time of 10 min during the

last cycle. The PCR-amplified DNA fragments were visu-

alized after 1.5% agarose electrophoresis on a UV trans-

illuminator and photographed. Amplicon sizes for LPAR1,

LPAR2, LPAR3, and HB-EGF were 392, 521, 382, and

627 bp, respectively.

Immunofluorescence microscopy, image processing, and

deconvolution

Embryos were fixed for 30 min at room temperature in

PBS containing 3% paraformaldehyde (Polysciences, Inc.).

Some embryos were permeabilized by incubation for 10

min in PBS containing 0.1% Triton X-100 (Sigma). The

embryos were washed free of fixative or detergent by

transfer five times through 0.15 M glycine, pH 7.0. After

further transfer through three drops of 1 mg/ml BSA in

PBS (PBS–BSA), the embryos were incubated overnight

at 4jC in 10 Ag/ml of primary antibody in PBS–BSA.

Both recombinant human HB-EGF and purified mouse

monoclonal antibody (anti-HB-EGF) against the extracel-

lular domain of human HB-EGF were purchased from

R&D Systems. To test the specificity of labeling, anti-HB-

EGF was absorbed for 30 min with a 10-fold excess of

soluble recombinant human HB-EGF. After incubation

with primary antibody, the embryos were transferred

through five drops of PBS–BSA and incubated for 60

min at 37jC with 10 Ag/ml FITC-labeled donkey anti-

mouse IgG (Jackson ImmunoResearch Laboratories) in

PBS–BSA. Images were captured with a Hamamatsu

Orca digital camera (Hamamatsu City, Japan). Image

deconvolution was conducted using SimplePCI (C-Imag-

ing Systems, Cranberry Township, PA) imaging software

and a ‘‘no neighbor’’ algorithm. The fluorescence inten-

sities of blastocysts labeled with anti-HB-EGF were de-

termined using SimplePCI software, which measured the

total fluorescence of each embryo delineated within a

digital image. The mean intensity of control embryos

treated for 45 min with HB-EGF and labeled with

antigen-adsorbed anti-HB-EGF was subtracted from the

mean intensity for each treatment group to derive the

specific fluorescence intensity.

Statistics

All experiments were repeated at least three times. For

immunofluorescent staining, there were at least 15 embryos

in each treatment group. Mean specific fluorescence inten-

sities, in arbitrary units, were compared using ANOVA and

Duncan’s post hoc test. Probit analysis was used to compute

T50 values and statistically compare trophoblast outgrowth

rates [6,8]. Error bars shown in figures represent 95%

Fig. 2. LPA dose response. Blastocysts were treated on GD 5 with the

indicated concentrations of LPA for 1 h and the outgrowth rate (T50) was

determined, as in Fig. 1B. Each treatment was performed with at least 40

embryos in three independent experiments. *P < 0.05 compared to 0 AMLPA.

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326320

confidence intervals for T50 values and the SEM for specific

fluorescence intensity values. P values less than 0.05 were

considered significant.

Results

LPA accelerates trophoblast differentiation

The rate of trophoblast differentiation in culture can be

determined by assessing blastocysts for the first sign of

trophoblast outgrowth, which is observed as a spreading

monolayer of cells around the base of the embryo con-

comitantly with disappearance of the spherical blastocyst

morphology. Because total LPA in human serum is 0.85 F0.22 AM in males and 1.57 F 0.56 AM in females [31],

embryos were initially treated without or with 1 AM LPA

(Fig. 1A). LPA shifted the onset of blastocyst outgrowth to

an earlier time, with a significant reduction in the T50

Fig. 3. Expression of mRNA for LPA receptors in preimplantation embryos. Emb

RT-PCR to assess expression of LPAR1 (A), LPAR2 (B), and LPAR3 (C). The PCR

kidney (B and C) total RNA provided positive controls (left lane). DNA ladders

(34.75 h for the control and 24.45 h for LPA, P < 0.05).

To identify the developmental stage(s) when embryos can

be stimulated by LPA, morulae collected on GD 3 were

treated with 1 AM LPA at specific times during culture and

the subsequent rate of outgrowth was determined. All

treatments from GD 3 onward significantly shorten the

T50 (Fig. 1B). To determine whether rapid intracellular

signaling induced by LPA is sufficient to accelerate tro-

phoblast development, embryos were exposed to LPA for a

brief interval. Treatment with 1 AM LPA on GD 5 for 1

h significantly accelerated outgrowth comparably to con-

tinuous LPA exposure from GD 5 to 8. A dose-response

study established the effective concentration range of LPA

to be at or above 100 nM, with maximal stimulation at 1

AM or higher (Fig. 2).

Preimplantation embryos express LPA receptors

The biological effects of LPA are mediated in mammals

through the GPCRs, LPAR1 (EDG-2), LPAR2 (EDG-4), and

LPAR3 (EDG-7) [12]. The mRNAs for each receptor were

examined by RT-PCR throughout trophoblast differentiation

(Fig. 3). Positive control tissues demonstrated our ability to

detect all three receptors by RT-PCR. LPAR1 was constitu-

tively transcribed from the morulae stage on GD 3 until the

adhesion-competent late blastocyst stage. LPAR2 was

expressed only in late blastocysts on GD 6 and 7, while

no LPAR3 was found at any stage.

LPA accelerates trophoblast differentiation by inducing

Cai2+ signaling

LPA induces Cai2+ transients through all three of its

receptors [12]. Previously, we found that LPA induces

Cai2+ transients in preimplantation mouse embryos [32].

Because Cai2+ transients induced by calcitonin or HB-EGF

are required for acceleration of trophoblast differentiation

[6,7], we examined the role of Cai2+ transients in LPA-

stimulated outgrowth. Embryos cultured to GD 5 were

ryos collected at the indicated times during development were subjected to

product sizes were 392, 521, and 382 bp, respectively. Mouse lung (A) or

(100 bp) appear in the right lane; arrowheads indicate 500 bp.

Fig. 4. Effect of BAPTA-AM on the acceleration of outgrowth by LPA.

Blastocysts were first treated for 30 min on GD 5 with 0 or 5 AM BAPTA-

AM, and then for an additional hour with 0 (open bars) or 1 (solid bars) AMLPA. After subsequent culture, the outgrowth rates (T50) were determined,

as in Fig. 1B. Each treatment was performed with at least 40 embryos in

three independent experiments. *P < 0.05 compared to 0 AM LPA.

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326 321

treated for 30 min with 5 AM BAPTA-AM, a chelator of free

cytosolic Ca2+, before a 1-h exposure to LPA. BAPTA-AM

treatment alone had no effect on the basal rate of outgrowth;

however, it attenuated the acceleration of outgrowth by LPA

Fig. 5. Transactivation of ErbB receptors by LPA. Blastocysts were treated for 1 h

10 Ag/ml of a function-blocking polyclonal antibody against HB-EGF (A), 0 or 0.

1478, or 10 Ag/ml of function-blocking antibodies against ErbB1 or ErbB4 (C). Af

1B. Each treatment was performed with at least 40 embryos in three independen

(Fig. 4). This demonstrates that, like calcitonin and HB-

EGF, LPA stimulation of trophoblast differentiation is

dependent on its ability to induce Cai2+ transients.

LPA bioactivity requires HB-EGF

Because LPA can transactivate the ErbB1 by inducing

HB-EGF shedding [21], we examined the acceleration of

trophoblast outgrowth by LPA using specific reagents to

inhibit the function of HB-EGF or its receptors. A function-

blocking antibody against HB-EGF had no effect on the basal

rate of outgrowth compared to the untreated control; howev-

er, using the 1-h LPA treatment regime onGD 5, this antibody

attenuated the increase in outgrowth rate (Fig. 5A). Treatment

with nonimmune IgG resulted in a T50 (40.56 F 4.47) that

was not significantly different (P < 0.05) than the controls.

Heparan sulfate is a cofactor for HB-EGF binding and

potentiates its biological effects [19]. Accordingly, hepariti-

nase treatment attenuated the acceleration of outgrowth

induced by LPA (Fig. 5B). To demonstrate that HB-EGF

accelerated trophoblast differentiation specifically through its

receptors, ErbB1 and ErbB4 [24], embryos were stimulated

with LPA after treatments to interfere with receptor function.

on GD 5 with 0 (open bars) or 1 (solid bars) AM LPA in the presence of 0 or

1 U/ml heparitinase I (B), 10 nM of the ErbB1 tyrosine kinase inhibitor AG

ter subsequent culture, the outgrowth rates (T50) were determined, as in Fig.

t experiments. *P < 0.05 compared to 0 AM LPA.

Fig. 6. Expression of HB-EGF in preimplantation embryos. Morulae collected on GD 3 were cultured until GD 7 and assessed at daily intervals for HB-EGF

mRNA (A), cellular HB-EGF protein (B), and accumulation of HB-EGF on the apical surface of the trophoblast (C). The PCR product size for HB-EGF

mRNAwas 627 bp, indicated by an arrowhead. The brightest band in a 100-bp DNA ladder shown in the left lane is 600 bp. Representative images are shown

of detergent-permeabilized (B) or nonpermeabilized (C) fixed embryos labeled with or without (Controls) 10 Ag/ml of antibody recognizing the extracellular

domain of HB-EGF, followed with Texas Red-conjugated secondary antibody. Fluorescent images of the permeabilized embryos (B) were deconvolved to

reveal cell-specific expression. Images in C were not deconvolved to show total surface staining. The location of the ICM is indicated in B with an asterisk.

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326322

ErbB1 inhibitory antibody, ErbB4 inhibitory antibody, or the

ErbB1 tyrosine kinase inhibitor AG1478 did not alter the

basal outgrowth rate; however, they each attenuated the

Fig. 7. Translocation of HB-EGF in LPA-stimulated blastocysts. Blastocysts were

or without (B) permeabilization, and labeled with anti-HB-EGF, as in Fig. 6. Flu

fluorescence intensity of embryos labeled with antibody to detect total HB-EGF, as

image analysis, as described in Materials and methods (C). *P < 0.05 compared

ability of LPA to accelerate the outgrowth rate, as did a

combination of the two antibodies (Fig. 5C). These findings

demonstrate that LPA advances trophoblast differentiation

treated on GD 5 with 1 AM LPA for 0–60 min, as indicated, fixed with (A)

orescent images of the permeabilized embryos (A) were deconvolved. The

in A (open bars), or surface HB-EGF, as in B (solid bars), was quantified by

to 0 min treatment.

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326 323

through a mechanism mediated by HB-EGF interaction with

either of its receptors, ErbB1 or ErbB4.

HB-EGF is expressed in murine preimplantation embryos

To clarify whether HB-EGF is expressed in mouse

preimplantation embryos, we examined its mRNA and

protein during blastocyst development. HB-EGF mRNA

was detected by RT-PCR throughout trophoblast differen-

tiation from GD 3 to 7 (Fig. 6A). The expression of HB-

EGF protein was examined in whole embryos by immu-

nofluorescence. Deconvolved images showed that HB-EGF

was expressed at a relatively low level, mainly in the

cytosol of both the ICM and trophoblast, from GD 3 to 5

Fig. 8. Regulation of HB-EGF surface mobilization by LPA. Blastocysts were trea

HB-EGF mRNA by RT-PCR (A) or for surface expression of HB-EGF by im

amplified by PCR, as in Fig. 6, for 25 cycles (A). A 100-bp DNA ladder appears

exposure, some blastocysts were treated for 30 min with 10 Ag/ml cycloheximid

antibody was added to embryos exposed to LPA for 45 min in the Control group

(open bars), C (solid bars), or D (hatched bars) was quantified by image analysis

and at a higher level on GD 6 and 7 (Fig. 6B). At the late

blastocyst stages, there appeared to be more staining along

the free plasma membrane, which was confirmed by

staining embryos without permeabilization (Fig. 6C). Sur-

face staining of HB-EGF was weak from GD 3 to 5 and

increased on GD 6–7. These data demonstrate that HB-

EGF is indeed available for transactivation by LPA during

blastocyst differentiation.

LPA induces transient accumulation of HB-EGF on the

apical surface

To obtain further evidence that LPA transactivates ErbB

signaling through HB-EGF mobilization, we directly mon-

ted on GD 5 with 1 AM LPA for 0–60 min, as indicated, and assessed for

munofluorescence without permeabilization (B–E). HB-EGF mRNA was

in the right lane with an arrowhead marking the 600-bp band. Before LPA

e (B), 10 Ag/ml brefeldin A (C), or 5 AM BAPTA-AM (D). No primary

. The fluorescence intensity of embryos labeled with anti-HB-EGF as in B

, as in Fig. 7C (E). *P < 0.05 compared to 0 min treatment.

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326324

itored HB-EGF expression and trafficking during blastocyst

exposure to LPA on GD 5. According to semiquantitative

immunofluorescence analysis, there was no change in HB-

EGF staining of permeabilized embryos up to 1 h after LPA

treatment (Figs. 7A, C); however, LPA induced a transient

increase (P < 0.05) in surface HB-EGF accumulation,

peaking after 45 min of LPA treatment (Figs. 7B, C). The

peak staining intensity of the surface HB-EGF was reduced

by adsorption with recombinant human HB-EGF, demon-

strating the specificity of HB-EGF staining (data not

shown). The adsorbed antibody control value (2.8) was

subtracted to calculate the specific fluorescence intensities

reported in Fig. 7C. Surface accumulation of HB-EGF

increased more than 2-fold at 45 min, then returned to basal

levels by 60 min, suggesting that the protein was secreted

after trafficking to the surface.

Regulation of HB-EGF translocation induced by LPA

To establish a mechanism for LPA-induced accumulation

of HB-EGF on the embryo surface, we examined the effects

of LPA on HB-EGF biosynthesis and trafficking in GD 5

blastocysts. HB-EGF mRNAwas amplified linearly between

25 and 35 cycles by PCR (data not shown), providing a semi-

quantitative estimate of relative mRNA level within that

amplification range. No discernable difference was found in

the HB-EGF amplicon band size after LPA treatment using 25

amplification cycles (Fig. 8A). Inhibition of protein synthesis

with cycloheximide failed to block LPA-induced accumula-

tion of HB-EGF on the embryo surface (Figs. 8B, E),

suggesting that LPAwas not increasing HB-EGF biosynthe-

sis. When brefeldin Awas used to prevent protein trafficking

within the cells [33], the accumulation of HB-EGF on the

embryo surface after LPA treatment was inhibited (Figs. 8C,

E). Similarly, interference with Cai2+ signaling by treating

embryos with BAPTA-AM also blocked LPA induction of

transient surface accumulation of HB-EGF (Figs. 8D, E).

These data demonstrate that LPA, possibly through its effect

on Cai2+ mobilization, initiates HB-EGF trafficking from an

intracellular location to the apical surface of trophoblasts,

where it becomes available for ectodomain shedding and

transactivation of ErbB1 and ErbB4.

Discussion

Cai2+ signaling accelerates early blastocyst differentiation

and the onset of adhesive capacity [6,7]. In this report, we

used LPA as a ligand to induce Cai2+ transients, which

accelerated trophoblast differentiation by transactivating

HB-EGF receptors, ErbB1 and ErbB4. Because chelation

of Cai2+ blocked both acceleration of the rate of develop-

ment by LPA and trafficking of HB-EGF, it is feasible that

other agents capable of mobilizing Cai2+ also accelerate

blastocyst development by transactivating ErbB receptors

through HB-EGF. Additional studies will be required to

fully test this hypothesis. If substantiated, it would indicate a

key role for ErbB1 and ErbB4 in regulating blastocyst

development.

A regulatory function for LPA during early mammalian

development has been suggested in previous studies. LPA

stimulates oocyte maturation [14], cavitation [15], and

embryo transport in the oviduct [16]. We now report that

it also accelerates blastocyst differentiation based on blas-

tocyst outgrowth rates. At least one LPA receptor, LPAR1,

was constitutively expressed during blastocyst formation

and differentiation, while another, LPAR2, appears during

the late blastocyst stage. The optimal effective LPA con-

centration correlated well with its physiological concentra-

tion in serum. We previously showed that LPA induces a

Cai2+ transient in mouse preimplantation embryos [32]. In

this study, we found that the developmental effects of LPA

at the blastocyst stage were attenuated by BAPTA-AM,

providing additional evidence that Cai2+ signaling modulates

blastocyst differentiation.

Our observations using immunofluorescence suggest that

intracellular HB-EGF trafficks to the plasma membrane in

response to LPA. The mobilization of HB-EGF appears to

take place without additional biosynthesis of the protein,

based on our observations of unaltered levels of HB-EGF

mRNA and the failure of cycloheximide to prevent surface

accumulation of HB-EGF after LPA exposure. Evidence of

HB-EGF trafficking was provided by the ability of brefeldin

A to prevent surface accumulation of HB-EGF after LPA

treatment. Trafficking of HB-EGF was also dependent on

Cai2+ signaling based on inhibition by BAPTA-AM. Cai

2+

transients can induce fusion of intracellular vesicles with the

plasma membrane [34]. Integrin-induced Cai2+ signaling

mediates vesicle trafficking in trophoblast cells at the late

blastocyst stage through a biochemical cascade that

increases trophoblast adhesion to fibronectin [35]. Similarly,

Ca2+-mediated trafficking appears to be induced by the LPA

receptor, positioning HB-EGF at the apical surface of

trophoblasts.

We observed that LPA-induced accumulation of HB-

EGF on the blastocyst surface was transient. Limitations

in the amount of available embryonic protein impede our

ability to determine whether HB-EGF was release into the

medium; however, there is now abundant data in other cell

types supporting a mechanism involving HB-EGF shedding

through ectodomain proteolysis to transactivate ErbB recep-

tors [20,21]. Our report that function-blocking antibodies

against HB-EGF, ErbB1, or ErbB4 each interfered with LPA

stimulation of blastocyst development is consistent with a

transactivation mechanism, although not conclusive. Re-

moval from the cell surface of heparan sulfate, which is

essential for HB-EGF activation of its receptors [36], also

prevented LPA stimulation of development. HB-EGF is the

only member of the EGF family that binds both ErbB1 and

ErbB4 in a heparin-dependent manner [37]. Clearly, HB-

EGF interaction with its receptors is necessary for stimula-

tion of mouse blastocyst differentiation, but the physical

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326 325

nature of the interaction is uncertain. Either proHB-EGF is

shed from the surface or it interacts in a juxtacrine manner

with receptors on adjacent trophoblast cells. Further studies

will be required to determine if the reduced expression of

HB-EGF on the surface of LPA-treated embryos after 60

min is the result of its shedding or endocytosis.

Antibodies against either ErbB1 or ErbB4 attenuated

LPA stimulation of blastocyst development, suggesting that

the transactivation of both receptors is required. Our data do

not reveal whether one receptor is more important than the

other. Both ErbB1 and ErbB4 are expressed in murine

blastocysts; however, ErbB4 is not translocated from the

cytosol to the apical surface of blastocysts until late on GD

4, while ErbB1 is resident on blastocyst apical surface much

earlier [7]. ErbB4 has a much higher affinity for HB-EGF

than ErbB1 [38]. Therefore, HB-EGF accelerates murine

trophoblast differentiation only after ErbB4 is translocated

to the apical surface [7]. In humans, HB-EGF significantly

improves blastocyst development and hatching [39]. Be-

cause ErbB1 is only expressed by the inner cell mass of

human blastocysts, while ErbB4 is expressed in trophoblast

[40], HB-EGF most likely functions through ErbB4 to

activate trophoblasts. These data suggest that ErbB4 has a

critical role in regulating trophoblast differentiation both in

mice and humans. However, our new data suggest that in

mice HB-EGF activity also requires ErbB1. Members of the

ErbB family differ in their initiation of downstream signal-

ing pathways [41]. Acceleration of blastocyst differentiation

may require the activation of pathways regulated by both

ErbB1 and ErbB4. Furthermore, ErbB1/4 heterodimers can

induce a more potent biological response and activate more

intracellular pathways than either homodimer [42], suggest-

ing a possible explanation for the requirement of both

receptors. Inhibition of ErbB1 kinase activity by AG1478

blocked the stimulation of blastocyst outgrowth perhaps

because it prevented transphosphorylation of ErbB4 in

ErbB1/4 heterodimers.

Our study demonstrates that, as in the hamster [25],

murine preimplantation embryos synthesize HB-EGF,

which could both regulate blastocyst development through

an autocrine mechanism and have profound paracrine

effects on the adjacent uterine epithelium. HB-EGF accu-

mulated at low levels between the morulae and mid-

blastocyst stages, and increased markedly in the late,

adhesion-competent blastocyst. While there is strong evi-

dence for a functional role of uterine epithelial HB-EGF in

maternal – embryonic interactions during implantation

[5,43], HB-EGF accumulating on the surface of the tro-

phoblast could provide an important local signal to the

endometrium. The uterus undergoes cellular and molecular

changes as it becomes receptive for blastocyst implanta-

tion; however, the presence of a blastocyst in the uterine

lumen specifically induces localized physiological changes

in uterine tissue through altered gene expression [44].

Among the uterine genes induced by the embryo are

Bmp2 and cyclooxygenase-2, as well as HB-EGF. Beads

coated with HB-EGF and placed in the mouse uterus to

simulate a blastocyst specifically induce many of the same

discrete local responses elicited by the embryo, including a

localized increase in vascular permeability, decidualization,

and expression of Bmp2 [45]. Therefore, trophoblast-de-

rived HB-EGF could operate as an autocrine embryotro-

phic factor during the preattachment phase of development

and later as a paracrine initiator of uterine differentiation at

the implantation site.

Acknowledgments

This research was supported by National Institutes of

Health grants HD 36764 and AA12057 to D.R.A. We thank

Brian Kilburn and Po Jen Chiang for their expert technical

assistance, Michael Kruger for help with statistics, and Dr.

Richard Leach, Wayne State University, for helpful

discussions.

References

[1] D.D. Carson, I. Bagchi, S.K. Dey, A.C. Enders, A.T. Fazleabas, B.A.

Lessey, K. Yoshinaga, Embryo implantation, Dev. Biol. 223 (2000)

217–237.

[2] D.R. Armant, J. Wang, Z. Liu, Intracellular signaling in the develop-

ing blastocyst as a consequence of the maternal–embryonic dialogue,

Semin. Reprod. Med. 18 (2000) 273–287.

[3] J. Wang, D.R. Armant, Integrin-mediated adhesion and signaling

during blastocyst implantation, Cells Tissues Organs 172 (2002)

190–201.

[4] Y.Q. Ding, L.J. Zhu, M.K. Bagchi, I.C. Bagchi, Progesterone stim-

ulates calcitonin gene expression in the uterus during implantation,

Endocrinology 135 (1994) 2265–2274.

[5] S.K. Das, X.N. Wang, B.C. Paria, D. Damm, J.A. Abraham, M.

Klagsbrun, G.K. Andrews, S.K. Dey, Heparin-binding EGF-like

growth factor gene is induced in the mouse uterus temporally by

the blastocyst solely at the site of its apposition: a possible ligand

for interaction with blastocyst EGF-receptor in implantation, Devel-

opment 120 (1994) 1071–1083.

[6] J. Wang, U.K. Rout, I.C. Bagchi, D.R. Armant, Expression of calci-

tonin receptors in mouse preimplantation embryos and their function

in the regulation of blastocyst differentiation by calcitonin, Develop-

ment 125 (1998) 4293–4302.

[7] J. Wang, L. Mayernik, J.F. Schultz, D.R. Armant, 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, Development 127

(2000) 33–44.

[8] J.J. Stachecki, F.D. Yelian, R.E. Leach, D.R. Armant, Mouse blas-

tocyst outgrowth and implantation rates following exposure to eth-

anol or A23187 during culture in vitro, J. Reprod. Fertil. 101 (1994)

611–617.

[9] W.H. Moolenaar, Lysophosphatidic acid, a multifunctional phospho-

lipid messenger, J. Biol. Chem. 270 (1995) 12949–12952.

[10] A. Tokumura, M. Miyake, Y. Nishioka, S. Yamano, T. Aono, K.

Fukuzawa, Production of lysophosphatidic acids by lysophospholi-

pase D in human follicular fluids of in vitro fertilization patients, Biol.

Reprod. 61 (1999) 195–199.

[11] Y. Xie, T.C. Gibbs, K.E. Meier, Lysophosphatidic acid as an auto-

crine and paracrine mediator, Biochim. Biophys. Acta 1582 (2002)

270–281.

Z. Liu, D.R. Armant / Experimental Cell Research 296 (2004) 317–326326

[12] J.J. Contos, I. Ishii, J. Chun, Lysophosphatidic acid receptors, Mol.

Pharmacol. 58 (2000) 1188–1196.

[13] X. Ye, I. Ishii, M.A. Kingsbury, J. Chun, Lysophosphatidic acid as a

novel cell survival/apoptotic factor, Biochim. Biophys. Acta 1585

(2002) 108–113.

[14] K. Hinokio, S. Yamano, K. Nakagawa, M. Iraharaa, M. Kamada, A.

Tokumura, T. Aono, Lysophosphatidic acid stimulates nuclear and

cytoplasmic maturation of golden hamster immature oocytes in vitro

via cumulus cells, Life Sci. 70 (2002) 759–767.

[15] T. Kobayashi, S. Yamano, S. Murayama, H. Ishikawa, A. Tokumura,

T. Aono, Effect of lysophosphatidic acid on the preimplantation de-

velopment of mouse embryos, FEBS Lett. 351 (1994) 38–40.

[16] K. Kunikata, S. Yamano, A. Tokumura, T. Aono, Effect of lysophos-

phatidic acid on the ovum transport in mouse oviducts, Life Sci. 65

(1999) 833–840.

[17] A. Tokumura, S. Yamano, T. Aono, K. Fukuzawa, Lysophosphatidic

acids produced by lysophospholipase D in mammalian serum and

body fluid, Ann. N. Y. Acad. Sci. 905 (2000) 347–350.

[18] X. Fang, M. Schummer, M. Mao, S. Yu, F.H. Tabassam, R. Swaby, Y.

Hasegawa, J.L. Tanyi, R. LaPushin, A. Eder, R. Jaffe, J. Erickson,

G.B. Mills, Lysophosphatidic acid is a bioactive mediator in ovarian

cancer, Biochim. Biophys. Acta 1582 (2002) 257–264.

[19] R. Iwamoto, E. Mekada, Heparin-binding EGF-like growth factor: a

juxtacrine growth factor, Cytokine Growth Factor Rev. 11 (2000)

335–344.

[20] K.L. Pierce, A. Tohgo, S. Ahn, M.E. Field, L.M. Luttrell, R.J. Lef-

kowitz, Epidermal growth factor (EGF) receptor-dependent ERK ac-

tivation by G protein-coupled receptors: a co-culture system for

identifying intermediates upstream and downstream of heparin-bind-

ing EGF shedding, J. Biol. Chem. 276 (2001) 23155–23160.

[21] T. Umata, M. Hirata, T. Takahashi, F. Ryu, S. Shida, Y. Takahashi, M.

Tsuneoka, Y. Miura, M. Masuda, Y. Horiguchi, E. Mekada, A dual

signaling cascade that regulates the ectodomain shedding of heparin-

binding epidermal growth factor-like growth factor, J. Biol. Chem.

276 (2001) 30475–30482.

[22] J. Dong, H.S. Wiley, Trafficking and proteolytic release of epidermal

growth factor receptor ligands are modulated by their membrane-an-

choring domains, J. Biol. Chem. 275 (2000) 557–564.

[23] S.M. Dethlefsen, G. Raab, M.A. Moses, R.M. Adam, M. Klagsbrun,

M.R. Freeman, Extracellular calcium influx stimulates metalloprotei-

nase cleavage and secretion of heparin-binding EGF-like growth fac-

tor independently of protein kinase C, J. Cell. Biochem. 69 (1998)

143–153.

[24] G. Raab, M. Klagsbrun, Heparin-binding EGF-like growth factor,

Biochim. Biophys. Acta 1333 (1997) F179–F199.

[25] X. Wang, H. Wang, H. Matsumoto, S.K. Roy, S.K. Das, B.C. Paria,

Dual source and target of heparin-binding EGF-like growth factor

during the onset of implantation in the hamster, Development 129

(2002) 4125–4134.

[26] D.R. Armant, H.A. Kaplan, W.J. Lennarz, Fibronectin and laminin

promote in vitro attachment and outgrowth of mouse blastocysts, Dev.

Biol. 116 (1986) 519–523.

[27] J.F. Schultz, D.R. Armant, Beta1- and beta3-class integrins mediate

fibronectin binding activity at the surface of developing mouse peri-

implantation blastocysts. Regulation by ligand-induced mobilization

of stored receptor, J. Biol. Chem. 270 (1995) 11522–11531.

[28] C. Pages, D. Daviaud, S. An, S. Krief, M. Lafontan, P. Valet, J.S.

Saulnier-Blache, Endothelial differentiation gene-2 receptor is in-

volved in lysophosphatidic acid-dependent control of 3T3F442A pre-

adipocyte proliferation and spreading, J. Biol. Chem. 276 (2001)

11599–11605.

[29] J.J. Contos, J. Chun, The mouse lp(A3)/Edg7 lysophosphatidic acid

receptor gene: genomic structure, chromosomal localization, and ex-

pression pattern, Gene 267 (2001) 243–253.

[30] Y. Umeda, Y. Miyazaki, H. Shiinoki, S. Higashiyama, Y. Nakanishi,

Y. Hieda, Involvement of heparin-binding EGF-like growth factor and

its processing by metalloproteinases in early epithelial morphogenesis

of the submandibular gland, Dev. Biol. 237 (2001) 202–211.

[31] D.L. Baker, D.M. Desiderio, D.D. Miller, B. Tolley, G.J. Tigyi, Direct

quantitative analysis of lysophosphatidic acid molecular species by

stable isotope dilution electrospray ionization liquid chromatography-

mass spectrometry, Anal. Biochem. 292 (2001) 287–295.

[32] J.J. Stachecki, D.R. Armant, Regulation of blastocoele formation by

intracellular calcium release is mediated through a phospholipase C-

dependent pathway, Biol. Reprod. 55 (1996) 1292–1298.

[33] L. Orci, M. Tagaya, M. Amherdt, A. Perrelet, J.G. Donaldson, J.

Lippincott-Schwartz, R.D. Klausner, J.E. Rothman, Brefeldin A, a

drug that blocks secretion, prevents the assembly of non-clathrin-

coated buds on Golgi cisternae, Cell 64 (1991) 1183–1195.

[34] J. Zimmerberg, F.S. Cohen, A. Finkelstein, Micromolar Ca2+ stim-

ulates fusion of lipid vesicles with planar bilayers containing a calci-

um-binding protein, Science 210 (1980) 906–908.

[35] J. Wang, L. Mayernik, D.R. Armant, Integrin signaling regulates

blastocyst adhesion to fibronectin at implantation: intracellular calci-

um transients and vesicle trafficking in primary trophoblast cells, Dev.

Biol. 245 (2002) 270–279.

[36] S. Higashiyama, J.A. Abraham, M. Klagsbrun, Heparin-binding EGF-

like growth factor stimulation of smooth muscle cell migration: de-

pendence on interactions with cell surface heparan sulfate, J. Cell

Biol. 122 (1993) 933–940.

[37] Y. Yarden, The EGFR family and its ligands in human cancer. sig-

nalling mechanisms and therapeutic opportunities, Eur. J. Cancer 37

(Suppl. 4) (2001) S3–S8.

[38] B.C. Paria, K. Elenius, M. Klagsbrun, S.K. Dey, Heparin-binding

EGF-like growth factor interacts with mouse blastocysts indepen-

dently of ErbB1: a possible role for heparan sulfate proteoglycans

and ErbB4 in blastocyst implantation, Development 126 (1999)

1997–2005.

[39] K.L. Martin, D.H. Barlow, I.L. Sargent, Heparin-binding epidermal

growth factor significantly improves human blastocyst development

and hatching in serum-free medium, Hum. Reprod. 13 (1998)

1645–1652.

[40] K. Chobotova, I. Spyropoulou, J. Carver, S. Manek, J.K. Heath, W.J.

Gullick, D.H. Barlow, I.L. Sargent, H.J. Mardon, Heparin-binding

epidermal growth factor and its receptor ErbB4 mediate implantation

of the human blastocyst, Mech. Dev. 119 (2002) 137–144.

[41] Y. Yarden, M.X. Sliwkowski, Untangling the ErbB signalling net-

work, Nat. Rev., Mol. Cell Biol. 2 (2001) 127–137.

[42] I. Alroy, Y. Yarden, The ErbB signaling network in embryogenesis

and oncogenesis: signal diversification through combinatorial li-

gand– receptor interactions, FEBS Lett. 410 (1997) 83–86.

[43] G. Raab, K. Kover, B.C. Paria, S.K. Dey, R.M. Ezzell, M. Klagsbrun,

Mouse preimplantation blastocysts adhere to cells expressing the

transmembrane form of heparin-binding EGF-like growth factor, De-

velopment 122 (1996) 637–645.

[44] B.C. Paria, J. Reese, S.K. Das, S.K. Dey, Deciphering the cross-talk

of implantation: advances and challenges, Science 296 (2002)

2185–2188.

[45] B.C. Paria, W. Ma, J. Tan, S. Raja, S.K. Das, S.K. Dey, B.L. Hogan,

Cellular and molecular responses of the uterus to embryo implantation

can be elicited by locally applied growth factors, Proc. Natl. Acad.

Sci. U. S. A. 98 (2001) 1047–1052.