505CHENG ET AL. Biol Res 42, 2009, 505-516Biol Res 42: 505-516, 2009 BRCharacterization of Calreticulin Expression in MouseEndometrium during Embryo Implantation

SHU-QUN CHENG, JUN-LIN HE, YAN-LING DONG, XUE-QING LIU, YU-BIN DING,RU-FEI GAO, YI TAN, QIAN YE, ZHEN-LING TIAN and YING-XIONG WANG

Laboratory of Reproductive Biology, Chongqing Medical University, No.1 Yixueyuan Road, YuzhongDistrict, Chongqing 400016, People’s Republic of China

ABSTRACT

Calreticulin (CRT), a Ca2+-binding storage protein and chaperone in the endoplasmic reticulum, modulatescell adhesiveness and integrin-dependent Ca2+ signaling. However, the role of CRT during implantationremains poorly understood. In the present study, we characterized the expression of CRT mRNA and theprotein in mouse endometria from pregnancy D1 to D7. Real-Time PCR and in situ hybridization resultsshowed that the levels of CRT mRNA in the endometria of pregnant mice were significantly higher than thoseof non-pregnant mice (P<0.05), and increased gradually from pregnancy D1 to D4, reaching the maximumlevel on D4, followed by a plateau from D4 to D7. Using immunofluorescence histochemistry and westernblot, changes of CRT expression in the endometria of pregnant mice were consistent with the expression ofCRT mRNA. Furthermore, antisense CRT oligodeoxynucleotide was injected into the uterus horns of pregnantmice (D3) to investigate its effect on embryo implantation. The result showed that the number of implantedembryos markedly decreased in the side of uterine horns receiving antisense CRToligodeoxynucleotide(P<0.05). These findings suggest that CRT may play an important role in embryoimplantation in mice.

Key terms: Calreticulin; Endometrium; Embryo implantation; Mouse.

Corresponding authors: Prof. Ying-Xiong WANG and Jun-Lin HE, Laboratory of Reproductive Biology, ChongqingMedical University, No.1 Yixueyuan Road, Yuzhong District, Chongqing 400016, People’s Republic of China. Phone:86-23-68485926; Fax: 86-23-68485008; Email: wyx61221@yahoo.com.cn. hejunlin_11@yahoo.com.cn

Received: February 18, 2009. In Revised form: August 11, 2009. Accepted: October 3, 2009.

INTRODUCTION

Signals and molecular pathways responsiblefor endometrial development andreceptivity to embryonic implantation areimportant issues in reproductivebiomedicine because of their implicationfor treatment of infertility and assistedfertilization and implantation (Wang andDey, 2006). Successful implantationdepends on synchronization between thecomplex molecular and cellular eventsassociated with development anddifferentiation of the blastocyst, and thereceptivity of the endometrium. Theseevents are finely regulated by a series ofsignals and molecules, including ovariansteroids and cytokines (Krüssel et al, 2003).Calreticulin(CRT) is an ancient and highly

conserved Ca2+-binding chaperone withmore than 90% amino acid homology inmammals (including human, rabbit, rat andmouse), which is expressed in various cellsand tissues (Bedard et al, 2005). Previousstudies have indicated that expression ofCRT is associated with increased Ca2+

storage capacity of the endoplasmicreticulum (ER), Ca2+-binding chaperones inthe lumen of the ER affect luminal Ca2+

concentrations [Ca2+ER] (Mesaeli et al,1999; Meldolesi et al, 1998) and thatagonist-induced changes in [Ca2+ER] affectER function (Corbett et al, 2000).Coppolino and Papp et al reported that CRTis involved in cell adhesion via theinduction of vinculin and N-cadherinexpression (Coppolino et al, 1995; Papp etal, 1999). It also affects cell adhesion

CHENG ET AL. Biol Res 42, 2009, 505-516506

because transient down-regulation of CRTreduces attachment of cells to extracellularmatrix substrata (Otteken et al, 1996). Inaddition, CRT modulates integrin-dependent Ca2+ signaling and steroid-sensitive gene expression (Michalak et al,1999). During the initial stages of embryoimplantation, integrins have central roles inthe regulation of adhesion and migratoryevents at the embryo-maternal interface(Burrows et al, 1996). Integrins are capableof influencing and mediating adhesion,migration, invasion, cytoskeletonreorganization and cellular signaling, whichare all critical for embryo implantation, viabinding to various extracellular matrix(ECM) components and cell adhesionmolecules (Burghardt et al, 2002). Thesefindings suggest that CRT may play animportant role in the process of embryoimplantation. However, the expression andthe role of CRT in the uterus during earlypregnancy remain unknown. This studyaims to examine the expressioncharacteristics of CRT in the mouseendometrium during early pregnancy andthe effect of CRT on embryo implantationin order to probe the role of CRT inimplantation.

MATERIALS AND METHODS

Animals and endometria isolation

Female and male NIH mice (6-to-8-wk-old)were obtained from the Laboratory AnimalCenter of Chongqing Medical University[Chongqing, China (Certificate: SCXK(YU) 2006002]). All mice were housed inspecific pathogen free (SPF) rooms with acontrolled light schedule (12L: 12D) andcontrolled temperature range (22-25ºC)with free access to food and water. Allanimal procedures were approved by theEthics Committee, State Key Laboratory ofReproductive Biology, Institute of Zoology,Chinese Academy of Sciences. The estrusfemales, whose estrous cycles wereidentified by analysis of exfoliated vaginalcells under a microscope, were bred withfertile males (female to male: 2 to 1)overnight (12h) and the vaginal plugs were

checked the following morning. The day ofvaginal plug was considered as Day 1ofpregnancy (D1). Pregnant mice weredivided into seven groups (D1, D2, D3, D4,D5, D6, and D7) with 20 mice in eachgroup, 20 non-pregnant estrus mice werechosen as controls (D0). Twentysupernumerary D3 mice were used for CRTfunction study.

Mice in each group were killed between0800-0900 h by cervical dislocation afterintraperitoneal injection of 1% butaylone.Uterine horns were immediately dissectedand flushed with saline, and thenendometrial tissues were squeezed out fromuterine horns by use of fine forceps under ananatomical microscope according to theprocedures described by Tan Y (Tan Y et al,2001). The isolated endometrial tissues of 10mice in each group were immediately storedin liquid nitrogen for Real-Time PCR andwestern blot. The tissues of the other 10mice in each group were fixed in 4%paraformaldehyde, embedded in paraffin forimmunofluorescence histochemistry and insitu hybridization. In addition, another 20D3 mice received uterine horns injection ofCRT antisense oligonucleotide for CRTfunction study.

Real-Time PCR of CRT mRNA in miceendometria

RNA extraction

Total RNA of mouse endometria wasextracted using the total RNA isolationreagent (TRI reagent, Promage, Madison,WI, USA) according to the manufacturer’sinstructions. Quantification and purity ofthe total RNA was examined using UVspectrophotometer, A260/A280≥ 1.8 wasthe standard. Integrity of the total RNA wasidentified by agarose gel electrophoresis.

Real-Time PCR

The reverse transcription of cDNA wasdone in a 25μl reaction system, containing1 μl RNA (1μg/μl), 500 ng Oligo-dT, 5 ×buffer 5 μl, M-MLV (200 U/μl) 1 μl, RNAi(25 U)0.5 μl, dNTPs (10 mmol/L) 5 μl,added DEPC-H2O to 25 μl. The reaction

507CHENG ET AL. Biol Res 42, 2009, 505-516

conditions were for 5 min at 70 ºC, 60 minat 42ºC and 10 min at 70 ºC. The primersfor CRT and β-actin were designed usingthe primer design program (Primer Premier5) and synthesized by DingguoBioTechnologies Co, Ltd., as shown intable 1. Quantification of the transcriptswas carried out in the iQ5 Real-Time PCRDetection System (Bio-Rad, Hercules, CA).Templates for PCR were obtained byreverse transcription of cDNA. Real-TimePCR was performed in a total 25 μl reactionsystem (the SYBR Green I PCR Reagentkit , TaKaRa, Dalian). The reactionconditions were for 3min at 94°C, 5s at94°C and 30s at 54°C, for 40 cycles. Thefluorescence signal of 3rd to 15th cycleswas settled as a fluorescent backgroundsignal and the baseline was adjustedaccordingly. Relative quantification of theexpression level of CRT and β-actin mRNAwere calculated by the following method:for each reaction to amplify CRT targetgene and β-actin gene, the Ct value wasdetermined. The ratio of each CRT targetmRNA to β-actin mRNA was calculated bythe formula 2-ΔΔCt. The gene expression ratewas obtained by normalizing the amount ofCRT mRNA with that of β-actin mRNA.

In situ hybridization

Digoxigenin-labeled sense and probes ofwere designed and synthesized by DingguoBioTechnologies Co, Ltd. The sequence ofantisense CRT probe is: 5’-tgggactttctgccacccaagaagataaaggaccctgatgctgccaagcc-3’. Insitu hybridization was carried out accordingto the manufacturer’s instructions. Themouse endometria samples were fixedimmediately in 4% paraformaldehyde and

embedded in paraffin. Sections (7μm) werecut, de-paraffinized and rehydrated. Thesections were digested with 0.1 mg/mlproteinase K for 7 min at 37°C, and washedthree times with PBS. The next stepconsisted of the prehybridization inhybridization buffer for 3 hr at 42°C andhybridization at 42°C overnight inhybridization buffer containingdigoxigenin-labeled sense and antisenseprobe of CRT at a final concentration of20ng/ul. After hybridization, the sectionswere washed for 10 min with 2×SSC,0.5×SSC for 15 min and 0.2×SSC for15min at 37°C and incubated in 5% (w/v)bovine serum albumin for 30 min at roomtemperature, then incubated respectively inanti-rabbit CRT antibody labeled withdigoxigen (1: 100) and goat anti-rabbit IgGconjugated to alkaline phosphatase(1: 100)for 60 min at 37°C, washed in PBS for 5min. The signal was visualized with 5-bromo-4-chloro-3-indolyl phosphate andnitroblue tetrazolium. The signal intensityof CRT mRNA expression was quantifiedusing Chengdu Jinpan Media Mix ImageAnalysis Software 1.0 version (Chengdu,China).

Immunofluorescence histochemistry

The mouse endometria samples were fixedimmediately in 4% paraformaldehyde andembedded in paraffin. Sections (8μm) werecut, deparaffinized and rehydrated. Thesections were soaked in 3% H2O2 solutionfor 15 min at room temperature to eliminateendogenous peroxidase activity. Thesections were placed in 0.01M citrate buffer(pH 7.0) for 15 min at 95°C for antigenretrieval, and then rinsed at least three

TABLE 1

Sequences of primers for real-time PCR

Gene Primer sequence Product size

Calreticulin F: 5’-TACGCACTGTCCGCCAAAT-3’ 132bp

R: 5’-GTCCAAACCACTCGGAAACAG-3’

β-actin F: 5’-CCTGAGGCTCTTTTCCAGCC-3’ 110bp

R: 5’-TAGAGGTCTTTACGGATGTCAACGT-3’

CHENG ET AL. Biol Res 42, 2009, 505-516508

times with PBS. The sections wereincubated with monoclonal anti-rabbit CRTantibody (1: 30 dilution with PBS) (fromSanta Cruz Biotechnology Inc) for 2 h at37°C, washed for 10 min with PBS. Thesections were incubated with a secondaryantibody of goat anti-rabbit IgG labeledwith fluorescein isothiocyanate (FITC) for2 h at 37°C (Santa Cruz, CA, USA),following extensive washing with PBS. Asa negative control, CRT primary antibodywas replaced with PBS. The localization ofCRT protein expression was analyzed undera confocal laser scanning microscope(LEICA TCS SP2, Germany) and the levelsof CRT expression were quantified withChengdu Jinpan Media Mix Image AnalysisSoftware 1.0 version (Chengdu, China).

Western blot analysis

Total protein was extracted from the mouseendometria of each group using the totalprotein Tripure reagent kit (Dalian TaKaRa,China) according to the manufacturer’sinstructions. Protein concentration wasmeasured by the Bradford assay. Proteinsamples (30μg each sample) were subjectedto 10% SDS-polyacrylamide gelelectrophoresis and then transferred ontopolyvinylidene difluoride membrane using aBio-Rad electroblot apparatus (Bio-Rad,Beijing, China). The membrane wasincubated in blocking PBS containing 0.05%Tween-20(TBST) and 5% nonfat milk for 1h at room temperature. Immunoblot analyseswere accomplished by incubating themembranes with monoclonal rabbit anti-CRT antibody (1: 2000 dilution in 5% milk-TBST) (Santa Cruz, CA, USA) overnight at4°C, followed by incubating with secondaryantibody (goat anti-rabbit IgG) conjugatedwith horseradish peroxidase (ZhongshanBiosciences Inc, Beijing, China) for 2 h at37°C. After a final washing in TBST for 1 h,protein bands were visualized bydiaminobenzidine tetrahydrochloride (DAB).Densitometry analysis of CRT expressionwas performed using Quantity One version4.4.0 analysis software. Primary rabbit anti-CRT antibody was replaced with rabbit anti-actin antibody (Zhongshan Biotechnology)as an internal control.

CRT function study during early embryoimplantation

To further determine whether the CRT mayplay a role in embryo implantation, thefemale mice of pregnancy D3 were given auterine horn injection of antisense CRToligodeoxynucleotide according to theprocedures described by Zhu (Zhu LJ, et al.1998). The oligodeoxynucleotides weredesigned and synthesized (Dalian TaKaRa,China) according to the sequence of mouseCRT gene cDNA. The sequences ofoligodeoxynucleotides were: sense CRToligodeoxynucleotide: 5’-CGGCCCGCCATGCTCCTTTCG-3’; antisense CRToligodeoxynucleotide: 5’-CGAAAGGAGCATGGCGGGCCG-3’. The antisense CRToligodeoxynucleotide was complementary tobases 125-145 bp within exon I of the mouseCRT (GenBank Accession Number:NM_007591.3). Analysis of homologybetween the synthesized oligomer and therodent sequences present in GenBank databases by Genetics Computer Group sequenceanalysis software package revealed that thesynthesized oligomer was specificallycomplementary only to CRT mRNA.

Antisense CRT oligodeoxynucleotide of81ug was dissolved in 50ul 0.9% NaClsolution for stock solution (solution A).Lipofectamine2000 6ul was diluted with50ul 0.9% NaCl solution (solution B). Aworking solution was prepared by mixture ofthe two solutions before injection. Afterfemale mice were anesthetized, disinfectedand underwent midventral laparotomy on D3of pregnancy between 0800-0900 h, the rightuterine horn of each mouse (n=10) was fixedwith forceps, and the needle was very slowlyintroduced as close as possible to the cervixin the uterine horn. The working solution(8μl) containing antisense CRToligodeoxynucleotide was injected into theright uterine horn of each animal. Eachanimal served as her own control, with theleft uterine horn receiving the same volumeof sense CRT oligodeoxynucleotide solution.In addition, the two sides of uterine of eachmouse (n=10) on pregnancy D3 were onlyinjected with 8μl of 0.9% NaCl solution as aloading control. On D8 of pregnancy, allanimal were killed between 8: 00AM-9:

509CHENG ET AL. Biol Res 42, 2009, 505-516

00AM by cervical dislocation afterintraperitoneal injection of 1% butaylone,and then the uterine were removed bylaparotomy and the number of implantedembryos was recorded in each uterine horn.

Statistical analysis

All statistical data obtained from theexperiment were analyzed using theStatistical Package for the Social Sciencessoftware (SPSS for Windows packagerelease 10.0, SPSS Inc., Chicago, IL, USA).Statistical significance was determined usingone-way ANOVA. Post-Hoc comparisonsbetween groups were made by Fisher’sprotected least-significance-difference test.Values were means ±S.D. Differences wereconsidered significant at P<0.05.

RESULTS

CRT mRNA expression in mouseendometria during early pregnancy

The level of CRT mRNA was detected inendometria of pregnant and non-pregnantmice by Real-Time PCR. The result showedthat the relative ratio of CRT mRNA on D1,

D2, D3, D4, D5, D6, D7 was 1.45±0.013,1.53±0.015, 1.65±0.011, 1.98±0.009,1.97±0.010, 1.88±0.012 and 1.87±0.010 ,respectively (relative ratio of CRT mRNAon D0 was set 1 as a reference, Fig.1). Thelevel of CRT mRNA of pregnant miceshowed a gradual increasing trend as dayspassed, reaching the maximum level on D4,followed by a plateau in CRT mRNAexpression. Analysis of variance showedthat the expression of CRT mRNA hassignificant differences between pregnantand non-pregnant mice (P<0.05). In situhybridization of the endometria sectionsfurther confirmed the results obtained byReal-Time PCR. CRT mRNA was mainlyexpressed in the luminal epithelium,glandular epithelium and stroma cellsduring this period (Fig. 2). No signal wasdetected when the antisense probe wasreplaced with the sense probe (Fig. 2NC).The weak signal of CRT mRNA was presentin the endometrial gland epithelium andstroma cells of non-pregnant mice (Fig.2).

Expression of CRT protein in theendometrium

The quantitative change of CRTexpression in endometria was estimated by

Figure 1: Real-Time PCR analysis of CRT mRNA expression in mouse endometria Real-TimePCR analyses of CRT mRNA expression from D0 to D7. Gene expression values from pregnancyendimetria were expressed relative to D0 values and represent the means ±SD from 20 separatemouse endometria of each group, respectively. Levels of β-actin mRNA were used as an internalcontrol. D0 means non-pregnancy; D1 to D7 mean day 3, day 4, day 5, day 6, day 7 of pregnancy,respectively. D1, D2, D3 and D4 versus D0, respectively (* P<0.05).

CHENG ET AL. Biol Res 42, 2009, 505-516510

Figure 2: In situ hybridization for CRT mRNA expression in mouse endometria The locationand the level of CRT mRNA were examined by in situ hybrididization using DIG-labeledoligonucleotide probe. Blue staining was determined as positive. (A): NC represented negativecontrol; D0 means non-pregnancy; D1, D2, D3, D4, D5, D6 and D7 represented Day 1, Day 2, Day3, Day 4, Day 5, Day 6 and Day 7 of pregnancy, respectively. The level of CRT mRNA increasedgradually from D1 to D7, reached a maximum level on D4, followed by a plateau was maintainedfrom D4 to D7. L= luminal epithelium; S= stromal cell; G= glandular epithelium. (B): The signalintensity of CRT mRNA was quantified using Chengdu Jinpan Media Mix Image Analysis Software1.0 version (Chengdu, China). D1, D2, D3 and D4 versus D0, respectively (* P<0.05).

511CHENG ET AL. Biol Res 42, 2009, 505-516

Western blot, as shown in Fig. 3. Theprotein level was gradually increased fromD1 to D7 in a time-dependent manner,reached the maximal level on D4 (Fig. 3),just the time of implantation window.Densitometric analysis indicated that theintensity of CRT expression was higher inpregnant mice than in non-pregnant mice(P<0.05). Immunofluorescencehistochemistry analysis showed that TheCRT protein was mainly located in theuterine luminal, glandular epithelium andstromal cells. The highest expression wasalso observed on D4 (Fig. 4). The levels

and change of CRT protein expression alsocoincided with the mRNA levels detectedby Real-Time PCR.

Functional study of CRT during embryoimplantation

The effect of blockage of CRT geneexpression on embryonic implantation wasinvestigated using an antisense CRToligodeoxynucleotide as a blocker on D3.The results indicated that the number ofimplanted embryos in the right uterinehorns receiving antisense CRT

Figure 3. Western blot for CRT protein expression in mouse endometria The levels of CRTexpression during early pregnancy were determined by western blotting using anti-CRT antibody(from Santa Cruz Biotechnology Inc). (A): d0: non-pregnancy; d1, d2, d3, d4, d5, d6 and d7represented Day 1, Day 2, Day 3, Day 4, Day 5, Day 6 and Day 7 of pregnancy, respectively;Levels of β-actin were used as internal control. (B): Densitometry analysis of CRT expression wasperformed using Quantity One version 4.4.0 analysis software. The level of CRT expressionincreased gradually from D1 to D7, reaching a maximum level on D4, followed by a plateau thatwas maintained from D4 to D7, D1, D2, D3 and D4 versus D0, respectively (*P<0.05)

CHENG ET AL. Biol Res 42, 2009, 505-516512

Figure 4. Imunofluorescence histochemistry for CRT protein expression in mouse endometriaThe location and the level of CRT protein in endometria during early pregnancy were determinedby Imunofluorescence histochemistry using anti-rabbit CRT antibody (Santa Cruz BiotechnologyInc). Green staining was determined as positive. (A): D0: non-pregnancy; D2, D4 and D7represented Day 2, Day4 and Day7 of pregnancy, respectively; L= luminal epithelium; S= stromalcell; G= glandular epithelium. (B): The fluorescence intensity of CRT protein expression wasquantified using Chengdu Jinpan Media Mix Image Analysis Software 1.0 version (Chengdu,China). D1, D2, D3 and D4 versus D0, respectively (*P<0.05)

513CHENG ET AL. Biol Res 42, 2009, 505-516

Figure 5. Effect of antisense CRT oligodeoxynucleotide on number of implanted embryos:The function of CRT was blocked by uterine horn injection of antisense CRT oligodeoxynucleotideat 0800-0900 h on D3, and then the number of implanted embryos was recorded on D8 bydissection. (A): Representative uterine horns from one mouse of the treatment group (n=10), theright horn: receiving antisense CRT oligodeoxynucleotide; the left horn: receiving sense CRToligodeoxynucleotide. The number of implanted embryos in the right uterus was markedly lowerthan that of the left (*P<0.05). (B): Representative uterine horns from one mouse of the loadingcontrol group (n=10) by injecting 0.9% NaCl solution on both sides of the uterus. The saline did notaffect implantation (P>0.05). (C): The mean number (±SD) of implanted embryos in the uterinehorns of the treatment- and control groups. Unilateral antisense CRT oligodeoxynucleotideadministration significantly reduced implantations in the injected horn (*P<0.05). The numbers ofimplanted embryos in the uterine horns receiving sense CRT oligodeoxynucleotide and salinesolution treatment were not significantly different (P>0.05).

CHENG ET AL. Biol Res 42, 2009, 505-516514

oligodeoxynucleotide was significantlylower than in the left uterine hornsreceiving sense CRT oligodeoxynucleotide(the right: 2.92±1.02; the left: 7.66±0.93;P<0.05; Fig. 5A). In addition, the numberof implanted embryos in uterine horns ofthe right and the left receiving 0.9% NaClsolution did not decrease (7.93±0.86 and7.45±0.87, respectively. P>0.05; Fig. 5B).These data showed that the total number ofimplanted embryos was affected by uterushorn injection with antisense CRToligodeoxynucleotide (*P<0.05; Fig. 5C).

DISCUSSION

Endometrial receptivity towards embryoimplantation is a complex event thatinvolves various processes, such asblastocyst adhesion, trophoblast invasion,decidualization and cell-to-cell interaction,controlled by a variety of moleculesproduced by endometrium, embryo andovary (Liu YX, 2004). The dialog betweenthe ovary and the endometrium provides thehormonal environment for establishment ofa successful pregnancy. The hormonesestrogen and progesterone act coordinatelyto stimulate the expression of keymolecules necessary for embryos to attachto and invade. It is thought that initialattachment of the embryo involves celladhesion events. The best characterized celladhesion molecule on the luminal surface ofthe endometrium is the alpha-beta 3integrin. CRT can interact with a-subunitsof integrin receptors via the highlyconserved KXGFFKR sequence in thecytoplasmic domain and is considered apotential integrin regulatory protein. Thisinteraction is important in mediating cellattachment and spreading on extracellularmatrix (EMC) substrates (Coppolino et al.,1999; Reilly et al, 2004). Integrins are alarge family of transmembraneheterodimeric proteins that mediate cell-celland cell-ECM adhesion by its interactionwith the ECM including vitronectin,fibronection. Integrins are considered asmarkers of the implantation window andcontribute to the transformation ofendometrium from the non-adhesion to the

adhesion status for embryo adhesion andimplantation (Wang and Armant, 2002;Lessey and Arnold, 1998). These findingsstrongly imply that CRT may be involvedin embryo implantation. Consequently, weexamined the characteristics of CRTexpression in the endometrium during earlypregnancy. Our results demonstrated thatlevels of CRT mRNA and its protein in theendometria of pregnant mice weresignificantly higher than those of non-pregnant mice, and increased graduallyfrom D1 to D7, reaching the maximumlevel on D4 and D5 (just aroundimplantation window), followed bymaintaining high levels on D6 and D7.Expression of this molecule were mainlyobserved in the luminal, glandular epitheliaand stromal cells and showed a temporaland spatial pattern of relationship with theestablishment of early pregnancy, thehighest level appeared on D4. The previousstudy indicated that the adhesive reaction ofblastocyst trophectoderm with the luminalepithelium initiate the implantation. Earlyevent of implantation occurs on themidnight of D4 or early morning of D5 inmice (Wang X, et al, 2004). The expressionof integrins is necessary for the adhesionand invasive capability of blastocyst toendometrium during the implantationwindow (Lessey et al, 1992; Lessey et at,1994) and changes in CRT expression inendometrium may regulate integrin functionand cell adhesion to facilitate theimplantation process (Leung-Hagestei et al,1994; Fadel et al, 1994; 2001). Our resultsshow that the highest level of CRTexpression in endometrium is concordantwith “the implantation window” (on D4 andD5) and a role for CRT interacting withintegrins to regulate blatocyst adhesion andinvasion into uterine endometrium. Theobservation strongly implied that CRT maybe a critical factor during implantation.Given the increase of the CRT levels in theuterus, even when the embryos are still inthe oviduct on pregnancy, we think that theembryo, as a stimulus, may induce a seriesof physiologic reactions of endometrium,including production of a variety ofmolecules benefitting from endometrial celldifferentiation, blastcyst adhesion,

515CHENG ET AL. Biol Res 42, 2009, 505-516

trophblast invasion and decidualization.The event is very important forimplantation priming and protection ofimplanted embryo when the embryo entersthe uterus. Although the embryos are still inthe oviduct on pregnancy D1, a series ofmolecules related to implantation, includingCRT, begin to be expressed in endometriumin order to provide a suitable endometrialcircumstance for the implantation.

The results of the present study indicatethat a temporal-spatial pattern of CRTexpression in endometria during theimplantation process might be essential forimplantation. The presence of the embryoin the uterus as a stimulus may be importantfor increasing CRT expression. One canexpect that a reduction of CRT expressioncould prevent acquisition of a receptivestate of the endometrium, leading to afailure of implantation. Therefore, wedesigned an experiment directly injectingantisense CRT oligodeoxynucleotide intopregnant mouse uterus on D3 in order toinvestigate the function of CRT duringimplantation. Previous study confirmed thatit is effective for inhibition of geneexpression through uterine horns injectingoligodeoxynucleotide (Zhu et al, 1998).Especially recently, Yuan et al certifiedfurther that oligodeoxynucleotide can beeffectively taken up by the uterine luminal,glandular epithelium and stromal cells(Yuan JX et al, 2009). It is known thatantisense oligodeoxynucleotide has a half-life of 24-28 hours in certain tissues(Wagner RW, 1994). Therefore, one mayexpect that the antisense CRToligodeoxynucleotide by uterine hornsinjection on D3 is effective for suppresionof CRT expression in the subsequent Day4-Day 5 of pregnancy (implantationwindow). In this study, the number ofimplanted embryos was significantly lowerin the uterus horn injection of antisenseCRT oligodeoxynucleotide on D3. Thestatistical analysis of the difference in thenumber of implanted embryos between theantisense- and sense oligodeoxynucleotide-treated groups showed that embryoimplantation was indeed inhibited with thetreatment of CRT antisenseoligodeoxynucleotide (P<0.05). Our

findings strongly imply that CRT may beinvolved in the initial stages of embryoimplantation. However, it can not be ruledout that the CRT antisense oligonucleotidemay have direct inhibitory effects on theadhesive character of integrins.

In summary, our data have demonstrateda significant increase of CRT expression onD4 and D5. It seems that CRT plays animportant role in the implantation processin rodents and this may be similar inhumans. However, further studies arenecessary to reveal the molecularmechanisms of CRT involvement in thisprocess.

ACKNOWLEDGEMENTS

We are grateful to Prof. Thomas G forcritically reading the manuscript and themembers of our laboratory for many helpfuldiscussions. This study was supported byNational Natural Science Foundation(No.30270510, 30700898) and NaturalScience Foundation of Chonqqing (No.2004-47, CSTC2007BB5286).

REFERENCES

[1] Wang H , Dey SK. Roadmap to embryo implantation:clues from mouse models. Nat Rev Genet, 2006, 7:185-199

[2] Krüssel JS, Bielfeld P, Polan ML, Simón C. Regulationof embryonic Implantation. European Journal ofObstetrics and Gynecology and Reproductive Biology,2003, 110: 2-9

[3] Bedard K, E Szabo E, Michalak M, Opas M. Cellularfunctions of endoplasmic reticulum chaperonescalreticulin, calnexin, and ERp57. Int Rev Cytol, 2005,245: 91-121

[4] Mesaeli N, Nakamura K, Zvaritch E, Dickie P, Dziak E,Krause KH, Opas M, MacLennan DH, Michalak M.Calreticulin is essential for cardiac development. J CellBiol.1999, 144: 857-868

[5] Meldolesi J, Pozzan T. The endoplasmic reticulumCa2+ store: a view from the lumen. Trends BiochemSci, 1998, 23: 10-14

[6] Corbett, EF, Michalak M. Calcium, a signalingmolecule in the endoplasmic reticulum? TrendsBiochem Sci, 2000, 307-311

[7] Coppolino M, Leung-Hagesteijn C, Dedhar S, WilkinsJ. Inducible interaction of integrin a2131 withcalreticulin-dependence on the activation state of theintegrin. J Biol Chem, 1995, 270: 23132-23138

[8] Papp S, Fadel MP, Kim H, McCulloch CA, Opas M.Calreticulin affects focal contact-dependent but notclose contact-dependent cell-substratum adhesion. JBiol Chem, 1999, 274: 15085-15094

CHENG ET AL. Biol Res 42, 2009, 505-516516

[9] Otteken A, Moss B. Calreticulin interacts with newlysynthesized human immunodeflciency virus type 1envelope glycoprotein, suggesting a chaperone functionsimilar to that of calnexin. J Biol Chem, 1996, 271: 97-103

[10] Michalak M, Corbett BF, Mesaeli N. Calreticulin: oneprotein, one gene, many functions. Biochem J, 1999,344: 281-292

[11] Burrows TD, King A, Loke YW. Trophoblastmigration during human placental implantation. HumReprod, 1996, Update 2 : 307-321

[12] Burghardt RC, Johnson GA, Jarger LA, Ka H, GarlowJE, Spencer TE, Bazer FW. Integrins and ExtracellularMatrix Proteins at the Maternal-Fetal Interface inDomestic Animals. Cell Tissues Organs, 2002, 172:202-217

[13] Tan Y, Gu ML, Wang ZB. Collection of uterineendometrium of mouse during peri-implantationperiod. Chinese J of Lab Animal Sci, 2001, 9: 40-44

[14] Zhu LJ, Bagchi MK, Bagchi IC. Attenuation ofcalcitonin gene expression in pregnant rat uterus leadsto a block in enbryonic implantation. Endocrinology,1998, 139(1): 330-339

[15] Liu YX. Plasminogen activator/plasminogen activatorinhibitors in ovarian physiology. Front Biosci, 2440, 9:3356-3373

[16] Coppolino MG, Dedhar S. Ligand-specific, transientinteraction between integrins and calreticulin duringcell adhesion to extracellular matrix proteins isdependent upon phosphorylation/dephosphorylationevents, Biochem J, 1999, 340: 41-50

[17] Reilly D, Larkin D, Devocelle M, Fitzgerald D J,Moran N. Calreticulin-independent regulation of theplatelet integrin alphaIIbbeta3 by the KVGFFKRalphaIIb-cytoplasmic motif, Platelets, 2004, 15: 43-54

[18] Wang J, Armant DR. Integrin-mediated adhesion andsignaling during blastocyst implantation. Cells TissuesOrgans, 2002, 172: 190-201

[19] Lessey BA, Arnold JT. Paracrine signaling in theendometrium: integrins and the establishment of uterinereceptivity. J Reprod Immuno, 1998, 39: 105-116

[20] Wang X, Matsumoto H, Zhao X, Das SK, Paria BC(2004) Embryonic signals direct the formation of tightjunctional permeability barrier in the decidualizingstroma during embryo implantation. J Cell Sci, 1998,117: 53-62

[21] Lessey BA, Damjanovich L, Coutifaris C, CastelbaumA, Albelda SM, Buck CA. Integrin adhesion moleculesin the human endometrium. Correlation with thenormal and abnormal menstrual cycle. J Clin Inves,1992, 90: 188-195

[22] Lessey BA, Castelbaum AJ, Sawin SW, Buck CA,Schinnar R, Bilker W, Strom BL. Aberrant integrinexpression in the endometrium of women withendometriosis. J Clin Endocrinol Metab, 1994, 79: 643-649

[23] Leung-Hagesteijn CY, Milankov K, Michalak M,Wilkins J, Dedhar S. Cell attachment to extracellularmatrix substrates is inhibited upon downregulation ofexpression of calreticulin, an intracellular integrinalpha-subunit-binding protein. J Cell Sci, 1994, 107:589-600

[24] Fadel MP, Dziak E, Lo CM, Ferrier J, Mesaeli N,Michalak M, Opas M. Calreticulin affects focalcontact-dependent but not close contact-dependent cell-substratum adhesion. J Biol Chem, 1994, 274: 15085-15094

[25] Fadel MP, Szewczenko-Pawlikowski M, Leclerc P,Dziak E, Symonds JM, Blaschuk O, Michalak M, OpasM. Calreticulin affects beta-catenin-associatedpathways. J Biol Chem, 2001, 276: 27083-9

[26] Jin-Xiang Yuan, Li-Juan Xiao, Cui-Ling Lu, Xue-SenZhang, Tao Liu, Min Chen, Zhao-Yuan Hu, Fei Gao,Yi-Xun Liu.Increased expression of heat shock protein105 in rat uterus of early pregnancy and i tssignificance in embryo implantation. ReproductiveBiology and Endocrinology; 2009, 7: 23

[27] Wagner RW. Gene inhibit ion using antisenseoligodeoxynucleotide. Nature, 1994, 372(6504): 333-335