Glycodelin A is expressed differentially in normal human endometrial tissue throughout the menstrual cycle as assessed by immunohistochemistry and in situ hybridization

Gah4m(acd

eU

RAR

Glycodelin A is expressed differentially in normalhuman endometrial tissue throughout the menstrualcycle as assessed by immunohistochemistry andin situ hybridizationIoannis Mylonas, M.D.,a Udo Jeschke, Ph.D.,a Christiane Kunert-Keil, Ph.D.,b

Naim Shabani, M.D.,a Darius Dian, M.D.,a Ingo Bauerfeind, M.D.,c Christina Kuhn,a

Markus S. Kupka, M.D., Ph.D.,a and Klaus Friese, M.D., Ph.D.a,c

a First Department of Obstetrics and Gynecology and c Department of Obstetrics and Gynecology, Ludwig MaximiliansUniversity of Munich, Munich; and b Institute of Pathophysiology, Ernst Moritz Arndt University of Greifswald, Karlsburg,Germany

Objective: To [1] evaluate glycodelin A immunolabeling in normal endometrium with specific monoclonal (mAb)and polyclonal peptide (pAb) antibodies, [2] to assess glycodelin messenger RNA (mRNA) by in situ hybrid-ization, and [3] to conduct deglycosylation experiments to evaluate the recognized epitope of the mAb vs. pAb.Design: Retrospective immunohistochemical analysis.Setting: University institute and hospital in Germany.Patient(s): Normal human endometrial tissue from the proliferative (PP), early secretory, and late secretoryphases were obtained from patients undergoing surgery for benign diseases.Intervention(s): Generation of a pAb in rabbit, immunohistochemistry, and in situ hybridization.Main Outcome Measure(s): Semiquantitative and computerized analysis.Result(s): A statistically significant increase of the glycodelin A immunolabeling in the late secretory phasecompared with PP was demonstrated by using the mAb. Polyclonal-peptide antibody immunolabeling alsoshowed a rise between the PP and late secretory phases, but without statistical significance. In situ hybridizationdemonstrated a statistically significantly higher mRNA content during the early secretory phase compared withduring PP.Conclusion(s): Glycodelin was demonstrated in normal endometrium at the protein and mRNA levels. The mAbmay be more useful in assessing glycodelin expression in endometrium, because it probably can bind toglycodelin A–unique glycan structures, in contrast to the peptide pAb. This is of major interest because it mayreveal possible structural and functional relationships in different parts of this molecule and elucidate possiblefunctions of this glycoprotein in human endometrial tissue. (Fertil Steril� 2006;86:1488–97. ©2006 by AmericanSociety for Reproductive Medicine.)

Key Words: Glycodelin, endometrium, immunohistochemistry, monoclonal antibody, polyclonal-peptide anti-body, in situ hybridization

t(m(aetsomt

awdba

lycodelin, formerly named placental protein 14 (PP14), isn approximately 28-kDa glycoprotein with a unique carbo-ydrate configuration, consistent with sialylated GalNAc1-GlcNAc (LacdiNAc) structures that are very unusual forammals (1). Glycodelin, isolated from amniotic fluid

glycodelin A), is composed of two identical subunits thatre connected closely by noncovalent bonds and have aarbohydrate content of 17.5%, with a unique carbohy-rate configuration (2).

Glycodelin is a major reproductive glycoprotein with sev-ral functions in cell recognition and differentiation (3).nder physiological conditions, glycodelin mainly is syn-

eceived July 18, 2005; revised and accepted March 31, 2006.uthors I.M., U.J., and C.K.-K. contributed equally to this work.eprint requests: Udo Jeschke, Ph.D., Ludwig Maximilians University ofMunich, First Department of Obstetrics and Gynecology, Maistrasse11, 80337 Munich, Germany (FAX: 0049-89-5160-4916; E-mail: udo.

[email protected]).

1488 Fertility and Sterility� Vol. 86, No. 5, November 2006Copyright ©2006 American Society for Reproductive Medicine,

hesized in secretory endometrial glandular epithelial cells4, 5), gestational decidua (6), the ovary (7), and inegakaryocytic–erythroid precursors of the bone marrow

8). There is substantial evidence that glycodelin A may bemediator for immunomodulatory and immunosuppressive

ffects on several human tissues. Glycodelin A suppresseshe release of interleukin-2 and interleukin-2 receptor fromtimulated lymphocytes (9–11). It also inhibits the activityf natural killer cells (12) and suppresses both the allogenicixed-lymphocyte reaction and lymphocyte responsiveness

o phytohemagglutinin (9, 10).

During pregnancy, the protein is secreted mainly into themniotic fluid, with increasing concentration until the 10theek of gestation (13). Glycodelin A is involved in theifferentiation of trophoblast cells (14), and a relationshipetween low serum levels of glycodelin A and threatenedbortion also has been suggested (15). In addition, reduced

0015-0282/06/$32.00Published by Elsevier Inc. doi:10.1016/j.fertnstert.2006.03.062

gl

gdtaiditc(iwfTroKidwumtsepm

emcpuca

MTSplbcu

bFwSac

sabac(oa

m1lhecrldtepsatp

GPOadttNtdbGw(wwtaituPti(pfg

F

rowth restriction and to HELLP (hemolysis, elevated-liver,ow platelets) syndrome during pregnancy (16).

Although differences have been observed in the majorlycodelin isoforms purified from secretory endometrium,ecidua, amniotic fluid, and serum from pregnant women,he dominant glycan types are sufficiently similar that theyll may be called glycodelin A (17). However, glycosylations recognized as an important factor in cell recognitionuring early reproductive events and in signaling pathwaysnvolved in carcinogenesis. Interestingly, the importance ofhe glycodelin glycosylation recently has been observed inomparative studies with human and hamster spermatozoa18). In addition, we recently demonstrated that glycodelin As an inhibitor of the E-selectin–mediated cell adhesion,hich is probably a result of its special glycosylation with

ucosylated N,N=-diacetyllactosediamine structures (19).hese results have prompted us to evaluate the staining

eactions of a native monoclonal glycodelin antibody andf a polyclonal glycodelin antibody generated in rabbits.nowledge of these staining reactions may be important

n assessing the structure and function that may exist inifferent parts of the glycodelin molecule. In addition, weanted to evaluate which of these antibodies might beseful in assessing glycodelin expression in human endo-etrial tissue. This may be very important especially for

he human endometrium, because glycodelin A is a majorecretory endometrial product during the luteal phase andarly pregnancy, and measurement of this protein mayrovide a simple marker of endometrial development andaturation (20).

Therefore, the aims of this study were the following: [1]valuation of the expression of glycodelin in normal endo-etrial tissue by immunohistochemistry with specific mono-

lonal and polyclonal antibodies, [2] evaluation of the ex-ression of glycodelin messenger RNA (mRNA) with these of in situ hybridization, and [3] evaluation of oligosac-haride binding of the monoclonal and polyclonal-peptidentibody by deglycosylation experiments.

ATERIALS AND METHODSissue Samplesamples of human endometrium were obtained from 54remenopausal, nonpregnant patients undergoing gyneco-ogical surgery either by dilation and curettage (n � 15) ory hysterectomy for benign diseases (n � 39, of which 29ases were for uterine leiomyomata and 10 cases were forterine prolapse).

All patients were included who underwent surgery forenign diseases between August 2003 and May 2004 at theirst Department of Obstetrics and Gynecology of the Lud-ig Maximilians University of Munich (Munich, Germany).trict criteria for inclusion in this analysis were chosen tovoid bias. All women had a normal and regular menstrual

ertility and Sterility�

urgery, as indicated by the chart notes of the physicians ands confirmed during the obtaining of informed consent. Noleeding at the time of operation and age of �45 years weredditional inclusion criteria. All pathological (one case ofarcinoma in situ III) and hyperplastic endometrial samples4 cases of nonsymptomatic endometrial polyp and 3 casesf simple hyperplasia) were excluded from this study tovoid additional bias.

Endometrial samples were classified according to ana-nestical and histological dating into proliferative (day

–14; n � 32), early secretory (day 15–22; n � 10), andate secretory (day 23–28; n �12) phases. This timeframeas been used previously in evaluating, besides glycodelinxpression in isolated endometrial glandular epithelialells (5), several other molecules as well, such as steroideceptors (21, 22), inhibin or activin subunits (23, 24), andeukemia inhibitory factor (25), with similar results, asemonstrated by other groups using different semiquanti-ative and computerized evaluation methods. Dating of thendometrial tissues was performed by a gynecologicalathologist (N.S.) according to histological criteria de-cribed elsewhere (26, 27). The local ethical committeepproved the obtaining and handling of human endome-rial material. Informed consent was obtained from allatients.

eneration of a Polyclonal Glycodelinolyclonal-Peptide Antibodyf the known 180 amino acids of the glycodelin protein (28),15–amino acid peptide was chosen as the antigen, as

escribed elsewhere (29). Anti-glycodelin polyclonal pep-ide antibodies were generated in rabbits against a polypep-ide of 15 amino acids of glycodelin (polypeptide sequence,H2-CKKVLGEKTENPKKFK-COOH; amino acid posi-

ion, 87 to 101; Genbank accession number, A39167) asescribed elsewhere (29). The antibody was raised in rabbitsy using standard procedures by BioScience (Göttingen,ermany), in contrast to the case in previous reports, inhich a polyclonal-peptide antibody was raised in chicken

29). A primary dose of 200-�g of glycodelin polypeptideas emulsified in Freund’s complete adjuvant (Sigma) andas administered subcutaneously in rabbits. Two doses of

he peptide emulsified in Freund’s incomplete adjuvant weredministered in intervals (6 wk). After the second boosternjection (at 14 d), blood was collected from the rabbit, andhe serum was separated. Antibodies were isolated by col-mn chromatography with a protein A column (Amershamharmacia Biotech, Freiburg, Germany). The specificity of

he antiserum was determined by using Western blot andmmunohistochemical analysis, as described elsewhere30). In Western blots, this antibody was reactive tourified glycodelin of human amnion fluid; to glycodelinrom pooled ascites of ovarian cancer patients; and tolycodelin isolated from blood plasma of pregnant women

1489

aw

IIddmaasrim3oat(wsHmtatiurPcld

PA(tNtR(dR

IoNadsmfti

ifH1smawiscw(iz

DPdwei4swfgais

IoTcqseutoIS2cc0a

witp

ccordance with previous reports regarding the moleculareight of glycodelin (31).

mmunohistochemistrymmunohistochemistry on paraffin sections (7 �m) of theifferent endometrial tissue specimens was performed asescribed elsewhere (30). Briefly, sections were incubated inethanol–H2O2 (30 min) to inhibit endogenous peroxidase

ctivity, washed in phosphate-buffered saline (PBS; 5 min),nd treated with goat serum (20 min, 22°C) to reduce non-pecific background staining. Incubation with the primaryabbit polyclonal-peptide antibody against glycodelin (1/500n PBS) was performed overnight at 4°C. Alternatively, aonoclonal mouse anti-glycodelin A antibody (2 �g/mL [5,

2]) served as primary antibody. Sections then were thor-ughly incubated with the biotinylated secondary anti-rabbitntibody or with the biotinylated secondary anti-mouse an-ibody (1 h, 22°C) and with avidin-biotinylated peroxidase45 min, room temperature). Between each step, the sectionsere washed with PBS (pH 7.4) three times. Peroxidase

taining reaction was performed with diaminobenzidine–

2O2(1 mg/mL; 5 min) and was stopped in tap water (10in). Sections were counterstained in hemalaun (1 min) and

hen covered. In controls, the primary polyclonal-peptidentibody was replaced with preimmune serum of the respec-ive rabbit. As positive controls for the monoclonal antibod-es, placenta slides of the second and third trimesters weresed. In the control placental tissue, only decidual cellseacted with the antibodies, as described elsewhere (16).ositive staining reaction was visualized with a brownisholor. Normal mouse serum was used instead of the glycode-in antibodies for negative controls. Negative-reacted cellsemonstrated a blue color.

reparation of Riboprobes227-bp fragment of the glycodelin complementary DNA

position �41 to �268) was cloned into the EcoR1 restric-ion site of pBluescript SK� (Stratagene, Amsterdam, Theetherlands) and labeled with digoxigenin (DIG) by in vitro

ranscription with the DIG RNA labeling Kit (SP6/T7;oche Biochemicals, Germany), as described elsewhere

33). The antisense chromosomal RNA was used for theetection of glycodelin mRNA, and the sense chromosomalNA probe served as a negative control.

n Situ Hybridization Analysis of Glycodelinn Paraffin Sectionsonradioactive in situ hybridization was performed on par-

ffin sections (4 �m) that had been fixed in 4% paraformal-ehyde. Sections were rehydrated and permeabilized by pep-in digestion (750 �g/mL of pepsin in 0.2 M HCl; 37°C, 30in). Postfixation (4% paraformaldehyde; 20 min, 4°C) was

ollowed by acetylation using 0.25% acetic anhydride inriethanolamine (0.1 M, pH 8.0, 15 min). After dehydration

1490 Mylonas et al. Glycodelin expression in human endome

zed for 16 hours (56°C) in a solution containing 25%ormamide; 0.3 M NaCl; 10% dextran sulfate; 20 mM Tris-Cl, pH 7.5; 1 mM ethylenediaminetetraacetic acid;� Denhardt’s (Ficoll, polyvinylpyrrolidone, and bovineerum albumin); 0.1 ng/mL of herring sperm DNA; 0.5g/mL of transfer RNA; 0.1 mg/mL of polyuridylic acid;

nd 125 ng of DIG-labeled chromosomal RNA probe. Afterashing with 50% formamide in 1� SSC, sections were

ncubated with RNAase A, followed by additional washingteps and incubation with blocking reagent (Roche Bio-hemicals). Bound riboprobe was visualized by incubationith alkaline phosphatase–conjugated anti-DIG antibody

Roche Biochemicals) and subsequent substrate reaction us-ng 5-bromo-4-chloro-3-indolyl phosphate–nitroblue tetra-olium chloride.

eglycosylation of Endometrial Tissue Slidesreliminary classification of antibody epitopes into carbohy-rate vs. peptide structures was obtained as described else-here (34), with some modifications (35). In brief, dewaxed

ndometrial tissue slides from the late secretory phase werencubated with 10 mM NaIO4 in 0.1 M acetate buffer, pH.5, for 1 hour, at room temperature, in the dark. Controllides were covered with acetate buffer only. Wells wereashed with PBS, treated with 50 mM NaBH4 in PBS

or 30 minutes, washed again, and incubated with anti-lycodelin monoclonal antibody or peptide glycodelinntibody in PBS for 90 minutes at room temperature. Themmunohistochemical staining was carried out as de-cribed in Immunohistochemistry.

mmunohistochemical Evaluation, Computerized Analysisf Glycodelin mRNA Expression, and Statistical Analysishe intensity and distribution of the specific immunohisto-hemical staining reaction was evaluated by using a semi-uantitative method (immunoreactive score [IRS]) as de-cribed elsewhere (36) that has been used in the evaluation ofndometrial steroid receptor (21, 22), inhibin or activin sub-nits (23, 24), and glycodelin expression in human endome-rium (5, 32). The slides were examined by two independentbservers, including a gynecological pathologist (N.S.). TheRS score was calculated as follows: IRS � SI � PP, whereI is the optical stain intensity (graded as 0 � no, 1 � weak,� moderate, and 3 � strong staining), and PP, the per-

entage of positive stained cells. The PP was estimated byounting out approximately 200 cells, and it was defined as� no staining, 1 � �10%, 2 � 11%–50%, 3 � 51%–80%,

nd 4 � �81%.

The level of glycodelin protein and mRNA expressionas determined in a blinded fashion in one run with

dentical staff, equipment, and chemicals. From each sec-ion, five digital pictures were taken at random of differentlaces of the tissue (200-fold magnification; 3CCD color

trium Vol. 86, No. 5, November 2006

atwapaKtu

tI

RIAep

F

nd Axiolab, Carl Zeiss, Germany). For standardization ofhe measurement in each picture, the optical density ofhite background color was attuned to 250 greyscale. For

ll sections, the mean optical density and the quantity ofixels that had a positive reaction for glycodelin weressessed by using the KSRun software (imaging systemS400, release 3.0; Zeiss, Vision GmbH, Germany). The

wo-tailed P test of the Mann-Whitney rank-sum test wassed to compare the means of the different IRS scores and

FIGURE 1

Immunohistochemical expression of glycodelin. Wherexpression during the proliferative phase (A), the polystaining reaction (B). During the early secretory phaswas confined to glandular epithelial cells (C), and a mpolyclonal-peptide antibody (D). The most intense rethe monoclonal antibody (E), whereas the polyclonal-intensity during this menstrual phase (F). Magnificatio

Mylonas. Glycodelin expression in human endometrium. Fertil Steril 2006.

ertility and Sterility�

he results of the computerized analysis (SPSS, Chicago,L). Significance was assumed at P�.05.

ESULTSmmunohistochemistrylthough the monoclonal antibody showed no or minimal

xpression during the proliferative phase, the polyclonal-eptide antibody produced a clear staining reaction, albeit a

the monoclonal antibody showed no or minimalal-peptide antibody produced a clear, albeit weak,e staining reaction with the monoclonal antibodyintense staining reaction was observed for then was observed during the late secretory phase fortide antibody reacted also with the strongestll panels, �250.

easclon

e, thore

actiopepn, a

1491

wtcrbdbws

idptoaui

eak one (Fig. 1). Whereas during the early secretory phasehe staining reaction with the monoclonal antibody wasonfined to glandular epithelial cells, a less intense stainingeaction was observed for the polyclonal-peptide anti-ody. However, the most intense reaction was observeduring the late secretory phase for the monoclonal anti-ody, whereas the polyclonal-peptide antibody reactedith lesser intensity and with minimal diffuse positive

tromal reaction (Fig. 1).

FIGURE 2

Evaluation of immunohistochemical glycodelin staininbetween the proliferative and late secretory phases (whereas the polyclonal glycodelin peptide antibody s(B) The staining intensity demonstrated a significant(*P�.05) as well as between the proliferative and late� SEM. Significance was assumed at P�.05 (all aste

Mylonas. Glycodelin expression in human endometrium. Fertil Steril 2006.

1492 Mylonas et al. Glycodelin expression in human endome

By assessing the IRS, which takes in account the stainingntensity and the amount of positive reacted cells for glan-ular epithelial cells, a significant rise (P�.002) between theroliferative phase and late secretory phase was observed forhe glycodelin monoclonal antibody. A similar rise also wasbserved for the polyclonal glycodelin polyclonal-peptidentibody but was not statistically significant (Fig. 2). Eval-ating just the intensity, a significant rise of the stainingntensity was observed between the proliferative and early

action. (A) The IRS showed a significant rise.002) for the glycodelin monoclonal antibody,ed a similar rise but without statistical significance.between the proliferative and early secretoryretory phases (**P�.05). All data are meansed values).

g re*P�howrise

secrisk

trium Vol. 86, No. 5, November 2006

stps

IEpmtg

aeeptemid

F

ecretory (P�.05) as well as the proliferative and late secre-ory (P�.05) phases. A similar rise also was observed for theolyclonal glycodelin peptide antibody but again was nottatistically significant (Fig. 2).

n Situ Hybridizationxpression of glycodelin mRNA was detected in the cyto-lasm of epithelial cells in endometrial tissue samples withinimal staining reaction in stromal cells, especially during

he late secretory phase. By using the monoclonal anti-

FIGURE 3

Glycodelin mRNA was expressed minimally in endomB, sense). Glycodelin mRNA was expressed minimall(C, antisense; D, sense). Glycodelin mRNA decreaseF, sense). Magnification, all panels, �100.

Mylonas. Glycodelin expression in human endometrium. Fertil Steril 2006.

lycodelin and the polyclonal glycodelin polyclonal-peptide s

ertility and Sterility�

ntibody, specific staining reaction was observed in humanndometrial glandular epithelial cells. There was minimalxpression of glycodelin mRNA during the proliferativehase. The highest intensity of mRNA was observed duringhe early secretory phase (Fig. 3). Interestingly, the mRNAxpression was reduced during the late secretory phase, withinimal expression in stromal cells. All sections that were

ncubated with the sense chromosomal RNA probe wereevoid of any reaction.

The densitometric quantification of the glycodelin mRNA

al tissue from the proliferative phase (A, antisense;endometrial tissue from the proliferative phasering the late secretory phase (E, antisense;

etriy ind du

taining intensity showed the strongest mRNA expression in

1493

tt

DPdww(tdipsbpat

DTpeawcsh

magtebshdl

gaidto

tatctafii

he early secretory phase; it was significantly higher (P�.05)han in the proliferative phase (Fig. 4).

eglycosylationreliminary classification of antibody epitopes into carbohy-rate vs. peptide structures was obtained as described else-here (34, 35). Both antibodies reacted positively primarilyith glandular epithelial cells from the late secretory phase

Fig. 5). After the deglycosylation procedure, the staining reac-ion of the monoclonal antibody was minimal within the glan-ular epithelial cells, showing a substantial decrease in thentensity. However, the immunoreaction of the polyclonal-eptide antibody also was diffuse within the endometrialtroma. The staining reaction of the polyclonal-peptide anti-ody was not significantly decreased after the deglycosylationrocess, and the staining reaction with the polyclonal-peptidentibody was more intense after deglycosylation compared withhe case of the monoclonal antibody (Fig. 5).

ISCUSSIONhe endometrium is one principal target tissue of theituitary–gonadal axis but also has been recognized as anndocrine organ itself. There is a clinical need for a simplend efficient marker of the activity of this tissue, especiallyith regard to the endometrial physiology, pathogenesis, and

arcinogenesis (3, 20). We demonstrated glycodelin expres-ion in normal human endometrial tissue by using immuno-

FIGURE 4

Evaluation of glycodelin expression by in situ hybridizhybridization reaction in endometrial tissue as determrise of mRNA expression during the early secretory p(*P�.05). Data are means � SEM. Significance was a

Mylonas. Glycodelin expression in human endometrium. Fertil Steril 2006.

istochemistry and in situ hybridization. r

1494 Mylonas et al. Glycodelin expression in human endome

A significant rise of the glycodelin staining reaction foronoclonal antibody was observed between the proliferative

nd early and late secretory phases. With the polyclonallycodelin peptide, a similar rise could be observed, al-hough without statistical significance. In situ hybridizationxperiments confirmed on mRNA levels the results obtainedy immunohistochemistry. Glycodelin mRNA was demon-trated throughout the menstrual cycle and was significantlyigher during the early secretory phase, which is in accor-ance with previous results, because protein secretion fol-ows the mRNA synthesis.

These conflicting results between glycodelin peptide andlycodelin clearly demonstrate that anti–glycodelin peptidentibodies (29) are different from native glycodelin antibod-es (5, 31, 32, 37, 38). These assumptions were confirmed byeglycosylation experiments, in which classification of an-ibody epitopes into carbohydrate vs. peptide structures wasbtained.

Although both antibodies reacted positively with endome-rial glandular epithelial cells from the late secretory phase,fter the deglycosylation procedure, the staining reaction ofhe monoclonal antibody was minimal within the gland cells,ompared with the polyclonal-peptide antibody. In addition,he polyclonal-peptide antibody showed diffuse staining re-ction within endometrial stroma, which could not be con-rmed with the monoclonal antibody nor the in situ hybrid-

zation. Therefore, we assume that although both antibodies

n. Staining intensity of glycodelin mRNA by in situby computerized analysis showed a significant

e compared with during the proliferative phasemed at P�.05 (asterisked values).

atioinedhasssu

ecognize glycodelin within human endometrium, the mono-

trium Vol. 86, No. 5, November 2006

cwi

senvgtr(te

rri

digtntRsttc

twtsp

F

lonal antibody reacts primarily with glycan structures,hereas the main epitope of the polyclonal-peptide antibody

s the peptide sequence.

Actually, the differences in the staining reaction are noturprising in view of the fact that linear and conformationalpitopes are known to differ, because the antibody’s recog-ition of the antigen is specific and is affected by antigenicariation (39). Recently, an alternative splice variant oflycodelin mRNA was demonstrated in human breast cancerissue (37). Should this mRNA variant be translated, theesulting protein will lack three potential glycosylation sides37), which eventually would lead to a different folding ofhe protein. However, whether this also is true in humanndometrium still remains unknown.

These observations between the different epitopes that areecognized with both antibodies are of importance withegard to the glycodelin function. Meanwhile, glycosylation

FIGURE 5

Binding of glycodelin monoclonal and polyclonal pepBoth antibodies reacted positively primarily with endophase (A and C). After the deglycosylation procedureminimal within the glandular epithelial cells, showingimmunoreaction of the polyclonal-peptide antibody wreaction within the endometrial stroma (C). We foundthe deglycosylation process (D). Magnification, all pa

Mylonas. Glycodelin expression in human endometrium. Fertil Steril 2006.

s recognized as an important factor in cell recognition c

ertility and Sterility�

uring early reproductive events and in signaling pathwaysnvolved in carcinogenesis. Although the precise role oflycodelin is quite unknown, it is thought to participate inhe fetomaternal defense mechanisms because of its immu-osuppressive properties. This effect may be dependent onhe blocking of E-selectin–mediated cell adhesion (1).ecently, it was demonstrated that both glycodelin A and

erum glycodelin are in vitro very efficient inhibitors ofhe E-selectin–mediated cell adhesion, suggesting an impor-ant role in carcinogenesis and metastatic potential of can-erous cells (19).

Interestingly, the importance of the glycodelin glycosyla-ion recently has been demonstrated in comparative studiesith human and hamster spermatozoa (18). An overlap be-

ween the immune and gamete recognition systems has beenuggested specifically in terms of the glycans that are ex-ressed on CD45 in T cells and the glycans that are impli-

antibodies to deglycosylated endometrial tissue.trial glandular epithelial cells from the late secretorye staining reaction of the monoclonal antibody wasbstantial decrease in the intensity (B). Theocalized in epithelial cells, with a diffuse stainingsignificant decrease of the staining reaction after, �250.

tideme, tha suas lno

nels

ated in mediating gamete recognition (40). In addition, the

1495

sippmo

tkitatd

ncooawwg(dtwdtstisolapcgwc

pssatsnTao

ABs

oPaPs

R

1

1

1

1

1

1

1

1

1

ame unusual carbohydrate sequences associated with thesemmunosuppressive glycodelins also are specifically ex-ressed on intravascular helminthic parasites, Helicobacterylori, and HIV-infected T lymphocytes (41), suggesting aore complex function of this glycoprotein in pathogenesis

f other diseases.

Differences in glycosylation may explain different func-ions of glycodelin within several tissues. Therefore, thenowledge of the used glycodelin antibodies may be verymportant in elucidating the function of this important pro-ein. In addition, information about the glycodelin glycosyl-tion form is of extreme importance because it may revealhe structural and functional relationships that may exist inifferent parts of the glycodelin molecule.

One of the difficulties in this study was that of obtainingormal human endometrial tissue from healthy patients. Be-ause it is quite unclear whether endometrial benign leiomy-ma can influence glycodelin production, careful acquisitionf tissue samples outside the area of the myomas was envis-ged. Recently, a significant rise of plasma glycodelin levelsas demonstrated in patients with leiomyomata comparedith in normal controls, whereas no significant changes inlycodelin concentrations in uterine flushings were observed42). These results suggest that the higher glycodelin pro-uction noticed in human serum may be from another sourcehan endometrial tissue (42). In addition, infertility patientsere analyzed, which also may suggest an endocrinologicalysregulation affecting glycodelin expression. Furthermore,here are no data to suggest that muscle cells can produce andecrete glycodelin (Mylonas I et al., unpublished observa-ions). Finally, it may also be possible that leiomyomatancrease the endometrial surface, resulting in a higher totallyecreted glycodelin into human plasma but not in an increasef locally endometrial glycodelin production within glandu-ar cells. Therefore, we suggest that myomas do not have

directly major effect in endometrial cellular glycodelinroduction. Interestingly, these assumptions appear to beonfirmed by the statistically nonsignificant changes oflycodelin concentrations in uterine flushings in patientsho have leiomyoma, compared with individuals in the

ontrol group (42).

In conclusion, by using a monoclonal and polyclonal-eptide antibody raised against glycodelin, we observedome discrepancies regarding the immunohistochemical re-ults on the same specimens. We assume that these discrep-ncies were a result of the different glycosylated epitopeshat are recognized by the different antibodies. Additionaltudies are required to elucidate the relationship of endoge-ous glycodelin peptide immunoreactivity with glycodelin.his is of major interest because it may reveal the structuralnd functional relationships that may exist in different partsf the glycodelin molecule.

cknowledgments: The authors thank Dr. Uwe Karsten, Ph.D. (Glycotope,erlin, Germany), for providing the monoclonal antibodies used in this

tudy. We thank Susi Kunze (1st Department of Obstetrics and Gynaecol-

1496 Mylonas et al. Glycodelin expression in human endome

gy, LMU Munich, Munich, Germany) and Jutta Adolph (Institute ofathology EMAU Greifswald, Greifswald, Germany) for excellent technicalssistance. Furthermore, the authors are grateful to Dr. Reiner Jarchow,h.D. (Computing Centre, EMAU Geifswald), for his support and to Deut-che Forschungsgemeinschaft (DFG) for financial support.

EFERENCES1. Dell A, Morris HR, Easton RL, Panico M, Patankar M, Oehniger S,

et al. Structural analysis of the oligosaccharides derived from glycode-lin, a human glycoprotein with potent immunosuppressive and contra-ceptive activities. J Biol Chem 1995;270:24116–26.

2. Bohn H, Kraus W, Winckler W. New soluble placental tissueproteins: their isolation, characterization, localization and quantifi-cation. Placenta Suppl 1982;4:67– 81.

3. Seppala M, Taylor RN, Koistinen H, Koistinen R, Milgrom E. Gly-codelin: a major lipocalin protein of the reproductive axis with diverseactions in cell recognition and differentiation. Endocr Rev 2002;23:401–30.

4. Julkunen M. Human decidua synthesizes placental protein 14 (PP14) invitro. Acta Endocrinol (Copenh) 1986;112:271–7.

5. Mylonas I, Speer R, Makovitzky J, Richter DU, Briese V, Jeschke U,et al. Immunohistochemical analysis of steroid receptors and glycodelinA (PP14) in isolated glandular epithelial cells of normal human endo-metrium. Histochem Cell Biol 2000;114:405–11.

6. Julkunen M, Koistinen R, Sjoberg J, Rutanen EM, Wahlstrom T,Seppala M. Secretory endometrium synthesizes placental protein 14.Endocrinology 1986;118:1782–6.

7. Kamarainen M, Leivo I, Koistinen R, Julkunen M, Karvonen U, Ru-tanen EM, et al. Normal human ovary and ovarian tumors expressglycodelin, a glycoprotein with immunosuppressive and contraceptiveproperties. Am J Pathol 1996;148:1435–43.

8. Kamarainen M, Riittinen L, Seppala M, Palotie A, Andersson LC.Progesterone-associated endometrial protein—a constitutive marker ofhuman erythroid precursors. Blood 1994;84:467–73.

9. Pockley AG, Mowles EA, Stoker RJ, Westwood OM, Chapman MG,Bolton AE. Suppression of in vitro lymphocyte reactivity to phytohe-magglutinin by placental protein 14. J Reprod Immunol 1988;13:31–9.

0. Pockley AG, Bolton AE. Placental protein 14 (Glycodelin A) inhibitsthe synthesis of interleukin-2 and the release of soluble interleukin-2receptors from phytohaemagglutinin-stimulated lymphocytes. Clin ExpImmunol 1989:252–6.

1. Pockley AG, Bolton AE. The effect of human placental protein 14(Glycodelin A) on the production of interleukin-1 mitogenically stim-ulated mononuclear cell cultures. Immunology 1990:277–81.

2. Okamoto N, Uchida A, Takakura K, Kariya Y, Kanzaki H, Riittinen L,et al. Suppression by human placental protein 14 of natural killer cellactivity. Am J Reprod Immunol 1991;26:137–42.

3. Julkunen M, Rutanen EM, Koskimies A, Ranta T, Bohn H, Seppala M.Distribution of placental protein 14 in tissues and body fluids duringpregnancy. Br J Obstet Gynaecol 1985;92:1145–51.

4. Jeschke U, Richter DU, Reimer T, Bergemann T, Briese V, Karsten U,et al. Glycodelin A and differentiation of first trimester trophoblast cellsin vitro. Arch Gynecol Obstet 2005;272:151–9.

5. Tomczak S, Briese V, Kunkel S, Muller H. Serum placental protein 14(PP14) levels in patients with threatened abortion. Arch Gynecol Obstet1996;258:165–9.

6. Jeschke U, Kunert-Keil C, Mylonas I, Hammer A, Schiessl B, LombaI, et al. Expression of glycodelin A in decidual tissue of preeclamptic,HELLP and intrauterine growth-restricted pregnancies. Virchows Arch2005;446:360–8.

7. Koistinen H, Easton RL, Chiu PC, Chalabi S, Halttunen M, Dell A,et al. Differences in glycosylation and sperm-egg binding inhibition ofpregnancy-related glycodelin. Biol Reprod 2003;69:1545–51.

8. Dutta B, Ain R, Seshagiri PB, Karande AA. Differential influence ofrecombinant non-glycosylated and glycosylated glycodelin on humansperm function: comparative studies with hamster spermatozoa. Reprod

trium Vol. 86, No. 5, November 2006

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

F

9. Jeschke U, Wang X, Briese V, Friese K, Stahn R. Glycodelin and amniofluid transferrin as inhibitors of E-selectin mediated cell adhesion.Histochem Cell Biol 2003;119:345–54.

0. Seppala M, Suikkari AM, Julkunen M. Human endometrial proteins.Reprod Nutr Dev 1988;28:1649–54.

1. Mylonas I, Jeschke U, Shabani N, Kuhn C, Balle A, Kriegel S, et al.Immunohistochemical analysis of estrogen receptor alpha, estrogenreceptor beta and progesterone receptor in normal human endometrium.Acta Histochem 2004;106:245–52.

2. Mylonas I, Jeschke U, Shabani N, Kuhn C, Kriegel S, Kupka MS, et al.Normal and malignant human endometrium express immunohisto-chemically estrogen receptor alpha (ER-alpha), estrogen receptor beta(ER-beta) and progesterone receptor (PR). Anticancer Res 2005;25:1679–86.

3. Mylonas I, Jeschke U, Wiest I, Hoeing A, Vogl J, Shabani N, et al.Inhibin/activin subunits alpha, beta-A and beta-B are differentiallyexpressed in normal human endometrium throughout the menstrualcycle. Histochem Cell Biol 2004;122:461–71.

4. Mylonas I, Makovitzky J, Richter DU, Jeschke U, Briese V, Friese K.Expression of the inhibin-alpha subunit in normal, hyperplastic andmalignant endometrial tissue: an immunohistochemical analysis. Gy-necol Oncol 2004;93:92–7.

5. Mylonas I, Makovitzky J, Shabani N, Richter DU, Kuhn C, Jeschke U,et al. Leukaemia inhibitory factor (LIF) is immunohistochemicallyexpressed in normal, hyperplastic and malignant endometrial tissue.Eur J Obstet Gynecol Reprod Biol 2005;118:101–8.

6. Dallenbach-Hellweg G, Poulsen H. Atlas der Histopathologie des En-dometriums. Berlin: Springer, 1985.

7. Noyes RW, Hertig AT, Rock J. Dating the endometrial biopsy. FertilSteril 1950;1:3–15.

8. Huhtala ML, Seppala M, Narvanen A, Palomaki P, Julkunen M, BohnH. Amino acid sequence homology between human placental protein 14and beta-lactoglobulins from various species. Endocrinology 1987;120:2620–2.

9. Poddar AS, Kim JG, Gil KP, Bates BN, Santanam N, Rock JA, et al.Generation and characterization of a polyclonal antipeptide antibody tohuman glycodelin. Fertil Steril 1998;69:543–8.

0. Bergemann C, Reimer T, Muller H, Hosel A, Briese V, Friese K, et al.Stimulation of hCG protein and mRNA levels in trophoblast tumourcells Jeg3 and BeWo by glycodelin A. Anticancer Res 2003;23:

ertility and Sterility�

1. Riittinen L, Narvanen O, Virtanen I, Seppala M. Monoclonal antibodiesagainst endometrial protein PP14 and their use for purification andradioimmunoassay of PP14. J Immunol Methods 1991;136:85–90.

2. Mylonas I, Jeschke U, Makovitzky J, Winkler L, Richter DU, Friese K,et al. Immunohistochemical expression of steroid receptors and gly-codelin A in isolated proliferative human endometrial glandular cellsafter stimulation with tamoxifen and phytoestrogens (genistein anddaidzein). Anticancer Res 2003;23:1119–25.

3. Keil C, Husen B, Giebel J, Rune G, Walther R. Glycodelin mRNA isexpressed in the genital tract of male and female rats (Rattus norvegi-cus). J Mol Endocrinol 1999;23:57–66.

4. Woodward MP, Young WW, Bloodgood RA. Detection of monoclonalantibodies specific for carbohydrate epitopes using periodate oxidation.J Immunol Methods 1985;78:143–53.

5. Karsten U, Butschak G, Cao Y, Goletz S, Hanisch FG. A new mono-clonal antibody (A78-G/A7) to the Thomsen-Friedenreich pan-tumorantigen. Hybridoma 1995;14:37–44.

6. Remmele W, Stegner HE. Vorschlag zur einheitlichen Definierungeines immunreaktiven Score (IRS) für den immunhistochemischenÖstrogenrezeptornachweis (ER-ICA) im Mammakarzinomgewebe.Pathologe 1987;8:138–40.

7. Kamarainen M, Halttunen M, Koistinen R, von Boguslawsky K, vonSmitten K, Andersson LC, et al. Expression of glycodelin in humanbreast and breast cancer. Int J Cancer 1999;83:738–42.

8. Arnold JT, Lessey BA, Seppala M, Kaufman DG. Effect of normalendometrial stroma on growth and differentiation in Ishikawa endome-trial adenocarcinoma cells. Cancer Res 2002;62:79–88.

9. Mariuzza RA, Phillips SE, Poljak RJ. The structural basis ofantigen-antibody recognition. Annu Rev Biophys Biophys Chem1987;16:139 –59.

0. Clark GF, Dell A, Morris HR, Patankar MS, Easton RL. The speciesrecognition system: a new corollary for the human fetoembryonicdefense system hypothesis. Cells Tissues Organs 2001;168:113–21.

1. Clark GF, Dell A, Morris HR, Patankar M, Oehninger S, Seppala M.Viewing AIDS from a glycobiological perspective: potential linkages tothe human fetoembryonic defence system hypothesis. Mol Hum Reprod1997;3:5–13.

2. Richlin SS, Ramachandran S, Shanti A, Murphy AA, Parthasarathy S.Glycodelin levels in uterine flushings and in plasma of patients withleiomyomas and polyps: implications for implantation. Hum Reprod

1497