1 Supported by NIH grant HD.14966 (FT).

2 Correspondence to: Dr. RH. Renegar. Department of Anatomy and

Cell Biology, School ofMedicine, East Carolina University, Greenville, NC

27858.

935

0022-1554/90/$3.30

The Journal of Histochemistry and CytochemistryCopyright © 1990 by The Histochemical Society, Inc.

Vol. 38, No. 7, pp. 935-940, 1990

Printed in USA.

Original Article

Immunohistochemical Co-localization of Placental

Lactogen II and Relaxin in the Golden Hamster

(Mesocricetus a ura tus)’

RANDALL H. RENEGAR,2 JONATHAN N. SOUTHARD, and FRANK TALAMANTES

Department of Anatomy and Cell Biology, School of Medicine, East Carolina University, Greenville, North Carolina (RHR,);

and Department ofBiology, University of California, Santa Cruz, California (JNS,FT).

Received for publicationjuly 31, 1989 and in revised formJanuary 31, 1990; accepted February 2, 1990 (9A1749).

Two hormones with lactogenic activity are produced by thehamster placenta during the second halfofpregnancy. One

of these hormones, hamster placental lactogen II (haPL-II),has been well characterized; however, its cellular source is

not known. In the present study, haPLrII was localized inplacental tissues using a specific antibody and the avidin-bio-tin-peroxidase immunohistochemical technique. Becauserelaxin has been localized in the hamster placenta, it was

of interest to determine if haPL-II and relaxin are localizedin the same cells. haPL-II immunoactivity was observed in

primary and secondary giant trophoblast cells ofthe placentaon Days 12, 14, and 15 of pregnancy. On Day 15 positive

Introduction

Lactogenic activity in placental tissue extracts and blood from preg-

nant hamsters was first reported by Kelly et al. (1976). Serum (Kelly

et al. , 1976) and tissue (Soares and Talamantes, 1982) lactogenic

activity in the hamster is detectable beginning at midpregnancy

(Day 8). Serum concentrations increase to term (Day 16) and placen-

tal tissue concentrations increase to Day 14 but decrease thereafter

(Soares and Talamantes, 1982). Two proteins designated hamster

placental lactogen I (haPL-I, MW 37,000) and haPL-II (MW 22,000)

account for most of the lactogenic activity in serum and placental

extracts during the latter halfofpregnancy in this species (Southard

et al., 1987; Southard et al., 1986). The haPL are so designated

because of their similarity in structure to PL found in the mouse

(Jackson et al. , 1986; Soares et al. , 1983) and rat (Duckworth et

al., 1982; Robertson et al., 1982). Placental lactogen II is the most

extensively characterized ofthe hamster hormones (Southard et al.,

1986), and recently a specific antibody and a homologous radio-

immunoassay were developed (Southard and Talamantes, 1987).

Hamster placental lactogen II is first detected in serum on Day 10

staining was also observed in large cells located withinmesometrial arteries and in eosinophilic bodies associatedwith degenerating sheathed arteries of the decidua basalis.haPL-II-positive staining was not observed in placentae fromDays 8 or 10 of pregnancy. On Day 14, haPL-II was co-localized with relaxin in 75#{176}/oof the giant trophoblast cells

observed. Therefore, it is probable that these hormones are

synthesized and secreted by the same cell. (J HiswchemCyrochem 38:935-940, 1990)

KEY WORDS: Placenta; Placental lactogen; Hamster; Immuno-

histochemistry; Relaxin.

ofgestation, and concentrations increase to term (Southard et al.,

1987).

The objective ofthe present study was to immunolocalize haPL.II

in the hamster placenta throughout the latter half of gestation.

Because the placenta is the source ofrelaxin in the pregnant ham-

ster (Steinetz et al. , 1988; Renegar et al. , 1987), we also examined

whether haPL-II and relaxin are localized within the same placen-

tal cells.

Materials and Methods

Animals and Tissue Preparation. Pnimiparous female Golden (Syrian)

hamsters (120-150 g; Charles River, Kingston, NY) were maintained in the

animal housing facility of the Department of Comparative Medicine on

a 12L:12D (lights on at 0700 hn) schedule. Animals were checked daily (be-

tween 2000 and 2200 hr) for estrus, as determined by expression of lordosis

in the presence of a male. Females in estrus were housed overnight with

a male and the following day was designated Day 1 of gestation. Animal

maintenance and handling were performed in accordance with the institu-

tion’s guidelines for care and use oflaboratory animals. Animals were anesthe-

tized with a combination ofxylazine (Haven-Lockhart; Shawnee, KS) andketamine HCI (Bristol Laboratories; Syracuse, NY) and the reproductive

tract exposed through a midventral incision. Placentae were recovered onDays 8, 10, 12, 14, and 15 ofgestation (n = 3 or 4 animals/day) between

0800 and 1200 hr, except on Day 15 when tissues were removed between

1900 and 2100 hr to obtain tissues immediately before partunition (Day

by guest on March 16, 2016jhc.sagepub.comDownloaded from

936 RENEGAR, SOUTHARD, TALAMANTES

Hamster placental lactogen Ii immunoreactivity was not observed

Figure 1. Cross-section ofan implantation site on Day 12 ofgestation. The chonioallantoic placenta is well developed. The trophospongium (ts), which is composedof fetal trophoblast cells, including the secondary giant trophoblasts, has invaded the decidua basalis (db). The placental labyrinth (Ib), where nutrient exchange

between mother and fetus occurs, develops by extension of fetal vessels into the trophospongium. Primary giant trophoblast cells are located within the reticularmatrix (.) at the peripheral margins of the discoid placenta. A central placental artery (arrowheads) and peripheral venous channels course through the deciduaand mesometnium (mm). m, muscularis; ys, yolk sac; f, fetus. (Inset) High magnification of the decidual portion of a central artery. These arteries are characterizedby their sheath of connective tissue cells and minimal smooth muscle. A sheath forms around mesometnial arteries as pregnancy progresses. Eosin and hematoxy-in stained. Bar � 200 am; inset � 50 �tm.

16 = day ofpantunition). Tissues were fixed by immersion in Bouin’s fluid,

dehydrated, and embedded in paraffin. Tissues from Days 10 and 14 were

also fixed in Zamboni’s, Bouin-Hollande’s, or Carnoy’s solution to exam-inc the influence of fixation on immunostaining for haPL-II. Seven-�tm

sections were placed on glass slides for immunohistochemistry or for he-

matoxylin and eosin staining.

lmmunohistochemistry. Hamster placental lactogen II and relaxin were

localized in placental tissue sections using the avidin-biotin-peroxidase

technique (Vectastain ABC kit; Vector Laboratories, Burlingame, CA) as

previously described (Renegar et al., 1987). The antiserum to haPL-Il has

been previously used to develop a homologous haPL.II radioimmunoassay

and does not crossreact significantly with hamster prolactin or growth hon.

mone (Southard and Talamantes, 1987). The rabbit antiserum to porcine

relaxin (UF-1M.; kindly provided by Dr. Mj. Fields. University of Florida)

has been previously used to localize relaxin in the hamster placenta (Rene-

garet al., 1987). Diluted antisera to haPL.Il (U10,000) and relaxin (12500)

were incubated on tissue sections for 24 hr at 4C. Bound avidin-biotin-per-

oxidase complex was localized by incubating sections in a solution of 0.5%

(w/v) 3,3’-diaminobenzidine, 0.02% hydrogen peroxide in 0.1 M Tnis (hy-

droxymethyl) amino methane (Tnis) buffer (pH 7.2). Completed slides were

counterstained with hematoxylin.

Method specificity for haPL.Il staining was assessed (a) by treating ad-

jacent sections with specific antiserum and non-immune rabbit serum and

(b) by omission of individual components of the staining procedure. Posi-

tive staining was present only when specific antiserum and all components

of the staining procedure were used. Specificity of the antiserum for its

antigen was assessed by (a) exhaustive dilution to demonstrate the pres-

ence ofnonspecific low-titer antibodies and (b) absorption with each anti-

gen. Absorption ofhaPL.II-specific antiserum overnight with purified haPIAI

(0.15 riM) (Southard et al., 1986) but not porcine relaxin (NIH.RXN-P1;

2.1 saM) eliminated all positive staining. Likewise, absorption of relaxin.

specific antisera with porcine relaxin (1.0 �sM) but not haPL-II (1.69 �tM)

eliminated all positive staining.

Estimate of Co-localization. To evaluate haPL-II and relaxin co-local-ization in the hamster placenta, identical areas of adjacent tissue sections

from Day 14 ofgestation were stained for haPL.II or relaxin, photographed,

and printed (total magnification x 254). Immunostained cells (haPL.II

or relaxin) with the nucleus present in both sections were counted. Cells

identified in photomicrographs were confirmed by direct microscopic ob-

servation of tissue slides. Three or four microscopic fields (189 cm2) from

each of four different placentae were examined.

Results

by guest on March 16, 2016jhc.sagepub.comDownloaded from

..� .

.�.... 4 ij�

� ..

� ‘� � . .,it1�Z�Vf. . � � ,

.- �

. .5

- � .-.-.- ‘..,� .� -v,. � � - �

, � V �.�

; �

haN1-II IMMUNOLOCALIZATION

s.s):;,.. , lb ,

�

;:� � . a .

,. -...,.; . -. ..

‘5 a _ ‘-

fi .::� �- � �

. - #{149}�;�.#{149} .,._

I #{149}4 W4.’..’, ‘.. �#{149}�.4f�. -

� ,;,,,� � #{149}

. ‘,,##{149} �,- ,.Op_. .4�

.‘. � . . F k ‘

.�. ._.4. p�.#{248}ay ‘

a � ‘5 ‘#{149}‘“� ..,.

a � . . .

937

Figure 2. (A) Immunocytochemical localization of haPL-lI in the placenta on Day 14 of gestation. Positive staining is confined to GTC-2 within the trophospongium.Ib, labyrinth. (B) Section adjacent to that in A stained for haPL-II using normal rabbit serum in place of antiserum to haPL-Il. (C) High magnification of a secondary

giant trophoblast cell stained to localize haPL-II. Note close proximity to maternal venous space (s) and perinuclear localization of staining. n, nucleus. (D) Primarygiant trophoblast cells within the peripheral reticulum that contain light staining (arrowheads) for haPL-Il. Bars = 50 �m.

in placental or uterine tissues on Days 8 or 10 ofgestation. On Days

12, 14, and 15 PM haPL-Il immunoreactivity was localized in sec-

ondary giant trophoblast cells (G’Lt-2) in the placental trophospon-

gium (Figures 1 and 2A) and in primary giant trophoblast cells

(GTC-1) in the peripheral reticulum surrounding the chonioallan-

toic placenta (Figures 1 and 2D). Secondary giant trophoblast cells

containing haPL-II usually bordered a maternal venous space and

specific staining ofsubstances in these venous spaces also was ob-

served (Figure 2A). On Days 14 and 15, haPL-II was also immu-

nolocalized in individual or groups of giant cells within the laby-

ninth. These may represent Glt-2 cells incorporated into the

labyrinth by formation of new capillaries at the junctional zone.

With the haPL-II antiserum dilution utilized, which maximized

specific staining, the cytoplasmic distribution of brown reaction

product was relatively homogeneous. However, increased dilution

ofthe antiserum or processing of4-�.tm sections revealed that most

cells contained focal sites of intense staining in the peninuclear ne-

gion (Figure 2C). Because of placental growth, the number of haPL-

Il-positive cells was greater on Days 14 and 15 PM than on Day 12.

In addition to staining for haPL-II within primary and second-

ary GTC, on Day 15 PM staining was also localized within eo-

sinophilic bodies between the trophospongium and compacted

decidua basalis (Figure 3A). This staining appeared to be associated

with degenerating cells that surround the sheathed arteries of the

decidua basalis (Orsini, 1954). Staining for haPL-II was occasion-

ally observed in large cells with hetenochromatic nuclei located

within the lumen of mesometnial sheathed arteries (Figure 3B).

Because haPL-II was localized in cell types previously demon-

strated to contain relaxin (Renegar et al. , 1987), it was of interest

to determine if these two hormones are present within the same

cell. Hamster placental lactogen II and relaxin were co-localized

in primary and secondary giant trophoblast cells of the Day 14

placenta. Four hundred and thirty-three immunostained GTC-2

cells were counted, and haPL-II and relaxin were co-localized in

75% of these cells (Figures 4A and 4B). Hamster placental lacto-

gen II or relaxin only was detected in 17.9 and 7.1% ofstained cells,

respectively.

Discussion

Immunolocalization of haPL-II (this study) and relaxin (this study;

and Renegar et al. , 1987) in giant trophoblast cells suggests that

GTC are the source of these placental hormones in the hamster.

Organization of the cytoplasmic organdIes of the GTC, abundant

profiles ofendoplasmic reticulum, and an extensive Golgi appara-

tus Uohns and Renegar, unpublished data) indicate that these cells

are actively involved in protein synthesis. Intense peninuclear staining

for haPL.lI in Git may represent hormone contained within the

extensive peninuclear Golgi complex of these cells, since secretory

granules are not present Uohns and Renegar, unpublished data).

by guest on March 16, 2016jhc.sagepub.comDownloaded from

C.

‘1 v�’�4�

�e:i., �

4

‘5

A

SI

� �

.� .,,

�

V,. � �-4. _

Figure 4. (A) Day 14 placental tissue stained to localize haPL-Il. (B) Section adjacent to that in A stained to localize relaxin. Note co-localization in cells (arrow-heads) adjacent to the maternal venous space (s). Bars = 50 �m.

938 RENEGAR, SOUTHARD, TALAMANTES

,� 101” # �: � -� #{149}

1r, � � , �.,;.., � “;

4 �. �ti � , �,‘t ;;��‘ ,

... .� t � ‘-.- : � � r� ‘-

��:�:; ,�, , � � �

k,� � b 1� � #{149}

.4% ,,#{149} � �; #{149}( S , #{149}, � � ‘

.., � ,jt -�. �, � iq-. ‘ .‘ ,.v�, ;-� �%- �

‘‘�. � f -.� .- � . p� � .�. . . #{149}: ,�,#{149} d.,’�5�t’ �

�W;4#{149} � �s . #{149}::Sd�” .a� �

Figure 3. Portion of a Day 15 placenta immunostained to localize haPL-ll. (A) Note staining within cells and maternal venous spaces of the trophospongium (ts).

db, decidua basalis. Staining is also associated with eosinophilic bodies in degenerating sheath cells of a decidual artery (a). (B) Portion of a mesometrial sheathedartery with staining located in large cells adjacent to the vessel lumen (.). Bars = 50 �tm.

Giant trophoblast cells of the mouse placenta also contain immu-

noactive placental lactogen (Hall and Talamantes, 1984). Placental

lactogen has been localized in binucleate cells of the fetal choni-

onic epithelium and the syncytial layer ofthe placentomes of sheep

(Wooding, 1981; Reddy and Watkins, 1980; Watkins and Reddy,

1980; Carnegie et al., 1978; Martal et al., 1977). Immunoreactive

human placental lactogen (Watkins, 1978; de Ikonicoffand Cedard,

1973) and mRNA for human placental lactogen (Hoshina et al.,

1985; Boime et al., 1982; Hoshina et al., 1982; McWilliams and

Boime, 1980) have been demonstrated in the syncytium ofthe hu-

man placental villus.

Hamster placental lactogen II is first detectable in serum and

placental tissue extracts on Day 10 of gestation (Southard et al.,

1987). The reason that haPL-II immunoreactivity in placental tis-

sue sections was not detected on Day 10 is unknown. Watkins (1978)

reported that formalin fixation reduced the antigenicity of PL in

human villus samples but that fixation by dehydration using Car-

noy’s solution allowed for successful immunolocalization. Demon-

stration of haPL-II in tissue sections on Day 10 was unsuccessful

regardless of the fixative. Low serum hormone concentrations

(Southard et al., 1987) may reflect low rates of hormone synthesis

in trophoblast cells on Day 10. It is possible that the quantity of

.w

by guest on March 16, 2016jhc.sagepub.comDownloaded from

haPL-II IMMUNOLOCALIZATION 939

hormone within individual cells on Day 10 is below the sensitivity

of the methodology used.

Onsini (1954) concluded that primary and secondary giant

trophoblast cells of the hamster placenta are phagocytic and that

these cells may function in penetration ofthe uterine epithelium,

enlargement of the decidual cavity, and degeneration of maternal

tissue during implantation and placental growth. Association of

haPL-II with these cells may indicate a role for this hormone in

placental growth and development. Hamster placental lactogen II

binds to prolactin receptors in the mouse mammary gland and

stimulates a-lactalbumin synthesis by these same tissues (Southard

et al. , 1986). Therefore, haPL-II may have a role in lactogenesis in

the hamster.

Arteries that supply the hamster placenta are distinguished by

their prominent sheaths (Orsini, 1954). These arteries have an en-

dothelium and a single layer ofmuscle surrounded by layers of hyper-tnophied connective tissue cells. Sheathed arteries are located within

the decidua basalis early in pregnancy, and mesometnial arteries

undergo this transformation later in pregnancy. Near the time of

partunition, sheath cells surrounding arteries ofthe decidua basalis

undergo degeneration to form eosinophilic bodies. Although these

structures contained haPL-II immunoreactivity, the significance of

this observation is unknown.

Orsini (1954) described two types oflarge cells, the tertiary gi-

ant trophoblast cells and the endovascular cells, which reside within

the lumen ofmesometrial arteries during late pregnancy. Both cells

are of fetal origin; however, their function is not known. Endovas-

culan cells predominate at this stage of pregnancy and are probably

those stained for haPL-II in this study. The significance of haPL-II

staining in these cells is unknown.

Specificity control experiments demonstrated that staining with

each antiserum was specific for the antigen in question. Additional

evidence ofantibody specificity is implied by the fact that haPL-II

positive staining was not observed on Days 8 and 10 when relaxin

immunoreactivity is present (Renegar et al. , 1987). Data from this

study indicate that haPL-II and relaxin are localized within and

probably synthesized and secreted by the same cell. Localization

of a single hormone in any given cell is likely owing to limitations

of the techniques used to determine co-localization. Staining in-

tensity for haPL-Il was greater than that for relaxin and may ac-

count for the detection ofcells stained only for haPL-II. This differ-

ence in staining intensity may be due to haPL-II localization using

a homologous system (antibody to the native hormone) and relaxin

localization using a heterologous system (antibody to porcine

relaxin). Relaxin is detectable in GTC on Days 8 and 10 of gesta-

tion in the hamster (Renegar et al. , 1987), whereas haPL-II was not

detected in the present study until Day 12. Temporal differences

in the appearance of haPL-II and relaxin in the Git suggests that

different factors regulate synthesis and secretion of these hormones

on that hormone expression evolves as Git differentiation proceeds.

Prolactin and relaxin C-peptide have been immunolocalized in

cells of the panietal decidua adherent to fetal membranes and in

decidua-like cells of the placental basal plate of the human (Sak-

bun et al., 1987). Although, the significance ofthis co-localization

in the placenta is unknown, Sortino et al. (1989) recently reported

that human relaxin stimulated prolactin secretion from dispersed

pituitary cells. Therefore, a direct paracnine effect ofhuman placental

relaxin on decidual prolactin secretion is possible. Additional study

will be required to determine ifsuch a relationship exists between

hamster relaxin and haPL-II.

Literature Cited

Boime I, Boothby M, Hoshima M, Daniels-McQueen 5, Darnell R (1982):

Expression and structure ofhuman placental hormone genes as a function

of placental development. Biol Reprod 26:73

Carnegie JA, McCully ME, Robertson HA (1978): Localization of ovine

placental lactogen by immunofluorescence using sections of tissue embed-

ded in glycolmethacrylate. J Histochem Cytochem 26:223

de lkonicoffLK, Cedard L (1973): Localization ofhuman chorionic gonad-

otropic and somatomamotropic hormones by the peroxidase immunohisto

enzymes method in villi and amniotic epithelium ofhuman placentas (from

six weeks to term). Am J Obstet Gynecol 116:1124

Duckworth ML, Kirk KL, Fniesen HG (1986): Isolation and identification

of a cDNA clone of rat placental lactogen II. J Biol Chem 21:10871

Hall J, Talamantes F (1984): Immunocytochemical localization of mouse

placental lactogen in the mouse placenta. J Histochem Cytochem 32:379

Hoshina M, Boothby M, Hussa R, Pattillo R, Camel HM, Boime I (1985):

Linkage of human chonionic gonadotrophin and placental lactogen bio-

synthesis to trophoblast differentiation and tumorigenesis. Placenta 6:163

Hoshina M, Hussa R, Pattillo R, Camel HM, Boime I (1982)-� The role of

trophoblast differentiation in the control of the hCG and hPL genes. AdvExp Med Biol 176:299

Jackson U, Colosi P, Talamantes F, Linzer DIH (1986): Molecular cloning

of mouse placental lactogen cDNA, Proc NatI Acad Sci USA 83:8496

Kelly PA, Tsushima T, Shiu RPC, Fniesen HG(1976): Lactogenic and growth

hormone-like activities in pregnancy determined by radioreceptor assays.Endocrinology 99:765

MartalJ, DijianeJ, Dubois MP (1977): Immunofluorescence localizationof ovine placental lactogen. Cell Tissue Res 184: 427

McWilliams D, Boime I (1980): Cytological localization ofplacental lacto-

gen messenger nibonucleic acid in syncytiotrophoblast layers of humanplacenta. Endocrinology 107:761

Orsini MW (1954): The trophoblastic giant cells and endovascular cells as-

sociated with pregnancy in the hamster, Cricetus auratus. AmJ Anat 94:275

Reddy 5, Watkins WB (1978): Immunofluorescent localisation of ovine

placenta lactogen. J Reprod Fertil 52:173

Renegar RH, Cobb AD, Leavitt WW (1987): Immunocytochemical local-

ization of nelaxin in the golden hamster (Mesocricetus auratus) during the

last half of gestation. Biol Reprod 37:925

Robertson MC, Gillespie B, Fniesen HG (1982): Characterization ofthe two

forms ofrat placental lactogen (rPL): rPL.I and rPL-II. Endocrinology 111:1862

Sakbun V. Koay ESC, Bryant-Greenwood GD (1987): Immunocytochemi.

cal localization of prolactin and relaxin C.peptide in human decidua andplacenta. J Clin Endocninol Metab 65:339

Sortino MA, Cronin MJ, Wise PM (1989): Relaxin stimulates prolactin scene-

tion from anterior pituitary cells. Endocrinology 124:2013

Soares MJ, Colosi P. Ogren L, Llamantes F(1983): Identification and par-

tial characterization ofa lactogen from the midpregnant mouse conceptus.

Endocrinology 112:1313

Soares MJ, i�lamantes F(1982): Placental and serum hormone changes during

the second half of pregnancy in the hamster. Biol Reprod 7:523

SouthardJN, Campbell GT, ‘Illamantes F(1987): Hamster placental lacto-gen: gestational profiles and high molecular weight forms. Endocrinology121:900

by guest on March 16, 2016jhc.sagepub.comDownloaded from

940 RENEGAR, SOUTHARD, TALAMANTES

SouthardJN, l#{224}lamantesF(1987): Immunological studies ofrodent placental Watkins WB (1978): Use ofimmunocytochemical techniques for the local-

lactogens. Mol Cell Endocninol 50:29 ization of human placental lactogen. J Histochem Cytochem 26:288

SouthardJN, Thoudarson G, Talamantes F (1986): Purification and partial Watkins WB, Reddy S (1980): Ovine placental lactogen in the cotyledonary

characterization of hamster placental lactogen. Endocrinology 119:508 and intercotyledonary placenta of the ewe. J Reprod Fentil 58:411

Steinetz BG, O’Byrne EM, Goldsmith LT, Anderson MB (1988): The source Wooding FBP(1981): Localization ofovine placental lactogen in sheep placen-

of relaxin in pregnant Syrian hamsters. Endocrinology 122:795 tomes by electron microscope immunocytochemistry.J Reprod Fertil 62:15

by guest on March 16, 2016jhc.sagepub.comDownloaded from