Expression of lymphocyte perform in the mouse uterus during pregnancy

LI MOU ZHENG*, CHAU-CHING LIU, DAVID M. OJCIUS and JOHN DING-E YOUNG

laboratory of Cellular Physiology and Immunology, The Rockefeller University, 1230 York Avenue, New York 10021, USA

* Author for correspondence

Summary

In situ hybridization and immunofluorescence wereused to study the expression of a lymphocyte pore-forming protein (perforin) in the uterus of pregnantmice. Cells expressing perforin mRNA were detectedas early as gestation day 5, whereas perforin proteinwas detected one or two days later. Although thenumber of cells expressing both perforin mRNA andperforin protein varied subsequently with time, theywere consistently observed from the day of implan-tation until parturition. The highest levels of mRNA

expression were observed sometime during mid-gestation. The high abundance of this cytolytdcprotein in the metrial gland during pregnancy andthe time course of its expression thus suggest thatGMG perforin expression is tightly regulated, prob-ably hormonally, by the uterus, and that GMG cellperforin plays an important role in the normalcompletion of pregnancy.

Key words: perforin, GMG cells, pregnancy, uterus.

Introduction

As a fetus inherits paternal transplantation antigens, itcan be viewed as a semiallogeneic graft that survivesfastened to the maternal uterus until birth, but it is notunderstood how the fetus is protected from the maternalimmune response (Beer and Billingham, 1971; Pence et al.1975; Smith and Powell, 1977; Clark and McDermott,1981; Chaouat, 1986; Baines and Gendron, 1990; Redman,1990). In addition, barriers must exist to prevent trans-mission of viral infections to the fetus, and these have yetto be properly identified.

The metrial gland is a transient uterine tissue thatdevelops near the placenta in pregnant rodents andconsists primarily of two cell types, a conspicuousgranulated metrial gland (GMG) cell and a fibroblast-likestromal cell; both cell types are believed to be maternal inorigin. The metrial gland starts to develop near theembryo a few days after implantation and is associatedwith the major blood vessels supplying the placenta. Sincethe metrial gland is thought to be involved in theimmunology of pregnancy (Mitchell et al. 1981; Mitchelland Peel, 1987; Parr et al. 1987), it became of interest todetermine whether cytotoxic mediators from the cellularbranch of immunity might be present. Specifically, thepresence of a pore-forming protein, called perforin orcytolysin (Henkart, 1985; Podack, 1985; Young, 1989;Tschopp and Nabholz, 1990), was measured during thecourse of pregnancy.

Perforin had previously been identified in the cytoplas-mic granules of in vitro cell lines of cytotoxic Tlymphocytes (CTL) and natural killer (NK) cells. Thepurified protein has potent lytic activity, and it can lyse awide variety of targets by forming tubular lesionsultrastructurally resembling those produced by comp-lement (Henkart, 1985; Podack, 1985; Young, 1989;Tschopp and Nabholz, 1990); by primary sequence analy-sis, it was also found that perforin is homologous toJournal of Cell Science 99, 317-323 (1991)Printed in Great Britain © The Company of Biologists Limited 1991

complement (Stanley and Luzio, 1988). In vivo, perforinappears to be produced by the effector lymphocytesprimarily in situations where high local concentrations ofinterleukin-2 (IL-2) are generated (Ojcius and Young,1990), such as in autoimmune disorders, transplantrejections, cancer, and acute viral infections.

The observations described here are the outgrowth ofresearch in our laboratory on the in vivo function anddistribution of perforin. Using the in situ hybridizationtechnique, which allows for detection of perforin mRNAtranscription at the single cell level during differentstages of pregnancy, we here show that the metrial glandcontains larger numbers of perforin-synthesizing cellsthan formerly observed in any of the pathologicalconditions studied until now (Young et al. 1989a,6, 1990;Muller et al. 1989). These measurements were made forperforin mRNA from day 5 of gestation (just afterimplantation) to day 19 (right before parturition).

Materials and methods

AnimalsPregnant Swiss mice were used (Charles River BreedingLaboratories, Wilmington, MA). The day at which a vaginal plugwas found was designated as day 0 of gestation. The animals weredeeply anesthesized with methoxyflurane and killed by rapiddecapitation on the indicated day of gestation.

In situ hybridization techniqueSixty four uteri were taken at embryonic ages day 4 to day 19from 32 pregnant mice. Cryostat sections (6;an) were mountedonto polylysine-coated slides and fixed in 4 % paraformaldehyde,0 . 1 M phosphate-buffered saline (PBS) for 5min, rinsed, anddehydrated in increasing concentrations of ethanol. Slides werethen acetylated in 0.25% acetic anhydride containing 0 . 1 Mtriethanolamine (pH8.0, lOmin), rinsed for lOmin with 0.2xSSC(SSC is 0 . 1 5 M NaCl, 0 .015 M sodium citrate, pH7.0), anddehydrated in increasing concentrations of ethanol. Sections were

317

incubated for 3-4 h with prehybridization buffer, followed by aheat-denatured hybridization buffer (McCabe et al. 1986) contain-ing 36S-labeled sense or antisense ribonucleotide probe. Theriboprobes were transcribed from a 1600 base-pair (bp) fragmentof the mouse perforin gene (Kwon et al. 1989) cloned into atranscription vector with bacteriophage T7 and T3 promotersarranged to produce riboprobes of both orientations. Hybridiz-ation continued for 2 days at 50°C, and sections were sub-sequently rinsed with 2x, l x and 0.5xSSC, treated with RNaseA, rinsed with RNase-free buffer, rinsed again with SSC, andfinally incubated in O.lxSSC, 10mM dithiothreitol, at 45°Covernight. The sections were then rinsed for 1 min each in 300 mMammonium acetate:ethanol, 1:1, 3:7, 1:9 (v/v), and lastly inabsolute ethanol. After drying, the sections for dark-fieldphotomicroscopy were dipped in Kodak NTB2 emulsion (38 °C),and stored dry in light-tight boxes at 4°C for 4 days. Autoradio-grams were developed, fixed, hematoxylin counter-stained, andexamined with a Zeiss photomicroscope.

Northern blot analysisUteri were removed from pregnant mice at the indicated timesafter the beginning of gestation, dissected as described forpreparation of tissues for in situ hybridization, and stored frozenfor isolation and analysis of total RNA (Chirgwin et al. 1979). A20f/g sample of RNA was either transferred directly to aGeneScreen Plus membrane (New England Nuclear) for dotanalysis or fractionated on a 1 % agarose—formaldehyde gelbefore transferring to the membrane. In either case, themembrane was then probed with a ^-labeled cDNA probe forperforin (Kwon et al. 1989). Prehybridization and hybridizationwere performed at 42 °C in a solution containing 1 M NaCl, 50%formamide, 10% dextran sulfate, 1% SDS and 100 ng ml"1 ofsalmon sperm DNA. After hybridization, the blots were washedtwice with 2xSSC/l% SDS at room temperature for 16min,followed by two washes with 0.5xSSC/l % SDS at 60°C.

Immunofluorescence techniqueTissue preparation and immunostaining were performed by atechnique as previously described (Zheng et al. 1988), withpolyclonal rabbit antiserum against purified perforin (Young et al.1986a) as the primary antibody. Briefly, the cryostat sectionswere fixed either with 4% paraformaldehyde in 0 . 1 M PBS for10 min or Zamboni's fixative for 20 min. After rinsing, thesections were preincubated with 2% normal goat serum for30 min, after which they were incubated in the primaryantiserum for 48 h at 4°C. Finally, the sections were incubated in

a 1:1000 dilution of sheep anti-rabbit IgG coupled to fluoresceinisothiocynate (Aldrich Chemical Co., Milwaukee, WT) for l h atroom temperature.

The specificity of labelling was tested by two procedures. First,the tissue was incubated with rabbit pre-immune serum insteadof primary antiserum against perforin; and second, the incubationin primary antiserum was omitted. No staining was found ineither of these negative controls.

Results

Specificity of in situ hybridizationFig. 1A shows the histological structure of a whole uterusof a pregnant mouse at day 11 of gestation. At highermagnification, GMG cells are readily distinguished in themetrial gland and decidua basalis stained with theglycogen-specific dye PAS (data not shown; and Zheng etal. 1991), which reveals the presence of large numbers ofcells containing glycoprotein-rich granules. By in situhybridization with sense and anti-sense riboprobes forperforin mRNA, the specificity of the probe was next testedon two adjacent sections of the metrial gland. Incomparison with the section hybridized with the anti-sense riboprobe (Fig. IB), a dark-field photomicrograph ofthe gland hybridized with the control, sense riboprobe didnot yield any labeling (Fig. 1C). The cross-section(Fig. IB) shows that strongly labelled cells are present,indicating that the perforin message is highly abundant.Although perforin-positive cells were mostly visible in themetrial gland, scattered cells were also found to someextent in the decidua basalis. They were rarely seen toinvade the layer of spongy trophoblasts, and none werefound in the labyrinthine region of the placenta.

Time course of perforin gene expression in the uterusAs implantation normally takes place on day 4.5 ofgestation, the search for perforin-expressing cells wasbegun on day 4. No perforin-expressing cells were detectedin the uterus at day 4, whereas low levels of GMG cellsexpressing the perforin gene were already detected at day5 (data not shown). These levels increased significantly byday 6 (Fig. 2A). Thereafter, there is a dramatic rise in the

DB

Fig. 1. Hybridization of perforin riboprobes with murine metrial glands. Anatomic overview of a cross section of the pregnantuterus at day 11 of gestation (A); x30. At this stage, the fetus (F) is well defined. The first layer above the fetus is the labyrinthine(L) region of the placenta. The decidua basalis (DB) is the first layer above the labyrinthine region, while the uppermost layer iscomposed of the metrial gland (MG). The specificity of perforin riboprobe hybridization was determined using dark-fieldphotomicroscopy of metrial glands hybridized with the (B) antisense and (C) sense probes; X100.

318 L. M. Zheng et al.

Fig. 2. Time-course of perforin gene expression from day 6 through 19 of gestation. Dark-field photomicrographs were taken ofmetrial glands removed at the indicated times of gestation .and hybridized in situ with the antisense probe for perforin. Times ofgestation were as follows: (A) day 6; (B) day 8; (C) day 10; (D) day 11; (E) day 12; (F) day 13; (G) day 14; (H) day 15; (I) day 16; (J)day 17; (K) day 18; and (L) day 19. x60.

Perforin expression in pregnant uterus 319

number of perforin-expressing cells as midgestation isapproached (Fig. 2 B-D). At day 11 (Fig. 2D), labeled cellstend to be clustered and to localize preferentially towardsthe mesometrial side of the metrial gland. By day 12(Fig. 2E), the labeled cells become widely distributedthroughout the metrial gland. The maximal number ofperforin-positive GMG cells detected by in situ hybridiz-ation was observed sometime between days 9 and 12(Fig. 2B-E), after which the signal began to diminish,reaching a minimum immediately before parturition,usually on day 20 (Fig. 2H-L). These experiments wereperformed for each day with two separate uteri, andqualitatively reproducible results were obtained eachtime. In addition, all of the in situ hybridizationexperiments described above were done in parallel with anadjacent section hybridized with the sense control, whichwas negative in all cases (data not shown).

Northern blot analysis of perforin expressionThese results were complemented by Northern (RNA) blot(Fig. 3A) and dot blot (Fig. 3B) analysis. As noted with thein situ hybridization technique above, a faint signal isalready evident on day 6 (Fig. 3). The intensity of theperforin band varies with time, confirming the in situhybridization results.

The mRNA of a murine CTL line, CTLL-R8 (Kwon et al.

1

1989), was used as a positive control in Fig. 3A (lane 5),while mRNA extracted from the uterus before day 5 ofgestation did not hybridize with the perforin probe (datanot shown).

Relationship between perforin transcription andtranslationThe temporal relationship between perforin transcriptionand perforin translation is shown in the microphotographsin Fig. 4. A strong signal detectable by the in situhybridization technique with riboprobe (Fig. 4A) wasalready seen early on day 6, whereas the presence ofperforin protein as seen by immunofluorescence (Fig. 4B)is relatively sparse at the same time, suggesting thateither there is a lag time between the expression ofperforin mRNA and perforin protein or the amounts ofperforin protein are too low to be detectable by immuno-cytochemistry. By visual inspection, it is seen that amaximal signal from in situ hybridization is reached byday 11, while the maximal signal from immunohistologydoes not appear until day 13. Fig. 4C-D shows the mRNAand protein expression on day 11. On day 19, the last dayof gestation (Fig. 4E-F), there is a marked diminution inthe expression of perforin mRNA, while the number ofcells expressing perforin protein, although decreasing, isstill large. For unknown reasons, the perforin-expressingcells also begin to aggregate shortly before term (e.g.Fig. 4F).

Perforin » - • # - 1 8 S

Actin <~ - 1 8 S

B

E 8 9 10 11 12 13

E 14 15 16 17 18 19

Fig. 3. Northern (A) and dot (B) blot analysis of perforin geneexpression in the uterus of pregnant mice. y-Actinhybridization was used as a reference for relative mRNA loadper lane. Message sizes were 2.9kb (kilobases) for perforin and2.1kb for ><-actin. Lanes in A correspond to: 1, day 6; 2, day 8;3, day 12; 4, day 16; 5, CTLL-R8. In B, uteri and placentaswere removed from pregnant mice at the indicated days afterthe beginning of gestation (E, embryonic age in days).

Discussion

Perforin is a pore-forming protein known to be present inthe granules of cytolytic lymphocytes. The purifiedmonomer has been shown to bind to a wide array of cells,where it traverses the plasma membrane and polymerizesinto nonselective, heterogeneous pores as large as 20 ran indiameter (Henkart, 1985; Podack, 1985; Young, 1989;Tschopp and Nabholz, 1990). According to the granuleexocytosis model of cell-mediated killing, interactionbetween the effector and target cells leads to effector celldegranulation, which causes perforin to be released intothe extracellular space between the effector and targetcells. Pore formation on the target subsequently results insmall ions and other solutes diffusing down their electro-chemical gradient into the target, so that the cell swellsand eventually bursts.

As most of the evidence for this model had been derivedwith CTL and NK cell lines maintained in long-termcultures in IL-2-containing medium, it became crucial todetermine the distribution of perforin in primary lympho-cytes actively engaged in cell killing. Perforin has notbeen detected in resting cytolytic lymphocytes fromspleen, liver or peripheral blood of healthy animals. Invivo, its presence has rather been restricted to pathologicalconditions where high local concentrations of IL-2 aregenerated, such as lymphocytic choriomeningitis virus-infected tissues (Muller et al. 1989; Young et al. 1989a),pancreata from non-obese diabetic mice (Young et al.19896), and herpesvirus infections (Young et al. 1990).

The results presented in this work show the specificexpression of perforin mRNA in the metrial gland, asjudged from the absence of labelled cells with the control,sense riboprobe. Surprisingly, the density of perforin-containing cells in the uterus is higher than in anypathological condition studied to date (Young et al.1989a,6, 1990; Muller et al. 1989). Perforin message is

320 L. M. Zheng et al.

Fig. 4. Temporal relationship between perform transcription (mKNA level, A,C,E) and perform translation (protein level, B,D,F).Uteri were removed early on day 6 of gestation (A and B) and on days 11 (C and D) and 19 (E and F). Entire metrial gland cross-sections were used for A, C, E and representative metrial gland sections were used for B, D, F. The dark-field micrographs (A, C,E) were obtained as described in Materials and methods. Perform protein (B, D, F) was detected by immunofluorescence using IgGcoupled to fluorescein isothiocyanate; A, C, D, E, x l l5 ; B, F, X134.

Perforin expression in pregnant uterus 321

detectable as early as day 5, and the protein is expressed athighest levels during midgestation. Significantly, implan-tation usually takes place on day 4.5, which coincides withthe onset of both decidualization and perforin synthesis inGMG cells. These observations suggest that perforinexpression in GMG cells is most likely under the control ofthe same hormones that induce decidualization.

The in vivo target of this cytolytic protein has yet to bedetermined. Two possibilities come readily to mind: thatGMG cell perforin may be required (1) to prevent verticaltransmission of viral infections to the fetus or (2) to lyseaberrant trophoblasts that may otherwise invade thematernal side. Either of these possibilities would be inagreement with the view that GMG cells are NK or NK-like cells (Peel etal. 1983; Peel and Stewart, 1984; Mitchelland Peel, 1984; Mukhtar et al. 1989; Parr et al. 1990). Thisview has recently been reinforced through a study of thesubcellular location of perforin in GMG cells. By trans-mission electron microscopy (Zheng et al. 1991), it wasobserved that GMG cells contain an abundance ofgranules, which often have a characteristic vesicular capover a fine granular matrix, as previously found for NKcells (Young et al. 19866; Groscurth et al. 1987). Inaddition, the subcellular distribution of perforin in GMGcells was studied by immunoelectron microscopy usinganti-perforin antibodies and protein A-gold as a secondligand (Zheng et al. 1991). It was found that the goldparticles are distributed mainly in the granular homo-geneous core, and that all of the granules contain perforin.The pattern of immunogold labelling of perforin in GMGcells is thus similar to that observed in NK cells(unpublished observations), and would be consistent withperforin being released from GMG cells through degranu-lation.

Lastly, as abundant levels of GMG perforin are producedin a normal uterus during pregnancy, it is likely that theperforin is usually required for the successful completionof pregnancy. At the present time, however, one cannotexclude the possibility that, under pathological conditions,GMG cells may also be lytic to other cells besidestrophoblasts or virally infected cells. Accordingly, we arecurrently attempting to establish what role GMG cellsmay play in the pathology of pregnancy. For these studies,the most reliable and sensitive GMG cell marker currentlyavailable is the expression of the perforin gene.

We are grateful to Drs Zanvil A. Cohn and Ralph M. Steinmanfor constant support and encouragement. This work was sup-ported by grants from the Juvenile Diabetes Foundation, theAmerican Diabetes Association, the American Cancer Society,the Lalor Foundation, and the National Institutes of Health.

References

BAINES, M. G. AND GBNDHON, R. L. (1990). Are both endogenous andexogenous factors involved in spontaneous foetal abortion? Res.Immun. 141, 164-158.

BEER, A. E. AND BILLINGHAM, R. E. (1971). Immunobiology ofmammalian reproduction. Adv. Immun 14, 1-84.

CHAOUAT, G. (1986) La defense du foetus contre sa mere. La Recherche17, 570-584.

CHIRCWIN, M., PRZYBYLA, A. E., MACDONALD, R. J. AND RUTTER, W. J.(1979). Isolation of biologically active nbonucleic acid from sourcesenriched in ribonuclease. Biochemistry 18, 5294-5299

CLARK, D. A. AND MCDERMOTT, M. R. (1981). Active suppression of host-vs-graft reaction in pregnant mice. HI. Developmental kinetics,

properties, and mechanism of induction of suppressor cells during firstpregnancy. J. Immun. 127, 1267-1273.

GROSCURTH, P., QIAO, B.-Y., PODACK, E. R. AND HENGARTNER, H. (1987).Cellular localization of perforin 1 in murine cloned cytotoxic Tlymphocytes J. Immun. 138, 2749-2752.

HENKART, P. A. (1985). Mechanism of lymphocyte-mediated cytotoxicity.A. Rev. Immun. 3, 31-58.

KWON, B. S., WAKULCHIK, M., LIU, C.-C., PERSECHINI, P. M., TRAPANI, J.A., HAQ, A. K., KIM, Y. AND YOUNG, J. D.-E. (1989). The structure ofthe mouse lymphocyte pore-forming protein perforin. Biochem.bwphys. Res. Commun. 158, 1-10.

MCCABE, J. T., MORRELL, J I., IVBLL, R., SCHMALE, H., RICHTER, D. ANDPTAFF, D. W. (1986). In situ hybridization technique to localize rRNAand mRNA in mammalian neurons. J. Histochem. Cytochem. 34,45-50.

MITCHELL, B. S., CRACGS, R. I. AND PEEL, S. (1981). Localisation ofimmunoglobulin (IgG) in granulated metrial gland cells ofdeciduomata of the pseudopregnant rat. J. Reprod. Immun. 3,237-241.

MITCHELL, B. S. AND PEEL, S. (1984). Identification of cells bearingleucocyte surface antigens in metnal gland tissue from rats ofdifferent gestational ages, strains or parities. Immunology 53, 63-68.

MITCHELL, B. S. AND PEEL, S. (1987). Uptake of immunoglobulin andalbumin by granulated metrial gland cells in vitro. J. Reprod Fertd.81, 59-64.

MUKHTAR, D. D. Y., STEWART, I. J. AND CROY, B. A. (1989). Leucocytemembrane antigens on mouse granulated metrial gland cells. J.Reprod. Immun. 15, 269-279.

MOLLER, C, KAGI, D., AEBISCHER, T., ODERMATT, B., HELD, W., PODACK,E. R., ZINKERNAGEL, R. M. AND HENGARTNER, H. (1989). Detection ofperforin and granzyme A mRNA in infiltrating cells during infectionof mice with lymphocytic choriomemngitis virus. Eur. J. Immun. 19,1253-1259.

OJCIUS, D. M. AND YOUNG, J. D.-E (1990). Cell-mediated killing: effectormechanisms and mediators. Cancer Cells 2, 138-145.

PAKR, E. L , PARR, M. B. AND YOUNG, J. D.-E. (1987). Localization of apore-forming protein (perforin) in granulated metrial gland cells. Biol.Reprod. 37, 1327-1336.

PAER, E. L., YOUNG, L. H. Y., PARR, M. B. AND YOUNG, J. D.-E. (1990).Granulated metrial gland cells of pregnant mouse uterus are naturalkiller-like cells that contain perforin and senne esterases J. Immun.145, 2365-2372.

PEEL, S. AND STEWART, I. (1984). The differentiation of granulatedmetrial gland cells in chimeric mice and the effect of uterineshielding during irradiation. J. Anat. 139, 593-598.

PEEL, S., STEWART, I AND BULMBR, D. (1983). Experimental evidence forthe bone marrow origin of granulated metrial gland cells of the mou3euterus. Cell Tiss. Res. 223, 647-656.

PENCE, H., PETTY, W. M. AND ROCKJJN, R. E. (1975). Suppression ofmaternal responsiveness to paternal antigen by maternal plasma. J.Immun. 114, 525-528.

PODACK, E. R. (1985). Molecular mechanism of lymphocyte-mediatedtumor cell lysis. Immun. Today 6, 21-27.

REDMAN, C. W. G. (1990). Are there immunologically mediatedabortions? If so, which mechanisms? Res. Immun. 141, 169-175.

SMITH, R. N. AND POWELL, A. E. (1977). The adoptive transfer ofpregnancy-induced unresponsiveness to male skin grafts with thymus-dependent cells. J. exp. Med. 146, 899-904.

STANLEY, K. AND LUZIO, P. (1988). A family of killer proteins. Nature334, 475-476

TSCHOPP, J. AND NABHOLZ, M. (1990). Perforin-mediated target cell lysisby cytolytic T lymphocytes. A. Rev. Immun. 8, 279-302.

YOUNG, J. D.-E. (1989). Killing of target cells by lymphocytes: amechanistic view. Physiol. Rev. 69, 250-314.

YOUNG, J. D.-E., COHN, Z. A. AND PODACK, E. R. (1986a). The ninthcomponent of complement and pore-forming protein (perforin 1) fromcytotoxic T cells: structural, immunological and functionalsimilarities. Science 233, 180-190.

YOUNG, J. D.-E., HENGARTNER, H , PODACK, E. R. AND COHN, Z. A.(19866). Purification and characterization of a cytolytic pore-formingprotein from granules of cloned lymphocytes with natural killeractivity. Cell 44, 849-859.

YOUNG, L. H. Y., FOSTER, C. S. AND YOUNG, J. D.-E. (1990). In vivoexpression of perforin by natural killer cells during a viral infection.Studies on uveitis produced by herpes simplex virus type I. Am. J.Path. 136, 1021-1030.

YOUNG, L. H. Y., KLAVINSKIS, L. S., OLDSTONE, M. B. A. AND YOUNG, JD.-E. (1989a). In vivo expression of perforin by CD8+ lymphocytesduring an acute viral infection. J. exp. Med. 169, 2159-2171.

YOUNG, L. H. Y., PETERSON, L. B., WICKBR, L. S., PERSECHINI, P. M. ANDYOUNG, J. D.-E. (19896). In vivo expression of perforin by CD8+

lymphocytes in autoimmune disease. Studies on spontaneous and

322 L. M. Zheng et al.

adoptively transferred diabetes in nonobese diabetic mice. J. Immun. ZHBNO, L. M., OJCIUS, D. M., LIU, C.-C., KBAMBR, M. D., SIMON, M M.,143, 3994-3999. PARR, E L. AND YOUNG, J. D.-E. (1991). Immunogold labelling of

ZHENO, L. M., CALDANI, M. AND JOURDAN, F. (1988). perforin and serine esterases in granulated metrial gland cells.Immunocytochemical identification of luteiniiing hormone-releasing FASEB J. 6, 79-85.hormone-positive fibers and terminalB in the olfactory system of therat. Neuroscience 24, 667-678. {Received 18 February 1991 - Accepted 27 February 1991)

Perforin expression in pregnant uterus 323