Multiple Classes of Prostaglandin F2a Binding Sites in ... - CiteSeerX

BIOLOGY

OF REPRODUCTION

Multiple

41, 385-392

(1989)

Classes of Prostaglandin Subpopulations of Ovine

F2a Binding Sites Luteal Cells’

in

A. K. BALAPURE,3 I. C. CAICEDO,3 K. KAWADA,4 D. S. WATt4 C. E. REXROAD, JR.,5 and T. A. FITZ2’3 Department

of Obstetrics Uniformed

and

Gynecology3

Services

University of the Health Bethesda, Maryland

Sciences 20814

Department of Chemistr/ Parker Markey Cancer

Lucille

University Lexington,

Center

of Kentucky Kentucky 40506 and

Reproduction Laboratory5 Animal Science Institute United

Agriculture Research States Department Beltsville, Maryland

Service of Agri culture 20705

ABSTRACT A cryostorage procedure was developed to provide ovine luteal cells throughout the period of seasonal anestrus. Corpora lutea obtained from mid/weal phase, superovulated ewes were dispersed enzyma:ically. Some dispersed cells were fractionated into subpopulations by elutriation. Dimethylsulfoxide (7.5% final concentration) in Han/cs’ buffered saline was added to cells at 4 V, and dispersed cell preparations were frozen in a programmable cellfreezer and stored at -196 ‘C. After recovery from cryopreservation, cell viability and prostaglandin F2a (PGF2& binding characteristics of thawed cells were not different from those of corresponding fresh cells. Additionally, thawed cells retained the capacity to attach to culture dishes and retained responsiveness Qf progesterone secretion to prostaglandin E2 (PGE2) and ovine luteinizing hormone (LII), although rates of progesterone secretion were attenuated in thawed compared with fresh cells. The cryopreservation procedure will prove useful to relieve constraints in utilization of ovine luteal cells arising from reproductive seasonalisy in sheep. Cells retrieved from cryostorage were evaluated hy studying PGF2a binding characteristics. From saturation analyses (increasing amounts of radio/abe/ed PGF2& of PGF2a binding to unfractionated cells, we detected a single class of high affinity binding sites (Kd = 17.4 ± 2.3 nM) in addition to the nonspecific binding component. Using displacement analyses (constant radiolabeled PGF2a and increasing amounts of unlabeled PGF#{231}) and unfractionated cells, we detected additional binding sites of lower affinity (Kd = 409 ± 166 nM) as well as the nonspecific binding component. Small lzaeal cells obtained by elutriation, which were essentially devoid of large cell contamination, had only low affinity binding sites. Large luteal cells bound PGF2a predominantly at highaffinity sites but had a low affinity component that was commensurate with the proportion of small cells present in the preparation. In vivo, ovine large lutea! cells may achieve appreciable PGF2a receptor occupancy in the presence of nanomolar PGF2a concentrations. Sign4ficam occupancy of the lower affinity sites on small luteal cells would require higher concentrations of PGF2a approaching the micromolar range. Differential binding of PGF2a fry the two luteal cell types is suggestive of functional differences during events such as luteolysis that involve interaction of the corpus luteum and PGF2a.

INTRODUCTION Accepted

prill

388

tSupported

by NIH

HD20780.

Gynecology, Uniformed Services Jones Bridge Road, Bethesda, MD

Sheep tion ity of the Health 20814-4799.

Sc ences,

of prostaglandin

cipitous 1976),

4301

385

are one of many

luteolysis and PGF

species

F2a (POF2a) (reviewed has been

in which in vivo

adniinistracauses

pre-

by Horton and Poyser, identified as the physio-

386

BALAPURE

logic luteolysin in the ewe (McCracken et al., 1972). Despite intensive investigations, the mechanism by which PGF2a results in luteolysis remains unknown, although the luteolytic effects of PGF2a have been ascribed in part to changes in blood flow (Nett et al., 1976). Since ovine corpora lutea have abundant PGF2a receptors (Powell et a!., 1974), it seems likely that PGF2a effects on the corpus luteum are mediated at least in part by a direct interaction with luteal cells. A variety of changes in luteal tissues in vitro have been demonstrated after PGF2a-induced luteolysis, including effects upon luteinizing hormone (LH)-stimulaied activity of adenylate cyclase (Evrard et at., 1978; Fletcher and Niswender, 1982), membrane fluidity (Buhr et al., 1979; Riley and Carison, 1985), and activities of various steroidogenic enzymes (Behrrnan et a!., 1971; Calfrey et a!., 1979; Torday et a!., 1980). More recently, PGF2 has been implicated in turnover of phosphatidylinosito! and calcium mobilization in rat (Leung et al., 1986) and bovine (Davis et al., 1987) lutea! cells. However, none of the effects of PGF2a manifested by luteal tissue in vitro have been convincingly correlated with the rapid and marked degenerative events of luteolysis. Of concern in studies of POF2-induced luteolysis is the relative lack of PGF2a potency upon corpora luteal function in vitro. Intramuscular injection of PGF2a to ewes results within 12 h in a rapid decline in circulating progesterone to near-basal levels (22 jim) and SLC (12-22 rim) by using a modification of an elutriation procedure described earlier (Fitz et a!., 1982). Immediately after enzymatic dispersion, cells were applied to a Beckman e!utriator equipped with a Sanderson chamber, at a rotor speed of 1600 rpm in buffer (Ca2-, Mg2-free HBSS, 0.1% BSA, pH 7.35). Two hundred milliliters of buffer were eluted and discarded at a flow rate of 12 mi/mm, which removed most erythrocytes and other nonsteroidogemc cells. The SLC fraction was harvested at 1600 rpm in 200 ml of buffer eluted at 24 mI/mm. One hundred milliliters of buffer was harvested and discarded at 1600 rpm and 30 mI/mm; then the fraction containing LLC was eluted at 950 rpm and 24 mI/mm. Cell fractions were pelleted by centrifugation at 300 x g; resuspended in complete HBSS, 0.1% BSA, pH 7.35; and assessed for yield, purity and viability with a hemocytometer and trypan blue exclusion. Fractionated cells were stored overnight at 4#{176}C before cryopreservation or use in assays. Cryopreservation

Dimethyl

sulfoxide at 4#{176}C was added to a 7.5% final to cells in HBSS-0.1% BSA. One-milliliter aliquots containing 106 large cells or i#{248} small cells were added to cryotubes and frozen in a Cryo-Med Model 801 programmable cell freezer. The cooling gradients were linear from 4#{176}C to -38#{176}Cin 42 min and concentration

LUTEAL

CELLS

387

from -38#{176}Cto -90#{176}Cin 5 miii. Frozen cells were transferred to liquid nitrogen and stored at -196#{176}C. Upon retrieval from cryostorage, frozen cells were thawed rapidly in a water bath at 37#{176}C. After thawing, cells were placed on ice, pelleted by cenirifugaxion, washed, and resuspended in assay buffer. Viability of thawed !uteal cells was assessed with txypan blue exclusion. No decrease in viability of frozen cells was detectable after cryostorage for periods up to 9 mo.

Cell

Incubations

Parameters of progesterone secretion were compared between cells that were plated onto culture dishes on the day of tissue dissociation and cells from the same lot that had been stored at -196#{176}Cfor approximately 1 mo before use. In each case, !uteal cells were allowed to attach to 25-cm culture dishes by overnight incubation at 37#{176}C under 95% air:5% CO2 in TCM199 containing 25 mM HEPES, 0.1% BSA, 2% FBS, pH 7.35 (Ml 99). Small !uteal cells were seeded at a concentration of 2 x 1 5 cells/dish; large luteal cells were seeded at a concentration of 75 x i03 cells/dish; mixed cells utilized 37 x i03 LLC and approximately 18 x i0 SLC/dish. After attachment, dishes were washed three times with serum-free M199 before addition of treatment regimens in serum-free M199. Although cells were to be used primarily for studies of PGF receptors, responsiveness to POE2 was monitored since the stimulatory response to POE2 was more readily detectable than the inhibitory response to PGF2a (Fits et at., 1984a,b). Progesterone content of spent media was determined by a validated radioimmunoassay (Niswender, 1973). Radioreceptorassay

Procedure

The PGF2 radioreceptor assay used in this study was recently developed and validated (Balapure et at., 1989). Viable, dispersed cells were utilized in the assay to enable studies of POF2a actions upon cell fUiiCtiOfl in vitro. Receptor assay incubations were conducted at pH 5.75, since acidic incubations resulted in markedly greater binding of 3H-POF as reflected in increased abundance of both high- and low-affinity binding sites (Ba!apure et a!., 1989). Both saturation and displacement analyses were conducted in triplicate tubes of 210 jil total volume assay buffer (HBSS-BSA adjusted to pH 5.75). Saturation analyses used concentrations of 3H-PGF from 5 fmol

BALAPURE

388

to 5 pmo! per assay tube, and nonspecific binding was assessed at each concentration of 3H-PGF2a by addition of 16 jiM nonradioactive POF in parallel tubes. Displacement analyses used 250 fmol 3H-PGF2a per assay tube, and 0-16 j.tM concentrations of nonradioactive POF2a. In assays of unfractionated cells, a!iquots containing 5 x iO large luteal cells were added to each assay tube. Assays of enriched large luteal cells used 5 x iO LLC per assay tube, and assays of small luteal cells used iO SLC. Incubations were conducted in a slowly shaking water bath for 45 mm at 30#{176}C. IncubaDons were terminated by addition of 1.5 ml ice-cold wash buffer (25 mM Tris, 1 mM CaCl2, 0.1% BSA, pH 7.35); incubates were applied to glass microfiber filters on a Yeda filtration manifold and were washed twice with 1.5-rn.! volumes of cold wash buffer. Filters containing washed cells were placed in vials containing 10 ml of scintillation fluid, and were equilibrated for 16 h at 4#{176}C. Radioactivity was quantified with a Tracor Mark llI liquid scintillation counter having a tritium detection efficiency of 44%. Radioreceptor binding data were analyzed by LIGAND, which has subroutines for both saturation and displacement analyses. All experiments were repeated at least once and replicates yielded similar data. Figures were prepared from single representative experiments. Differences in progesterone secretion rates were assessed by 2-way analysis of variance within cell fractions.

‘Hot’ 0 “Cold”

0.06

ET AL.

RESULTS The procedures for tissue dispersion and elutriation were similar and yielded similar results to those reported earlier (Fits et a!., 1982). SLC were virtually devoid of contamination by LLC, and LLC contained 10-30% (by cell number) SLC. Since the large cells have a far greater mass than the small cells, it is estimated that LLC consistently contained less than 10% contamination on a mass basis. Saturation analyses (“hot” Scatchard plot) of PGF2a binding to mixed cells did not reveal the presence of a low affinity binding site, and displacement analyses (“cold” Scatchard plot) clearly revealed both a high and low affmity binding component (Fig. 1). The nonspecific binding component was removed from both types of binding analyses using routines in LIGAND. Hill plots of displacement binding data had slopes < 1.0 (data not shown). Subsequent receptor assays were conducted by using displacement analyses. Viability of cell preparations were assessed following dissociation procedures and again following thawing of frozen cell preparations. Since large lutea! cells were expected to be most sensitive to dissociation and freezing procedures, large lutea! cells were the focus of viability determinations. Cell viability in small steroidogenic cells was 90-95% for all procedures. In eight preparations on different days, large cell viability following dissociation varied from 82-95%, and following thawing after cryostorage varied from 70 to 99%. Within individual lots of cells, the cryostorage procedure had no discernible effect upon the proportion of

Scatchard Scatchard

0.20 Not frozen 0.16

U-

0

Frozen

then

thawed

0

0.04

0.12 1

0.02 0.08 0

O

0.00

0

3E-9

6E-9 Bound

9E-9

(M)

FIG. I. Scatchard plots of prostaglandin F2a (PGF) binding to dispersed ovine luteal cells using hot and cold analyses. Aliquots of freshly dispersed cells were incubated, in triplicate, with increasing amounts of 3H-PGF2 from I nCi to 2 tCi, with correction for nonspecific binding in para’lel tubes containing 16 xM unlabeled PGF2a (hot plot) or with constant 50 nCi H-PGF2 and increasing amounts of unlabeled PGF from 0 to 16 pM (cold plot). Binding data were convened to Scatchard plots using LIGAND.

0

0.04

nnn

I

0.0

5.OE-10

0o 1.OE-9

‘‘

A

1.5E-9 Bound

2.OE-9

2.5E-9

3.OE-9

(N)

FIG. 2. Prostaglandin F binding to fresh versus frozen cells. Half of the cells from a single dissociate were frozen and then thawed after! hat -196C. Cold Scatchard analyses were conducted on both unfrozen (triangle:) and frozen, then thad (circles) fractions.

PGF

1 800

0) C 0 0) _

I,

U)

TO OVINE

LUTEAL

CELLS

389

-

16001 400

BINDING

FRESH

-

-

FROZEN

1200-

I

0)

1000800

0

600

0

400#{149}-

200

H

12

LH (ng/ml) POE

Cell

(ng/ml)

fraction

0 -

10 -

O.05).

viable cells. In several lots, large cell viability even appeared to increase after cryostorage, due possibly to removal of dead cells when cells were washed immediately upon thawing (although decreased cell abundance in the thawed, washed preparations was not detectable). Freezing and thawing of dispersed luteal cells had no effect upon Scatchard analyses of PGF binding. Scatchard analyses from cells that had been frozen and maintained at -196#{176}Cfor 1 h were indistinguishable from analyses using cells from the same lot that were maintained at 4#{176}C during the freezing and thawing regimens (Fig. 2). After recovery from cryostorage, luteal cells retained responsiveness to secretagogues during subsequent incubation (Fig. 3). Effects of freezing and secretagogues were assessed by 2-way analysis of variance within each cell fraction. Freezing attenuated (p4z0.01) progesterone secretion by each cell fraction, but the patterns of responsiveness to secretagogues were not different

(p>O.OS) between fresh and frozen cells. Progesterone secretion during the second 1-h incubation was significantly (pO.05). Dispersed luteal cells before elutriation possessed both high and low affinity PGF2a binding sites (Fig. 4, middle panel). However, elutriated small cells possessed virtually no high-affinity sites (Fig. 4, bottom panel), whereas elutriated large cells contained predominantly high affinity sites (Fig. 4, top panel). The nonfractionated cells in Fig. 4 contained 5 x iO LLC + 186 x i03 SLC/assay tube; the elutriated large cell fraction contained 5 x i04 LLC + 12 x 10 SLC/assay tube; the e!utriated small cell fraction contained io SLC/assay tube. The experiment depicted in Figure 4 was replicated two additional times; in each case, the low affinity site in preparations of LLC was marginally detectable. Statistical correlation of abundance of low affinity binding in LLC preparations with abundance in

390

BALAPURE 0.04

0.03

Elutriated

o

Large Luteal

Celia

0.02

0

0.01

\o \0

0.00

I 2E-9

‘

0

IE-9

I

3t-9

4E -9

Li

cc

El

AL.

Scatchard plots of indomethacin-treated cells were indistinguishable from plots obtained from control cells that were not treated with indomethacin (data not shown). Nonsteroidogenic cells (