Formation of 6-keto-PGF1α by collecting tubule cells isolated from rabbit renal papillae

PROSTAGLANDINS

FORMATIONOF 6-KETO-PGFIcBY COLLECTINGTUBULE CELLS ISOLATED FROM RABBIT RENAL PAPILLAJZ Frank C. Grenier and William L. Smith Departmentof Biochemistry Michigan State University East Lansing, Michigan 48824 ABSTRACT Homogeneouspopulationsof collecting tubule epithelialcells have been isolated from rabbit renal papillae by a sequence of procedures involving:(a)dissociationof the tissue by mincing and treatment with trypsin; (b) destructionof contaminatingnon-collecting tubule cells by differentiallysis in hypotonicmedia and (c) collection and washing by repeated centrifugation. The isolated cells have been characterizedas being derived from the collectingtubules on the basis of anatomicalsource, size and histologicalstaining for both NADH diaphoraseactivity and cyclooxygenaseantigenicity. The cells are judged to be viable by several criteria includingtheir ability to exclude both trypsin and vital dyes, their capacity to metabolize glucose and leucine and their ability to retain distinctive morphology followinglo-14 days in culture media. Homogenates of freshly isolated collectingtubule cells when incubatedwith [3H]-arachidonic acid yielded radioactiveproducts identifiedby thin-layerchromatographicbehavior in multiple solvent systems as 6-keto-PGFlo,PGFzo, PGE2, PGDP and a monohydroxyacid, probably HHT. No lipoxygenase-likeactivity was detected, At arachidonateconcentrations of 2 uM or less, the major product was C-keto-PGFlo;while at substrate czcentrations of greater than 10 @f, PGE2 was the major radioactiveprostaglandinformed. Similar distributionsof products were observed when homogenatesof dissociatedrenal papillae enriched in medullary interstitialcells were incubatedwith arachidonicacid. Our results indicate that collectingtubule cells do contain significant prostacyclinsynthetaseactivity and suggest that PGI2 plays a role in the function of mammalian collectingtubules. INTRODUCTION Prostaglandinderivativesare formed by four cell types in the rabbit kidney (l-5). These include the medullary interstitialcells, the collecting tubule cells, the parietal layer of Bowman's capsule and the arterial endothelialcells. Homogeneouspopulationsof both medullary interstitialcells (6-8) and vascular endothelialcells (9,lO) have been prepared and studies with these isolated cells have provided important insights into how prostaglandinsynthesis is regulated in a unique fashion in different cell types. Parallel studies on isolated collectingtubule and parietal layer cells should lead to a further understandingof the parameters affecting prostaglandin formation in the kidney, Unfortunately,the only procedurespreviously

NOVEMBER 1978 VOL. 16 NO. 5

759

PROSTAGLANDINS

available for obtaining isolated, viable collecting tubule cells involve tedious microdissection techniques and provide rather small yields of these cells (11). We now report the development of a more rapid and convenient method for isolating viable collecting tubule cells from rabbit renal papillae in numbers useful for biochemical studies. A detailed analysis of the prostaglandin products formed by homogenates of isolated collecting tubule cells is also presented. METHODS OF PROCEDURE Materials. Samples of PGD2, PGE2, PGFze, TxB2, 6-keto-PGFIo and Flurbi~(d&d-(fluoro-4-biphenylyl) propionic acid) were a generous gift of Drs. John Pike and Udo Axen of the Upjohn Company, Kalamazoo, Michigan. Unlabeled arachidonic acid was obtained from Nu-Chek Prep, Inc., Elysian, Minnesota and [5,6,8,9,11,12,14,15-3H]arachidonic acid (60-100 Ci/mmol), &-[4,5-3H(N)]-leucine (40-60 Ci/mmol) and g-[14C(U)]-glucose (150-250 mCi/mmol) were purchased from New England Nuclear Corporation Boston, Massachusetts. Trypsin (l/250), Dulbecco's modified Eagle media, fetal calf serum, antibiotic-antimycotic (100x) and glutsmine were purchased from Grand Island Biological Company. Fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG was obtained from Miles Laboratories, Inc. Rabbit anti-cyclooxygenase serum and preimmune serum were prepared as described previously (1). Silica G was purchased from Merck. Nitroblue tetrazolium, NADH, bovine hemoglobin and BDTA were purchased from Sigma Chemical Company. Type CLS collagenase was from Worthington Biochemical Corporation. All other chemicals were reagent grade and were obtained from common commercial sources. Isolation of collecting tubule (CT) cells, New Zealand white rabbits (&-4 kg) were sacrificed by intravenous injections of lethal doses of Nembutal (0.5%) and the kidneys removed and placed in phosphate buffered saline (PBS). The papillae (0.7 g) were dissected frq each kidney, transferred to Krebs-Ringer buffer (less Ca2+ and IQ2 salts), pH 7.5 and minced finely into sections of 0.3-0.5 cm3 with a razor blade at room temperature. The minced tissue was placed in 3 ml of Krebs-Ringer buffer, pH 7.5 containing 0.5 mg of trypsin per ml. The tissue was drawn up and down a disposable pipet (3 mm bore) 15-20 times per min for 5 min and filtered through a stainless steel mesh (0.25 mm2 pore size). The initial filtrate normally contained 5-8 x 106 cells of which 95-99% were small non- collecting tubule cells. The remaining residue was removed from the filter and resuspended in 3 ml Krebs-Ringer buffer and again drawn up and down a disposable pipet 15-20 times per min (1 mm bore) until the tissue was completely dispersed (5 min). The trypsinized tissue was again filtered and the filtrate (containing l-2 x lo6 cells of which up to 40-80X were non-collecting tubule cells) diluted 1:2 with H20 and allowed to stand for 4-5 min. Reducing the osmolity caused the lysis of all small cells. Krebs-Ringer buffer (3-4 ml) was then added to the suspension and the collecting tubule cells were collected by centrifugation at 500 x g for 2 min. The cell pellet was washed twice by

760

NOVEMBER 1978VOL.16NO.5

PROSTAGLANDINS

resuspensionand centrifugationin Krebs-Ringerbuffer, pH 7.5. The final cell pellet was dispersed in Krebs-Ringeror another buffer and counted with a hemacytometerfollowing stainingwith either trypan blue and erythrosinered at a final concentrationof 0.04% (w/v) (12). The entire procedurenormally requires 90 min to complete. [3H]-leucinemetabolism. Suspensionsof CT cells were divided into 5 vials so that each contained l-4 x lo5 cells in 1 ml of KrebsRinger buffer, pH 7.5 containinga 1:50 dilution of antibiotic-antimycotic. Controls included 1 ml of buffer without cells and 1 ml of the cell suspensionto which cycloheximide(1 I.cg/ml) was added. i3Hlleucine (2-4 pCi) was added to each vial, and the cells were incubated at 37'C under a 10% CO2 atmosphere for different time intervals. To stop further metabolism of [3H]-leucine,100 umoles of unlabeled leucine and 1 mg of bovine serum albumin were added to each vial followed immediatelyby 5 ml of ice cold 10% trichloroaceticacid. Each Vial was left at 4'C for 40 minutes and centrifugedto pellet the protein. The supernatantwas removed and the pellet washed with 30 ml of ice cold 10% trichloroaceticacid on a Whatman GF/C glass fiber filter. The filter was dried and counted in 6 ml of Bray's scintillation fluid (13). [14C]-glucosemetabolism. CT cells were isolated u6ing.a Ca2+ , HC03 and glucose-freeKrebs-Ringerbuffer, pH 7.5. The final cell pellet was resuspendedin Krebs-Ringerbuffer lacking only HCO; and lucose. The cells were divided into 5 vials each containingl-4 x 10f cells in lml of buffer containinga 1:50 dilution of antibioticantimycotic. TWO controlswere performed,one without cells; the other, in which cells were incubated in the presence of NaN3 (0.02%). [14C]-glucose(5 uCi) was added to each vial and the vials sealed and incubated at 37'C. Glucose oxidationwas measured as 14C02 trapped as H14COj on KOH-soaked filter paper which was placed in a plastic ladle above the media. After different incubationtimes, 0.5 ml of 10% trichloroaceticacid was added to each vial and one hr later the filter paper was removed and counted in 6 ml of Bray's scintillation fluid. Mg2+,

NADH-diaphoraseand anti-cyclooxygenaseimmunohistochemical staining. Stainingwas performed on cells adhered to coverslips. A few drops of a CT cell suspension (5 x lo5 cells per ml of Dulbecco's modified Eagle media containing10% fetal calf serum, a 1:50 dilution of antibiotic-antimycotic and 2 mM glutamine)were placed on each coverslip, and the samples incubzed at 37'C under a 10% CO2 atmosphere for 6-8 hr at which time approximately40% of the cells had adhered. Coverslipswith cells attached were washed with PBS, frozen in isopentane (-6O'C) and then dried in a desiccatorunder water aspiration for 1 hr. The adhered CT cells were stained for NADH-diaphorase activity according to the procedure of Farber -et al. (14). Immunohistochemical staining for the prostaglandin-forming cyclooxygenasewas also performed on cells quick frozen, as above, except that coverslips to which cells were adhered were dipped into chloroform/methanol(2/l,

NOVEMBER 1978VOL. 16 NO. 5

761

PROSTAGLANDINS v/v) at -10% for 4-5 seconds and immediatelylyophilyzed. After drying for 1 hr the adhered CT cells were stained for cyclooxygenase antigenicityas describedpreviously (1,2). Fluorescenceand brightfield microscopywas performedusing a Leitz Orthoplanmicroscope. Photomicrographswere obtainedwith an Orthomat Camera using Kodak TriX Pan film (ASA 400). Prostaglandinsynthesisand characterization. CT cells or the small cells obtained in the first filtratewere susuendedin 2 ml of 0.1 g tris-chloride,pH 7.5 containing1 mM phenol.‘ Bovine hemoglobin (0.15 mg) was added to each sample of ccl%. The cells were homogenized in a glass pestle homogenizerat 4°C for l-14 minutes and then added to 1 ml of buffer containing [%I]-arachidonate diluted to varying specific activitieswith unlabeled acid. A sample to which Flurbiprofen(lo-%$ was added served as a control. The vials were incubatedwith shaking for 15 minutes at 37°C and the reactions terminated by adding 21 ml of chloroform/methanol(l/l, v/v). The protein was removed by centrifugationand 9 ml of chloroformand 4.5 ml of 0.05 M HCl were added to the supernatant. The chloroformphase was removed and dried under a stream of Np. The residueswere dissolved in 100 ul of chloroform/methanol(l/l, v/v) and applied to thin layers of silica G (0.3 mm). Chromatographywas performed in 2-butanone/isopropyl etherlmethylenechloride/benzene/HOAc (40/40/5/5/.1)-solvent system A-for 1% hr and standardsvisualized with Ip vapor. Regions containingthe PGFp,/6-keto-PGFlo, PGE2, PGD2, and TxB2 standardswere scraped into scintillationvials and counted or radioscanningwas performed on a Berthold Varian Aerograph Radioscanner. Individualprostaglandinderivativeswere further characterizedas follows. PGE2 and PGD2. Radioactivitychromatographingwith PGE2 and PGD2 in solvent system A was eluted from the silica gel with chloroform/methanol(l/l, v/v) and rechromatographedboth in hexaneiethyl acetate/HOAc(40/40/1),solvent system B-and chloroform/methanol/HOAc/ ?I20(90/9/l/0.65)-solvent system C on thin layers of silica G. In addition, aliquots of radioactivematerial from both the PGE2 and PGD2 regions of chromatogramsdeveloped in solvent system A were reduced by treatmentwith 5 mg of NaBlIt+ in 0.5 ml of methanol containing 10 mg each of carrier PGE2 and PGD2. After 40 minutes at room temperature,6.5 ml of chloroform/methanol(l/l, v/v), 3 ml of chloroform and 1.5 ml of 0.05 g HCl were added to each reaction vial. The lipid layer was removed, dried under Np, redissolvedin 100 ul of chloroform/methanol(l/l) and chromatographedin solvent system B. PGFza and 6-keto-PGFl,. Material chromatographingin the combined PGF&b-keto-PGFIo region of thin layer plates developed in solvent system A was eluted and rechromatographedin solvent systems B and C to resolve these two derivatives. The PGF&bketo-PGFI, region from chromatographyin solvent system A was also treated with N~BHI,,as described above, except that 10 ug of PGE:!,PGFzo, and 6-keto-PGFlowere used as carriers in each reaction vial. Another

762

NOVEMBER

1978 VOL. 16 NO. 5

PROSTAGLANDINS

aliquot of the PGFzo/C-keto-PGFloregion from chromatographyin solvent system A was methylatedby treatingwith ethereal diazomethane for 15 minutes at room temperature, The ether was removed under a stream of N2 and the residue dissolved in 100 ul of chloroform/ methanol (l/l) and chromatographedin solvent system C. m. Radioactivitymigratingwith TxB2 during chromatographyin solvent system A was rechromatographedin solvent systems B and C. RESULTS AND DISCUSSION Isolation and identificationof collecting tubule cells. Rabbit renal papillae contain four major cell types. Three of these types, the medullary interstitialcells, the vascular endothelialcells and the epithelialcells comprisingthe thin segment of Henle's loop are small (diameter< 6.5~) and thus can be readily distinguishedby microscopicexaminationfrom the fourth cell type, the collecting tubule epithelialcell (CT),whichhas a diameter of > 10 u. Dissociation of papillae by mincing and subsequent treatmentwith trypsin yielded mixtures of both large CT cells and contaminatingsmall cell types. All small cells disappearedwhen the isotonic preparative media was diluted 1:2 with water and incubatedfor 3 min. Apparently, the small cells were preferentiallylysed by this treatment while the number and morphology of the large cells was unaffected. The size and shape of the large cells (Fig, 1) were similar to collectingtubule cells observed in histologicalsections of renal papillae (15). Two or three CT cells were often seen adhered along their long axis forming a slight arc. Strings of CT cells lo-15 cells in length were also observed. Both the CT cells in histologicalsections and the isolated large cells stained positively for NADH-diaphoraseactivity. None of the small cells isolatedwere prominentlystained for this enzyme. These observationsare consistentwith the distributionof NADHdiaphorasein the renal medulla of the rabbit and serve to further confirm the identity of the isolated large cells as CT cells (14). Isolated CT cells were subjected to immunohistofluorescence staining with rabbit anti-cyclooxygenase serum and rabbit pre-immuneserum, respectively(Fig. 2). The CT cells stained much more intenselywith the immune serum in agreementwith staining seen in tissue sections (2,4). Metabolic characterizationof CT cells. The isolated CT cells excluded both trvnan blue and ervthrosinered dves. A maximum of lo6 CT cells were obtained per g of papillae by our-methodalthough the average yield was approximate1 half of that value. Figs. 3 and 4 illustratethe metabolism of [Y4C]-glucoseand [3H]-leucine,respectively, by isolated cells. CT cells both oxidized [14C]-glucoseto

NOVEMBER

1978 VOL. 16 NO. 5

763

PkOSTAGLANDINS

Fig. 2. Fluorescencephotomicrographsof isolated collectingtubule cells treated with (A) anti-cyclooxygenase or (B) rabbit preimmune serum then FITC-labeledgoat anti-rabbitIgG. The rounding of the cells as compared to those in Fig. 1 was caused by chloroform/methanol fixation prior to staining. Magnificationis x 500. 14C02 and incorporated[3H]-leucineinto trichloroaceticacid-precipitablematerial in time dependent fashion. Cells preincubatedfor 30 min with NaN3 (0.02%) showed a 75-90X reduction in [14C]-glucose oxidation. Incorporationof [3H]-leucineinto tricholoraceticacidprecipitableradioactivitywas inhibitedat least 80% when cells were preincubatedwith cycloheximide(1 ug/ml) for 30 min. In contrast, streptomycin,a prokaryoticprotein synthesis inhibitor,at a concentration of 100 vg/ml did not block [3H]-leucineincorporation. Isolated CT cells adhered to both glass and plastic petri dishes. Approximately40% of the isolated cells become attachedwithin a few hours when incubated in Dulbecco'smodified Eagle media supplemented with 10% fetal calf serum, 2 m& glutamine and antibiotic-antimycotic. The attached cells could be removed from these surfaces by treatment with trypsin (0.05% in PBS for 5 min) and subculturedon other petri dishes where they retained distinctiveCT cell morphologyand continued

764

NOVEMBER

1978VOL.16N0.5

PROSTAGLANDINS

0

/

/

0

0

/

/, 1 1

,

,

3

?

TIME(hr)Fig. 3.

[14C]-glucoseoxidationby isolated collectingtubule cells'. Collectingtubule cells were incubatedfor the indicated times with [14C]-glucoseand the formation of [14C]-C0, measured as described in Methods of Procedure, I

I

I

I

I

0

/

I_

0

/

I-

I0

L

/

/,

1

I

I

2

3

4

5

TIME (hr)

Fig. 4.

[3H]-leucineincorporationinto trichloroaceticacid precip: itable-products by isolated collectingtubule cells. Collecting tubule cells were incubated for the indicatedtimes with [3H]-leucineand the formation of trichloroaceticacid precipitable-radioactivity measured as in Methods of Procedure.

NOVEMBER

1978 VOL. 16 NO. 5

765

PROSTAGLANDINS

to exclude vital dyes. SubculturedCT cells when grown in a 10% CO2 atmospherestill excludedvital dyes after lo-14 days during which time only the culture media was exchanged. No increase in cell number occurred during this time. These results indicate that CT cells isolatedby trypsin dissociationcan be used for -in vivo metabolic studies. Attempts were also made to isolate CT cells followingtissue dissociationin which EDTA (O.OOOl-0.01%)and collagenase(0.02-0.1X) were substitutedfor trypsin. EDTA treatmentof minced papillae did provide relativelyhigh yields of CT cells (6-10 x lo6 cells per g papillae). The EDTA-isolatedcells also had the same capacity as trypsin-isolatedcells both to oxidize glucose and to convert exogenous leucine into protein. However, the EDTA-isolatedcells were permeable to vital dyes, were lysed rapidly by treatmentwith trypsin and were unable to repair their permeabilitydefect during a 48 hr incubationin culture media. CT cells isolatedby substitutingcollagenase for trypsin were easily lysed by relativelymild mechanical manipulationssuch as centrifugationand resuspensionproceduresand could not be maintainedovernight in a form which retained the capacity to exclude vital dyes. In addition, the yield of collagenase-isolated CT cells,apparentlybecause of their fragility,was usually less than lo5 cells per g of papillae. "Small" cell isolation. Relativelylarge numbers of small cells were obtained (5-8 x loo cells ner p;of papillae)and were contaminated with only l-5%.CT cells. The s&l-cells ;ere sphericaland approxi.. mately 6u in diameter. None of these cells stained with vital dyes or for NADH-diaphoraseactivity. We made no attempt to distinguish between vascular endothelialcells, medullary interstitialcells and the epithelialcells of the thin loop of Henle. Characterizationof arachidonicacid metabolitesformed by CT cells. Fig. 5 shows a radiochromatogramof the products synthesized from [3H]-arachidonic acid by CT cell homogenates. The major products synthesizedby homogenatesof both CT cells and small cells chromatographedwith authenticPGD2, PGE2 and PGF2, standardsas expected (16,17). The identity of radio-activitychromatographing with PGD2 and PGE2 in solvent system A was verified by chromatography in solvent systems B and C and by reductionwith NaBH4 and ahromatography in solvent system B. The two peaks of radioactivityseen in the PGF2o region of the thin layer plate (Fig. 5) were seldom resolved clearly. However, as describedbelow, further examinationindicated that 50% of the radioactivitymigrating with PGF2e was actually 13H]6-keto-PGFI,while only 30% was found to be due to [3H]-PGF2a. TxB2 was not synthesizedin significantquantities (c 3%). When cell homogenates were incubatedwith [3H]-arachidonic acid in the presence of Fluriprofen (10m4~) only unreacted arachidonatewas recovered indicating that all products were derived from the activity of the prostaglandin-forming cyclooxygenaseand not a lipoxygenase. Thus the small amount of radioactivitychromatographingin the monohydroxy

766

NOVEMBER

1978 VOL. 16 NO. 5

PROSTAGLANDINS

0

Fig. 5.

5 DISTANCE

10 FROM THE ORIGIN

15 (cm)

A radiochromatogram of the products formed upon incubationof [3H]-arachidonic acid (10 pM) with a collecting tubule cell homogenate. Products were separated by chromatography in solvent system A. Radioscanning was performed as described in Methods of Procedure. AA-arachidonlc acid; RC-ricinoleic acid.

acid region of the thin layer plate (Fig. 5) is likely K-IT and not a 20 carbon product (18). Data derived from incubations with five different CT homogenates are summarized in Table 1. TABLE 1.

Prostaglandin Products Formed From Arachidonic Acid Bg Cell Populations Isolated From Rabbit Renal Papillae

CT cell homogenates "Small" cell homogenates

PGF2a

6-keto-PGFI,

PGE2 ---

PGD2

TxB,

9-20x

30-47%

13-28%

S-9%

1-2x

12-13X

41-54x

17-26% lo-15%

aReactions were performed using [3H]-arachidonic acid (0.02 -PM). results of 5 separate experiments are summarized.

2-3% The

Material with Rf = 0.17-0.19 in solvent system A was eluted from the silica gel G with chloroform/methanol (l/l, v/v). The presence of both radioactive 6-keto-PGF1a and PGFza was indicated by comparing the chromatographic properties of the eluted material with those of various other prostaglandin derivatives in solvent systems B and C (Table 2). Treatment of the radioactive material with ethereal diazomethane converted 40% of the radioactivity into a product which also chromatographed with the methyl ester prepared from 6-keto-PGFlo. NaBI+ treatment of the radioactive metabolite and authentic 6-keto-PGFla failed to alter the chromatographic properties of either compound in

NOVEMBER

1978 VOL. 16 NO. 5

767

PROSTAGLANDINS

agreementwith the finding of Pace-Asciak (19), although PGD2 and PGE2 were reduced to PGF isomers by NaBI&,(>90%) under these conditions. TABLE 2.

Characterizationof 6-keto-PGFIc(Rf Values) Metabolite

6-k-PGFIoa

PGEz --

PGD2

PGF20a

Solvent System A

0.17-0.19

0.19(100%) 0.28

0.41

0.17(100%)

Solvent System B

0.65-0.69

0.68(57%)

0.68

0.80

0.52(25%)

Solvent System C

0.23-0.26

0.25(62%)

0.24

0.35

0.15(28%)

Solvent System B (NaBI&,treated)

0.65-0.69

0.68(51%)

0.52

0.51

0.52(35%)

Solvent System C (methyl esters)

0.40-0.43

0.41(40%)

0.41

0.51

0,24(36X)

aPercentagesin parenthesesindicate the percentageof radioactivity with Rf = 0.17-0.19 in solvent system A which chromatographswith 6-keto-PGFluor PGF2, in the subsequentsolvent systems, We performed two control experimentswhich verified that 6-keto-PGFI,formationwas actually being catalyzedby the biosynthetic machinery originatingfrom the intact CT cells and not by enzymes derived from cellular debris which might possibly have remained after hypotonic lysis of small papillary cells. In the first experiment,we examined the supernatantobtained from the final wash of CT cells, but found no prostaglandinbiosyntheticactivity. In the second experiment,we compared 6-keto-PGFI,synthesisby homogenates of two differentpreparationsof CT cells: one in which trypsin (0.5 mg/ml) was includedand the other in which trypsin was excluded during the hypotonic lysis step. We reasoned that trypsin would inactivateany prostaglandinbiosyntheticenzymes not sequestered within the intact plasma membrane. We found no differencesin the amounts of the different radioactiveproducts formed per cell by homogenatesof CT cells which had been isolatedby each of these two procedures. Fig. 6 shows a time course for the productionof various prostaglandin derivativeswhen isolated CT cell homogenateswere incubated with 2.5 uM arachidonicacid. PGE2 and 6-keto-PGFI,were synthesized at initialrates of 150 pmol/min/106cells and PGD2 and PGF2o at rates of 80 pmol/min/106cells. The reactionswere completewithin 10 min and prior to the complete utilizationof substrateacid apparently reflectingthe self-catalyzeddestructionof the cyclooxygenase(20). Fig. 7 compares the amounts of differentprostaglandinssynthesized at different initial concentrationsof arachidonate. At concentrationsof arachidt?nic acid of less than 2 G, 6-keto-PGFI,was the major product formed while at higher arachidonateconcentrations, PGE2 was the major product. Vane and coworkers (21) and Sun et al. (22) have made similar observationsof the effect of substratZ-cXiicentrationson the prostaglandinproduct distributionin other tissues.

768

NOVEMBER

1978 VOL. 16 NO. 5

PROSTAGLANDINS

2

6

4

8

IO

TIME(min)

_ Fig. 6.

Fig. 7.

Time course for the biosynthesis of various prostaglandin products by a collecting tubule cell homogenate incubated with [3H]-arachidonicacid (2.5 PM). Reactions were performed for the indicated times and the products analyzed as described in Methods of Procedure.

Prostaglandin biosynthesis by collecting tubule cell homogenates at different initial concentrations of [3H]-arachidonic acid. Incubations were performed for 15 min and the radioactive products measured as described in Methods of Procedure.

NOVEMBER

1978 VOL. 16 NO. 5

769

PROSTAGLANDINS

of the PGI2 synthetase is much lower than that of Apparently, the I$,, the isomerase catalyzing PGE2 formation. We suspect that the fall in overall 6-keto-PGFlo production noted at high concentrations of arachidonic acid (e.g. 100 pM) may have resulted from the presence of small but inhibitory levels 3 hydroperoxide contaminants (23) in the substrate solution. Although PGI2 is known to be formed by the renal cortex (24,25), apparently by renal vascular endothelial cells (26), the renal medulla has previously been reported to form only the classical prostaglandin derivatives (5,7,8,16,17,27). Thus, it was somewhat surprising to find that isolated collecting tubule cells as well as populations enriched in medullary interstitial cells have the capacity to produce significant quantities of PGI2. It seems reasonable to assume simply on the basis of cellular economy that the prostacyclin synthetase activity will be expressed at some stage during the function of the collecting tubule cell. It will be of interest to determine what stimuli cause the formation of PGI2 by these cells -in vivo. ACKNOWLEDGMENTS This work was supported in part by U.S.P.H.S. Grants No. HD-10013 and AM-22042. REFERENCES 1.

Smith, W.L., and T.G. Bell. ImmunohistochemicalLocalization of the Prostaglandin-FormingCyclooxygenase in the Mammalian Renal Cortex. Am. J. Physiol., in press.

2.

Smith, W.L., and G.P. Wilkin. Immunochemistry of Prostaglandin Endoperoxide-FormingCyclooxygenases: The Detection of the Cyclooxygenase in Rat, Rabbit and Guinea Pig Kidneys by Immunofluorescence. Prostaglandins 13:873-892, 1977.

3.

Janszen, F.H.A., and D.H. Nugteren in Advances in the Biosciences: International Conference on Prostaglandins (S. Bergstrom, ed.) Pergamon Press, Vieweg, 1973, pp. 287-292.

4.

Smith, W.L., F.C. Grenier, T.G. Bell and G.P. Wilkin in Prostaglandins in Cardiovascular and Renal Function (A. Scriabine, ed). Spectrum Publications, Inc. Holliswood, N.Y., in press.

5.

Bohman, S.O. Demonstration of Prostaglandin Synthesis in Collecting Duct Cells and Other Cell Types of The Rabbit Renal Medulla. Prostaglandins 14:729-744, 1977.

6.

Muirhead, E.E., G. Germain, B.E. Leach, J.A. Pitcock, P. Stephenson B. Brooks, W.L. Brosius, E.G. Daniels, and J.W. Hinman. Production of Renomedullary Prostaglandins by Renomedullary Interstitial Cells Grown in Tissue Culture. Circ. Res. 30/31 (Suppl. 2): 11-161-11-172, 1972.

770

NOVEMBER

1978 VOL. 16 NO. 5

PROSTAGLANDINS

7.

Eusman, R.M., and H.R. Keiser. Prostaglandin Biosynthesis by Rabbit Renomedullary Interstitial Cells in Tissue Culture. J. Clin. Invest. 60:215-233, 1977.

8.

Eusman, R.M., and H.R. Keiser. Prostaglandin Eg Biosynthesis by Rabbit Renomedullary Interstitial Cells in Tissue Culture. J. Biol. Chem. 252:2069-2071, 1977.

9,

Gimbrone, M.A., Jr., and R.W. Alexander, Angiotensin II Stimulation of Prostaglandin Production in Cultured Human Vascular Endothelium. Science 189:219-220, 1975.

10.

Harker, L. in Proceedings of the 1977 Winter Prostaglandin Conference. Prostaglandins 14:201-202, 1977.

11.

Burg, M., J. Grantham, M. Abramow, and J. Orloff. Preparation and Study of Fragments of Single Rabbit Nephrons. Am. J. Physiol. 210:1293-1298, 1966.

12.

Phillips, H.J., in Tissue Culture, Methods and Applications (P. Kruse, Jr. and M. Patterson, Jr., editors) Academic Press, London, 1973, p. 406.

13.

Bray, G.A. A Simple Efficient Liquid Scintillator for Counting Aqueous Solutions in a Liquid Scintillation Counter. Anal. Biochem. 1:279-285, 1960.

14.

Farber, E., W.H. Sternberg, and C.E. Dunlap. Histochemical Localiration of Specific Oxidative Enzymes: I. Tetrazolium Stains for Diphosphophyridine Nucleotide Diaphorase and Triphosphopyridine Nucleotide Diaphorase. J. Histochem. and Cytochem. 4:254-265, 1956.

15.

Bloom, W., and D.W. Fawcett, In: A Textbook of Histology. W. B. Saunders Co., London, p. 674, 1968.

16.

Anggard, E., S.O. Bohman, J.E. Griffin, III, C. Larsson, and A.B. Maunsbauch. Subcellular Localization of the Prostaglandin System in the Rabbit Renal Papilla. Acta Physiol. Stand. 84: 231-246, 1972.

17.

Blackwell, G.J., R.J. Flower, and J.R. Vane. some Characteristics of the Prostaglandin Synthesizing System in Rabbit Kidney Microsomes. Biochim. Biophys. Acta.'398:178, 1975.

18.

Hsmberg, M., and B. Samuelsson. Prostaglandin Endoperoxides. Novel Transformations of Arachidonic Acid in Human Platelets. Proc. Nat. Acad. Sci. (U.S.A.) 71:3400-3464, 1974.

19.

Pace-Asciak, C. A New Prostaglandin Metabolite of Arachidonic Acid. Formation of Ccketo-PGF1, by the Rat Stomach. Experientia 32/3:291-292, 1975.

NOVEMBER 1978VOL. 16 NO, 5

771

PROSTAGLANDINS

20. Smith, W.L. and W.E.M. Lands, Oxygenation of Polyunsaturated Fatty Acids during Prostaglandin Synthesis by Sheep Vesicular Gland. Biochemistry 11:3276-3285, 1972.

21. Cottee, F., R.J. Flower, S. Moncada, J.A. Salmon and J.R. Vane. Synthesis of 6-keto-PGFl, by Ram Seminal Vesicle Microsomes. Prostaglandins 14:413-423, 1977.

22. Sun, F.F., J.P. Chapman, and J.C. McGuire. Metabolism of Prostaglandin Endoperoxide in Animal Tissues. Prostaglandins 14:10551074, 1977.

23. Bunting, S., R. Gryglewski, S. Moneada, and J.R. Vane. Arterial Walls Generate from Prostaglandin Endoperoxides a Substance (Prostaglandin X) Which Relaxes Strips of Mesenteric and Coeliac Arteries and Inhibits Platelet Aggregation. Prostaglandins 12:897-913, 1976.

24. Needleman, P., S.D. Bronson, A. Wyche, M. Sivakoff and K.C. Nicolaou. Cardiac and Renal Prostaglandin 12.. J. Clin. Invest. 61:839-849, 1978.

25. Wharton, A.R., M. Smigel, J.A. Oates and J.C. Frolich. Evidence for Prostacyclin Production in Renal Cortex. Prostaglandins 5:1021, 1977.

26, Terragno, N.A, and A.J. Lonigro, personal communication. 27. Whorton, A..R.,M. Smigel, J.A. Oates and J.C. Frolich. Regional Differences in Prostacyclin Formation by the Kidney: Prostacyclin is a Major Prostaglandin of Renal Cortex. Biochim. Biophys. Acta 529:176-180, 1978.

Received

772

l/25/78 - Accepted

8/23/78

NOVEMBER

1978VOL.16NO.S