JOURNAL OF BONE AND MINERAL RESEARCH Volume 7, Number 5, 1992 Mary Ann Liebert, Inc., Publishers

Prostaglandin E2Promotes Osteoclast Formation in Murine Hematopoietic Cultures Through an Action on Hematopoietic Cells D.A. COLLINS and T.J. CHAMBERS

ABSTRACT Osteoclastic differentiation is induced from hematopoietic cells in the presence of 1,25-(OH),D3by stromal cells that are present in bone but not in hematopoietic spleen. Recent evidence suggests that prostaglandins (PGs) are essential for this process. In this communication we describe experiments in which we have examined further the role of PGE, in osteoclast formation. We found a marked reduction in basal, 1,25-(OH),D3, and IL-3-induced production of calcitonin receptor (CTR)-positive cells and bone resorption by cyclooxygenase inhibitors, which was restored by PGE, addition. Although some stromal cell types (ST2 cells) that support osteoclast formation from spleen cells produced PGs in response to 1,25-(OH),D3, others (ts8 and calvarial cells) did not, either alone or in combination with spleen cells. On the other hand, both bone marrow and spleen cells produced amounts of PGE, in response to 1,25-(OH),D3 that were sufficient to account for osteoclast formation. Osteoclast-inductive ts8 cells were able to support osteoclast formation from spleen cells in the presence of 1,25-(OH),D, or PGE, even if devitalized. Incubation of ts8 cells in these agents before devitalization did not avoid the requirement for the presence of PGE, or 1,25-(OH),D3 during subsequent incubation with spleen cells. Thus, hematopoietic cells produce sufficient PGE, for osteoclast formation, and the PGE, thus produced acts on hematopoietic precursors, which can be induced in the presence of PGE, to express CTR and resorb bone on contact with osteoclast-inductive stromal cells. The ability of osteoclast-inductive cells to support osteoclast formation appears not to rest on their ability to produce, induce, or respond to PGE,.

INTRODUCTION STEOCLASTS take ultimate origin from the hematopoietic stem and are formed by a contact-dependent interaction between hematopoietic precursors and stromal cells. ( 1 . 3 ) Osteoclast-inductive stromal cells are found in bone marrow rather than splenic stromal cell populations, and although several cell lines of accepted osteoblastic phenotype do not induce osteoclastic differentiation from spleen cell^,(^.^) at least some osteoclast-inductive stromal cell lines have been shown to be capable of osteoblastic differentiati~n(~) and therefore to belong to the osteoblastic lineage. Osteoclast formation in liquid cultures of bone marrow

0

cells has been shown to be stimulated by 1,25-dihydroxyvitamin D3 [1,25-(OH),D3].'6~7) It was recently shown that 1,25-(OH)2D3stimulates PGE, production in hematopoietic cultures; that osteoclast-like cell formation in response to 1,25-(OH)D3 is inhibited by indomethacin'8);and that PGE, induces osteoclast formation. (9.10) This dependence on PG for osteoclast formation is surprising. If the osteoclast-inductive stromal cell is of the osteoblastic lineage, this suggests that osteoclast formation occurs on or adjacent to bone surfaces; yet PGs are known to inhibit the function of mature osteoclasts disaggregated from rat and and also inhibit resorption for several hours in organ culture. (14.15) A role for PG in osteoclast formation, however, is consistent with evidence from longer organ

Department of Histopathology, St. George's Hospital Medical School, London, England.

555

COLLINS AND CHAMBERS

556

culture experiments in which P G production mediates increased bone resorption by epidermal growth factor (EGF),(I6’ platelet-derived growth factor (PDGF),(”’ transforming growth factors Q and /3 (TGF-a and /3),(”) tumor necrosis factor a (TNF-Q),(’~)and interleukin-l (IL- I ) . ( 2 0 ’ Much of the evidence for a role of PGs in osteoclast formation from hematopoietic tissue was based on counts of multinucleate or tartrate-resistant acid phosphatase-positive (TRAP) cell numbers. Because the evidence suggests that PGs may play an essential role in osteoclast formation, we elected to test the dependence of osteoclastic differentiation on PG production using CTR and bone resorption, which are more sensitive and more reliable markers of osteoclast formation than multinuclearity and T R A P . ( z 1 ~ zWe 2 1 also performed experiments to determine whether P G production is by stromal or hematopoietic cells and whether PGE, acts on stromal or on hematopoietic cells to induce osteoclast formation.

MATERIALS AND METHODS Prostaglandin E2, indomethacin (Sigma, Poole, UK), and 1,25-(OH),D3 (Roche, Welwyn Garden City, UK) were dissolved as stock solutions in absolute alcohol and diluted in culture medium immediately before use. Recombinant murine interleukin-3 (IL-3) was the generous gift of Steven C. Clark (Genetics Institute, Cambridge, MA) and was diluted in culture medium immediately before use. Parathyroid hormone (PTH; Dr. J. Zanelli, National Institute for Biological Standards, Hampstead, London, UK) was dissolved in 0.001070 acetic acid containing 1 mg/ml of bovine serum albumin (BSA; Sigma), subaliquoted into 5 IU and lyophilized. Aliquots were reconstituted with culture medium immediately before use.

Marrow cell culture Bone marrow was obtained from 6- to 8-week-old CBAca mice. The long bones were removed and dissected free of adherent tissue, the bone ends were cut across the epiphyses, and the marrow cavity flushed with HEPES-buffered medium 199 (M199; Flow Laboratories, Irvine, UK) by injecting the medium into one end of the bone using a 25 gauge needle. The collected cells were washed and suspended (2 x lo6 cells per ml) in Eagle’s minimum essential medium (EMEM, Flow) supplemented with 10% neonatal calf serum (SeraLab Ltd., Sussex, UK), M hydrocortisone (Sigma), 2 mM glutamine (GIBCO, Paisley, UK), penicillin, 50 IU/ml (Flow), and streptomycin, 50 pg/ml (Flow). The suspension was placed in the wells of a 24-well tissue culture plate (Falcon; 500 pl per well), each well containing a Thermanox coverslip (Lux, Flow) and a slice of devitalized cortical bone (see later). The cultures were then incubated in the presence of reagents at 37°C in 5% C 0 2 in humidified air. Four or six culture wells were used for each variable per experiment. Cells were fed every third day by replacing 250 pl medium with fresh medium and reagents. No attempt was

made to replace nonadherent cells. At the end of incubation the bone slices were taken for scanning electron microscopy and the coverslips for [1z51]salmon calcitonin autoradiography.

Calvarial cell stroma Calvarial cells were obtained from 7-day-old MF-1 mice. The calvariae were removed and cleaned of adherent tissue and periosteum and incubated in M199 containing 1 mg/ ml of collagenase (type 11, Sigma) for 2 h. The calvariae were then agitated with a plastic pipette and the suspended cells removed, washed twice in EMEM, suspended at lo6 cells per ml in 25 cm* tissue culture flasks, and incubated at 37°C in 5% CO,. The cells were subcultured when 95% confluent using trypsin and EDTA (GIBCO).

Stromal cell lines The osteoblastic osteosarcoma cell line UMR-106 (Dr. T. J. Martin, Melbourne, Australia) was maintained in EMEM at 37°C in 5% C 0 2 , the bone marrow-derived cell line ST2 (Riken Cell Bank, Tsukuba, Japan) was maintained in a-MEM (Flow) at 37°C in 5 % CO,, and the bone marrow-derived cell line ts8“’ was maintained in EMEM at 33°C in 5% CO,. All lines were subcultured with trypsin and EDTA when 95% confluent.

Coculture experiments Spleen cells were obtained from 7-day-old MF-I mice by piercing the spleen pulp and flushing with M199. The collected cells were washed and spun twice and suspended in EMEM. Spleen cells ( 5 x lo5 cells per well) were cocultured with stromal cells (4 x lo4 cells per well) in 200 p1 MEM supplemented with M hydrocortisone in individual wells of a 96-well tissue culture plate (Falcon), each well containing a Thermanox coverslip and a slice of devitalized cortical bone (see later). Cultures were maintained at 37°C (or 39°C in experiments using ts8 stroma) in 5% CO, in humidified air. Four or six culture wells were used for each variable per experiment. Cells were fed every third day by replacing 100 pl medium with fresh medium and reagents. No attempt was made to replace nonadherent cells. At the end of incubation the bone slices were taken for scanning electron microscopy and the coverslips for [‘251]SCTautoradiography. In some experiments the stromal cells were fixed with 1To glutaraldehyde (Sigma) in Hank’s balanced salt solution without calcium or magnesium (HBSS, GIBCO) for 2 minutes and then washed five times in HBSS before the addition of fresh medium and spleen cells. This procedure causes metabolic inactivation and fixation of the cells but allows the monolayer to support maturation of hematopoietic cells.(23’

PGE, assay Prostaglandin E2 was assayed in the supernatants from some cultures using a commercially available l z 5 I radioimmunoassay (Amersham, Amersham, UK) used according

557

PCE, INDUCES OSTEOCLASTS FROM HEMATOPOIETIC CELLS to the manufacturer's specifications. Culture supernatant (100 ml) was taken for assay immediately before feeding or after terminating the cultures after 14 days. 2

BONF IIESOHPIION

Jo 7

250

ZOO

Measurement of bone resorption

T

Slices of bovine cortcal bone were used as a substrate for osteoclastic resorption. Bone slices (2 x 2 x 0. I mm) were cut from cleaned adult bovine femora using a low-speed saw (Buehler, Lake Buff, IL), cleaned by ultrasonication, washed in acetone, and stored dry at room temperature. At the end of the incubations cells were removed from the bone slices by immersion for 10 minutes in 7% sodium hypochlorite. The slices were then washed, dehydrated in ethanol, and sputter coated with gold. Each slice was examined blind in a Cambridge S90 scanning electron microscope. The extent of bone resorption was measured by examining the whole surface of each bone slice, and the area of resorption was quantified using a 1 cm2 grid at x 100 screen magnification. The results were expressed as the percentage surface area showing bone resorption.

I !10

100

'10

0

ina

['2sI]sCTautoradiography Salmon calcitonin (sCT; Sandoz, Basel, Switzerland) was iodinated using a modification of the chloramine-T method, as previously described.(241 Coverslips were incubated with labeled sCT (0.2 nM) in medium 199 with 0.1% bovine serum albumin (Sigma) for 1 h at room temperature. After incubation they were washed with phosphate-buffered saline, fixed with formalin, air dried, and coated with K5 nuclear emulsion (Ilford, Ilford, UK). They were exposed for 14 days at 4°C and then developed and counterstained with Meyer's hematoxylin. CTR-positive cells were identified as those that demonstrated sufficient grain density to clearly outline the cells. ( 2 1 1 Nonspecific binding was assessed by including an excess (300 nM) of unlabeled sCT in some wells before labeling. Labeled mononuclear and multinuclear cells were counted in 10 random high-power fields per coverslip and expressed as the number of CTR-positive cells per cm'. Statistical comparisons of cell numbers and bone resorption were made using an unpaired Student's &test; differences were considered significant for p < 0.05.

w

2

FIG. 1. Both basal (a) and 1,25-(OH),D,-dependent (b) bone resorption and CTR-positive cell numbers were inhibited by the presence of indomethacin M) or ketoM) profen M). In both cases addition of PGE, restored CTR-positive cell production and bone resorption. [Indo, indomethacin; keto, ketoprofen; 1,25, 1,25(OH),D, M)]. *p < 0.05 versus control; O0p < 0.01 versus 1,25-(OH),D,. n = 12.

1

600

RONF RFSORPTION

RESULTS

Effects of inhibition of PG production Cultures of bone marrow cells incubated for 14 days developed CTR-positive cells and occasional foci of bone resorption (Fig. 1). Both CTR-positive cell numbers and bone resorption were increased in response to 1,25(OH),D, M) and to a lesser extent by IL-3 (1 ng/ml; Figs. 1 and 2). Simultaneous addition of indomethacin M) led to a reduction in basal, 1,25-(OH)2D,-, and

Control

IL 3

IL 3.1

I t 3+l+t

FIG. 2. CTR-positive cell numbers were increased in response to IL-3 (1 ng/ml) and inhibited by the presence of indomethacin M). Addition of PGE, M) restored cell differentiation. * p < 0.05 versus control; * p < 0.01 versus IL-3. n = 12.

COLLINS AND CHAMBERS

558

IL-3-dependent CTR-positive cell formation and bone resorption. Similar results were obtained using ketoprofen ( M), a structurally unrelated cyclooxygenase inhibitor (Fig. I). In all cases osteoclast differentiation was restored by the addition of PGE, M; Figs. 1 and 2); this suggests that PG synthesis is essential for basal, 1,25(OH),D,-, and IL-3-induced osteoclast formation in this culture system.

Assays of PGE, In cultures of bone marrow incubated for 14 days, 1,25(OH),D, caused a dose-dependent increase in PGE, production from 2.1 to 14.2 nM at 10-8M, with a dose responsiveness similar t o that of osteoclast induction (Fig. 3). We previously found that PGE, promotes CTR-positive cell formation in a dose-dependent manner.(lolIn those experiments we found that the addition of 10-7-10-8M PGE, induced similar numbers of CTR-positive cell production as did 10-O M 1,25-(OH),D,. Thus, the amount of PG production observed in our cultures can account for the extent of osteoclast formation in bone marrow cultures. We assayed PGE, production in response to 1,25(OH),D, in cultures of stromal cells that support CTR-positive cell formation (calvarial, ST2, and ts8) and in those that d o not (UMR)"] and found that neither calvarial cells nor ts8 cells produced amounts of PGE, comparable with that of spleen cell cultures, even with I ,25-(OH),D, (Table 1). This suggests that PG production is not an essential component of the ability of stromal cells to support CTRpositive cell formation (Table 1). We also assayed the production of PGE, in marrow cultures in response to other hormones and cytokines. PG production was increased in response to IL-3 (from 1.0 to 6.3 nM). Production was only minimally increased in response to PTH (from 1 .O to 2.2 nM); PTH has little effect on osteoclast formation in this culture system.(6)

TABLE1. PRODUCTION OF PGE,

Coculture experiments We found an increase in CTR-positive cell numbers and in bone resorption in cocultures of murine spleen cells and ts8 stroma in response to 1,25-(OH),D, and PGE, (Table 2 ) . CTR-positive cells were not seen in cultures of spleen cells alone or in coculture with UMR cells. Since ts8 cells d o not respond to 1 ,25-(OH),D3 with increased synthesis of PG but spleen cells do, these results suggest that 1,25(OH),D, acts on spleen cells to produce PG, which in the presence of ts8 cells leads t o osteoclast formation. In an attempt to discover whether PGE, induces osteoclast formation through an effect on the spleen cells or stromal cells, ts8 cells were incubated for 7 days in the presence of 1,25-(OH),D3, PGE,, or vehicle and then fixed with glutaraldehydes, and spleen cells were cultured on the

18

0

16

PGE, Produced

14

-

12

-

10

-

a6 4 -

2 -

0 4

FIG. 3. PGE, assayed on day 14 of bone marrow culture showed a dose-dependent increase in production in response to 1 ,25-(0H),D, together with a corresponding increase in bone resorption. * p < 0.05; **p < 0.01; * * * p < 0.001 versus control.

STROMAL CELL TYPESAND SPLEEN CELLS IN RESPONSE I ,25-(OH),D, (lo-' M)a

BY

TO

PGE,, Mean f SEM(nM) Cell type

T

0 Bone Resorollon

Control

I,25-(OH)zD3

Formation of CTR-positive cells in cultures incubated with 1,25-(oH)2D3

~~

Whole marrow Calvarial stroma Calvarial stroma + spleen UMR-106 UMR-106 + spleen ST2 ST2 + spleen ts8 ts8 + spleen Spleen

21 f 0.3 f 1.3 f 0.2 f 1.2 2.3 f 2.0 k 0.6 f 1.9 k

1.05 0.05 0.15 0.03 0.2 0.20 0.18 0.09 0.16 1 . 1 2 0.16

*

14.2 0.9 8.1 0.2 6.7 7.3 10.8 0.7 7.5

k

2.2b

f 0.09b +_

k

1.98b 0.01

f 1 3 f 1.52~

+ -

+ -

f 1.1lb f 0.07

+

f 1.88b 5.6 i 1 . 4 9

+

-

asamples were taken on day 14 of culture, 72 h after last feed. Results derived from four cultures per variable. Significance is indicated versus controls. SEM = standard error of the mean. bp

< 0.01.

cp

< 0.05.

PGE, INDUCES OSTEOCLASTS FROM HEMATOPOIETIC CELLS

559

hibited by the cyclooxygenase inhibitors indomethacin and ketoprofen in cultures of murine bone marrow cells. This inhibition appeared to be explicable as due to the inhibition of PG production, since replacement of PGE, restored osteoclastic differentiation to the cultures. Similar inhibition of production of osteoclast-like cells by indomethacin was previously reported in murine bone marrow cells.(') However, those authors found that levels of PGE, three orders of magnitude greater than those present in such cultures were required to induce numbers of osteoclast-like cells similar to those induced by 1,25(OH),D3 and concluded that PGE, could not account for osteoclast-like cell production. In those experiments, osteoclastic differentiation was measured as TRAP and multinucleate cell formation, criteria that can both underestimate osteoclastic differentiation [since CTR-positive DISCUSSION cells can be mononuclear but functionally osteo' ~ ~overestimate ~)] osteoclast formation [since We found that osteoclastic differentiation, both "spon- ~ l a s t i c ~ ~and taneous" and in response to IL-3 and 1,25-(OH),D3, is in- 1,25-(OH),D3 induces multinuclearity and TRAP in macrophages that remain CTR negative and nonre~orptive('~)]. PGE, production in our cultures, similar in quantity to that previously reported, 1') appears to be both necessary and sufficient('') for osteoclast generation in bone marrow OF CTR-POSITIVE CELLS AND BONE TABLE 2. PRODUCTION cultures by 1,25-(OH),D3. IN COCULTURES OF TS8 CELLS AND MURINE RESORPTION Spleen cells produced amounts of PGE, similar to those SPLEEN CELLSIN RESPONSETO PGE, M) AND in bone marrow in response to 1,25-(OH),D,, but unlike 1,25-(OH),D3 (lo-' M) AFTER14 DAYSIN CULTURE^ bone marruw cells did not produce osteoclasts. Osteoclasts CTR-positive Bone were produced, however, when spleen cells were coculcells per cmz, resorption tured with ts8 cells. Although ST2 cells, which like ts8 supMean + SEM (Yo) port osteoclast formation from spleen cells, produced considerable quantities of PGE, in response to 1,25Control 0 0 OH),D3, ts8 cells did not. Thus, osteoclast-inductive stro1,25-(OH),D3 36 f 13 5 ma1 cells do not seem necessary for the PGE, production PGE, 107 f 20 3 upon which osteoclast formation depends in such cultures. 1,25-(OH),D3 0 0 Rather they are required to provide an additional factor, + indomethacin M) not found in spleen cell cultures, that induces osteoclast formation. an = 12 cultures from three separate experiments. fixed monolayer in the presence or absence of 1,25(OH),D, M) or PGE, M). Hematopoiesis was reduced when compared with cultures using viable stroma, but after 28 days in culture characteristic resorption lacunae were seen in cultures treated with 1,25-(OH),D3 or PGE, after fixation, but not in those incubated with the hormones before fixation (Table 3). Resorption was also not seen when the cultures were repeated using UMR cells instead of ts8 (data not shown). No resorption was seen when ts8 cells were incubated with PGE, or 1,25-(OH),D, for 7 days and subsequently incubated without hormones with spleen cells, even if the ts8 cells were not fixed before addition of spleen cells (Table 3).

TABLE 3. DEVELOPMENT OF OSTEOCLASTS FROM SPLEEN CELLS INCUBATED ON FIXEDOR UNFIXED TS8 CELLSa

Before spleen cell addition Control Control Control PGE, (10-6 M) PGE, M) 1,25-(OH)iD, (lo-' M) 1,25-(OH),D, (lo-' M) 1,25-(OH),Dj (lo-* M)b PGE, M)b

After spleen cell addition

No. bone slices showing resorption

Control PGE, M) 1,25-(OH),D, (lo-' M) Control PGE, M) Control 1,25-(OH),D3 (lo-' M) Control Control

0/12 6/12 10/12 0/12 8/11 0/12 7/12 0/12 0/12

aThe ts8 cells were incubated for 7 days in the presence of PGE,, 1,25-(OH),D,, or vehicle and then fixed with glutaraldehyde for 2 minutes. Spleen cells were then cultured on the fixed or unfixed monolayer in the presence or absence of the hormones for a further 28 days. n = 12. bThese cells were allowed to remain viable.

COLLINS AND CHAMBERS

560

The additional factor required for osteoclast formation can be provided not only by viable but by devitalized stroma1 cells. Osteoclast production was much reduced on devitalized compared to viable stromal cells, presumably because the cell surface/extracellular matrix moieties responsible for osteoclast induction are damaged by glutaraldehyde fixation. However, the results demonstrate that expression of this osteoclast-inductive stromal cell factor by ts8 cells is constitutive and does not require incubation in 1,25-(OH),D, or PGE, for its expression (although either or both of these agents may increase this activity in stromal cells). Osteoclast production in cocultures of devitalized stroma and spleen cells remained dependent on the presence of 1,25-(OH),D, or PGE,. Thus, 1,25-(OH),D, induces osteoclast formation through induction of PG production by spleen cells. The PGE, thus produced then acts on hematopoietic precursors, which, in contact with ts8 cells, differentiate into osteoclasts. The cell type(s) that produce PGE, in spleen cell cultures in response to 1,25-(OH),D, is unknown. It may be a component of the hematopoietic stroma, which like ST2 cells responds to 1 ,25-(OH),D3 by PGE, production. Macrophages are also known to secrete PGE,.(26)It is intriguing to speculate that deficient osteoclast production in the op/ op mouse, which is due to absence of macrophage colonystimulating factor (M-CSF), ( 2 7 - 2 9 1 may be due to failure of bone marrow production of PGE, due to the simultaneous deficiency of macrophages in these animals. This is especially so since M-CSF, which enables osteoclast formation in op/op a n i m a l ~ , ( ~ Oinduces ,~') PGE, production in macrophages.

ACKNOWLEDGMENT This work was supported by the Arthritis and Rheumatism Council, UK.

REFERENCES I . Hattersley G , Chambers TJ 1989 Generation of osteoclasts from hemopoietic cells and a multipotential cell line in vitro. J Cell Physiol 140:478-482. 2. Hagenaars CE, Van der Kraan AAM, Kawilarang-de Haas EWM, Visser JWM, Nijweide PJ 1989 Osteoclast formation from cloned pluripotent haemopoietic stem cells. Bone Miner 6:187-190. 3. Takahashi N, Akatsu T, Udagawa N, Sasaki T, Yamaguchi A, Moseley JM, Martin T J , Suda T 1988 Osteoblastic cells are involved in osteoclast formation. Endocrinology 123: 2600-2602. 4. Hattersley G, Chambers TJ 1990 The role of bone marrow stroma in induction of osteoclastic differentiation. In: Cohn DV. Glorieux FH, Martin TJ (eds.) Calcium Regulation and Bone Metabolism: Basic and Clinical Aspects, Vol. 10. Elsevier Science Publishers, Amsterdam, pp. 439-442. 5. Yamashita T, Asano K , Takahashi N, Akatsu T, Udagawa N, Sasaki T, Martin T J , Suda T 1990 Cloning of an osteoblastic cell line involved in the formation of osteoclast-like cells. J Cell Physiol 145:587-595. 6. Fuller K, Chambers TJ 1987 Generation of osteocalsts in cultures of rabbit bone marrow and spleen cells. J Cell Physiol

132:441-452. 7. Takahashi N, Yamama H, Yoshiki S, Roodman DG, Mundy GR, Jones SJ, Boyde A, Suda T 1988 Osteoclast-like cell formation and its regulation by osteotropic hormones in mouse marrow cultures. Endocrinology 122: 1373-1382. 8. Shinar DM, Rodan GA 1990 Biphasic effects of transforming growth factor @ on the production of osteoclast-like cells in mouse bone marrow cultures: The role of prostaglandins in the generation of these cells. Endocrinology 1263153-

3158. 9. Akatsu T, Takahashi N, Debari K, Morita I , Murota S. Nagata N, Takatani 0, Suda T 1989 Prostaglandins promote osteoclast like cells formation by a mechanism involving cyclic adenosine 3',5'-monophosphate in mouse bone marrow cell cultures. J Bone Miner Res 4:29-35. 10. Collins DA, Chambers TJ 1991 Effect of prostaglandins E,, E,, and F,, o n osteoclast formation in mouse bone marrow cultures. J Bone Miner Res 6:157-164. 11. Chambers T J , McSheehy PMJ, Thomson BM, Fuller K 1985 The effect of calcium-regulating hormones and prostaglandins on bone resorption by osteoclasts disaggregated from neonatal rabbit bones. Endocrinology 116:234-239. 12. Arnett TR, Dempster DW 1987 A comparative study of disaggregated chick and rat osteoclasts in vitro: Effects of calcitonin and prostaglandins. Endocrinology 120:602-608. 13. Fuller K, Chambers TJ 1989 Effect of arachidonic acid metabolites on bone resorption by isolated rat osteoclasts. J Bone Miner Res 4:209-215. 14. Conaway HH, Diez LF, Raisz LG 1986 Effects of prostacyclin and prostaglandin E, (PGE,) on bone resorption in the presence and absence of parathyroid hormone. Calcif Tissue Int 38:130-134. 15. Lerner UH, Ransjo M, Ljunngren 0 1987 Prostaglandin E, causes a transient inhibition of mineral mobilization, matrix degradation, and lysosomal enzyme release from mouse calvarial bones in vitro. Calcif Tissue Int 40:323-331. 16. Tashjian AH Jr, Levine L 1978 Epidermal growth factor stimulates prostaglandin production and bone resorption in cultured mouse calvariae. Biochem Biophys Res Commun 85:966-975. 17. Tashjian AH Jr., Hohmann EL, Antoniades HN, Levine L 1982 Platelet-derived growth factor stimulates bone resorption via a prostaglandin-mediated mechanism. Endocrinology 111:118-124. 18. Tashjian AH Jr, Voelkel EF, Lazzaro M, Singer FR, Roberts AB, Derynck R, Winkler ME, Levine L 1985 a And @ human transforming growth factors stimulate prostaglandin production and bone resorption in cultured mouse calvaria. Proc Natl Acad Sci USA 82:4535-3438. 19. Tashjian AH Jr, Voelkel EF, Lazzaro M , Goad D, Bosma T , Levine L 1987 Tumor necrosis factor-a (cachectin) stimulates bone resorption in mouse calvaria via a prostaglandin mediated mechanism. Endocrinology 120:2029-2036. 20. Heath JK, Saklatvala J , Meikle MC, Atkinson SJ, Reynolds J J 1985 Pig interleukin-I (catabolin) is a potent stimulator of bone resorption in vitro. Calcif Tissue Int 37:95-97. 21. Hattersley G , Chambers T J 1989 Calcitonin receptors as markers for osteoclastic differentiation: Correlation between generation of bone resorptive cells and cells that express calcitonin receptors in mouse bone marrow cultures. Endocrinology 125:1606-161 2. 22. Hattersley G , Chambers TJ 1989 Generation of osteoclastic function in mouse bone marrow cultures: Multinuclearity and tartrate-resistant acid phosphatase are unreliable markers for osteoclastic differentiation. Endocrinology 124: 1689- 1696.

PGE, INDUCES OSTEOCLASTS FROM HEMATOPOIETIC CELLS 23. Roberts RA, Spooncer E, Parkinson EK, Lord BI, Allen TD, Dexter TM 1987 Metabolically inactive 3T3 cells can substitute for marrow stromal cells to promote the proliferation and development of multipotent haemopoietic stem cells. J Cell Physiol 132:203-214. 24. Nicholson GC, Moseley JM, Sexton PM, Mendelsohn FAO, Martin TJ 1986 Abundant calcitonin receptors in isolated rat osteoclasts. J Clin Invest 78:355-360. 25. Hattersley G, Chambers TJ 1990 Effects of interleukin 3 and of granulocyte-macrophage and macrophage colony stimulating factors on osteoclast differentiation from mouse hemopoietic tissue. J Cell Physiol 142:201-209. 26. Kurland HI, Bockman R 1978 Prostaglandin E produced by human blood monocytes and mouse peritoneal macrophages. S Exp Med 147:952-957. 27. Felix R, Cecchini MG, Hofstetter W, Elford PR, Stutzer A, Fleisch H 1990 Impairment of macrophage colony-stimulating factor production and lack of resident bone marrow macrophages in the osteopetrotic op/op mouse. J Bone Miner Res 5:781-789. 28. Wiktor-Jedrzejczak W, Bartocci A, Ferrante AW Jr, Ahmed-Ansari A, Sell KW, Pollard SW, Stanley ER 1990 Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic ( o.p / o.p ) mouse. Proc Natl Acad Sci USA 87:4828-4832. 29. Yoshida H, Hayashi S-I, Kunisada T, Ogawa M, Nishikawa S, Okamura H, Sudo T, Shultz LD, Nishikawa S-I 1990 The

561

murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345:442-

444. 30. Felix R, Cecchini MG, Fleisch H 1990 Macrophage colony stimulating factor restores in vivo bone resorption in the op/ op osteopetrotic mouse. Endocrinology 127:2592-2594. 31. Kodama H, Yamasaki A, Nose M, Niida S, Ohgame Y , Abe M, Kumegawa M, Suda T 1991 Congenital osteoclast deficiency in osteopetrotic (op/op) mice is cured by injections of macrophage colony-stimulating factor. S Exp Med 173:269212. 32. Kurland JI, Pelus LM, Ralph P, Bockman RS, Moore MAS 1979 Induction of prostaglandin E synthesis in normal and neoplastic macrophages: Role for colony-stimulating factor(s) from effects on myeloid progenitor cell proliferation. Proc Natl Acad Sci USA 76:2326-2331.

Address reprint requests to: T.J. Chambers Department of Histopathology St. George's Hospital Medical School Cranmer Terrace London S W17 ORE, UK Received for publication July 31, 1991; in revised form December 17, 1991; accepted December 18, 1991.