JOURNAL OF BONE AND MINERAL RESEARCH Volume 5, Number 7, 1990 Mary Ann Liebert, Inc., Publishers

Effects of Parathyroid Hormone, Calcitonin, and Dibutyryl-Cyclic AMP on Osteoclast Area in Cultured Chick Tibia SUDHA PANDALAI' and CAROL V. GAY','

ABSTRACT Slices of osteoclast-enriched endosteal surfaces of 3 week chick tibia were cultured for 1-3 days. Osteoclasts on the bone surface were made visible by acridine orange fluorescence. Osteoclast area was measured by image analysis. Parathyroid hormone (PTH) caused osteoclasts to increase in area about 40%, and calcitonin (CT) caused a decrease in area also of about 40%. Subsequent addition of dibutyryl cyclic AMP to PTH- or to CT-treated cells resulted in a further change of 40 and 30%, respectively. Application of the cyclic AMP analog alone had no effect. All responses were rapid, occurring in 2-4 minutes.

INTRODUCTION the major cellular agent of bone resorption during remodelling. It is well known that their activity is affected by the major calciumregulating hormones, calcitonin (CT) and parathyroid hormone (PTH). Less well defined is the mechanism by which osteoclasts resorb bone, although the picture beginning to emerge is that hydrogen ions are secreted by an ATPdriven proton pump located at the ruffled border. The protons are believed to be provided to the pump by carbonic anhydrase, which is also present at the ruffled

0

STEOCLASTS ARE CONSIDERED

Recent research has been aimed at understanding the action of CT and PTH in halting and triggering osteoclast activity, respectively. Cyclic adenosine monophosphate (CAMP), a mobile second messenger of many hormones, has been shown to serve in the action of PTH on osteoblasts in several species, including rats and chickens. ( 8 - 1 0 1 Rat osteoclasts have been shown to produce cAMP upon treatment with CT under certain conditions.'") Initial studies of avian osteoclasts, however, showed no change in cAMP levels when treated with CT.(9,11,12) However, when under the influence of a low calcium diet or a vitamin D-deficient diet, a cAMP response in chick osteoclasts has been r e p ~ r t e d . " ~ . ' By ~ ) histochemical means, adenylate cyclase was not demonstrable in osteoclasts from chicks on

a normal calcium diet(151 but could be shown under conditions of low calcium intake.(I6)Dibutyryl-CAMP has been shown to decrease cytoplasmic spreading of isolated rat osteoclasts whether or not osteoblasts were present.(''] It was also concluded that the rounding caused by CAMP, in the absence of osteoblasts, paralleled the osteoclastic inhibitory action of CT. This study focuses on the changes in the area of chick osteoclasts on the endosteal bone surface when treated with PTH, CT, and CAMP. The experimental design takes advantage of studying osteoclasts in a situation close to their natural environment. Visualization of the cells was based on staining with acridine orange, a relatively nontoxic dye that has orange-red fluorescence when protonated. Rapid changes in cell area following PTH and calcitonin were detected; cAMP enhanced the effects of both hormones but was ineffective alone.

MATERIALS AND METHODS

Reagents Salmon calcitonin and dibutyryl-CAMP (dbcAMP) were obtained from Sigma (St. Louis, MO). Bovine parathyroid hormone was provided by the National Hormone and Pituitary Program. All other chemicals were of reagent grade and were obtained through common commercial sources.

!Department of Molecular and Cell Biology, The Pennsylvania State University, University Park, PA 16802. 'Department of Poultry Science, The Pennsylvania State University, University Park, PA 16802.

701

702

Sample - -preparation The tibiae of 3-week-old chicks (Hubbard-Hubbard or Hubbard-Peterson commercial broiler strains raised by the Pennsylvania State University Poultry Farms) were removed under sterile conditions. The chicks had been fed chick starter for the first week, then a low-calcium diet (0.3%) for the remaining 2 weeks. Sacrifice was by decapitation. The tibiae were split longitudinally, placed in Tyrode’s buffer (GIBCO, Long Island, NY), and the marrow removed to expose the endosteal surface, as described previously.(7)The bones were placed in fresh Tyrode’s buffer and then transferred to Eagle’s minimum essential medium with Earle’s salts and L-glutamine (MEM), 1% penicillin-streptomycin (GIBCO), and 10% fetal bovine serum and maintained at 5% C 0 2 , 37°C. Samples were obtained by removing slices 0.5-1 mm thick with a surface area of approximately 4 mm2 from the endosteal surface. Approximately 50% of the surface was covered by osteoclasts. After 10-12 slices were removed from each longitudinal half of tibia, slices were maintained in culture for up to 3 days; osteoclasts remained viable as judged by acridine orange uptake and by responses to hormones.

Hormone treatment and staining procedure All samples were first stained in acridine orange (AO; 1 mg/ml) in MEM for 15 minutes. In initial studies samples were treated with PTH (2.5 pg/ml) or CT (1 mU/ml) for 15-30 minutes (Table 1, experiments l a and 3a). Controls were obtained from the adjacent sections of the same bones. Image analysis was performed on the samples placed in a conventional welled slide. A total of 30 cells, 15 experimental and 15 control, were measured for each experiment. The largest cells stained by acridine orange were selected for analysis. These cells fell in the top 10% of the cell size range. For the remainder of experiments listed in Table 1, a Teflon slide fashioned by the Penn State Physics Shop (Fig. 1) was used. This allowed sequential image analysis to be performed on the same bone slice. Samples were stained with A 0 in MEM for 15 minutes and then placed in the well of the Teflon slide, exposed to the first solution, as listed in Table 1 for 2-4 minutes; osteoclast area measurements were made. The first solution was flushed away with MEM and replaced by a second solution (Table l), and after 2-4 minutes the areas of the osteoclasts were measured again. The first solution was either MEM with no hormone added (experiments 1, 3, and 5 ) or with PTH (experiment 2) or calcitonin (experiment 4) added. The second solution was MEM plus PTH (experiment l), PTH + dbcAMP (experiment 2), CT (experiment 3), CT + dbcAMP (experiment 4), or dbcAMP (experiment 5). Acridine orange (1 mg/ml) was present in all treatment and control solutions. Solutions were added through 1 mm ID tubing on the right side of the slide. Solutions were removed by Pasteur pipette from the channel on the left side of this slide (Fig. 1). Solutions were moved gently so as not to disturb the bone samples, which were glued to the slide with rapid-drying cyanoacrylate cement. Again, cells in the

PANDALAI AND GAY top 10% of the size range were selected for analysis. To reduce the time the cells were on the microscope stage only two to three cells per bone slice were evaluated. Student’s t-test was used to test for significant differences between experimentals and controls. The image analysis system consisted of Bioquant System IV software and hardware (R & M Biometrics, Nashville, TN). Samples were viewed by light microscopy using epifluorescence (magnification = x 125). Filters employed were a neutral density filter (Leitz, N4), a yellow GG455, and a Leitz H2 filter cube (excitation filter, 390-490 nm; beam splitting mirror, 510 nm; and barrier filter, 515 nm). A video camera attached to the microscope fed the image to the monitor, and area measurements were made using a digitizing pad and movable cursor.

RESULTS Figure 2 shows typical acridine orange-stained osteoclasts on the endosteal surface of chick tibia. Osteoclasts retained a stable size and shape, took up acridine orange, and responded to hormone and second messenger treatment at similar rates over at least a 3 day culture period. The cells ranged from 50 to 80 pm in diameter among birds, and cell areas ranged from about 2000 to 5000 pm2. However, the variation of osteoclast area within each bird was only 7.2-13.2%, since only the largest cells were selected for analysis (Table 1). All cells examined were multinucleate. As shown in Table 1, osteoclasts treated with PTH had an average of 40% greater surface area than untreated cells (experiment 1) whereas calcitonin had the opposite effect (experiment 3). Addition of dbcAMP increased the surface area of cells already exposed to PTH by an additional 38% (experiment 2). The addition of dbcAMP to calcitonintreated osteoclasts further decreased the surface area by an additional 30% (experiment 4). When osteoclasts were treated with dbcAMP alone, no differences in cell areas were found (experiment 5 ) .

DISCUSSION A method was developed to study shape changes in osteoclasts as they remained attached to the endosteal surface of bone. The cells were identified as osteoclasts on the basis of acridine orange uptake, large size, multinuclearity, apposition to the bone surface, morphologic similarity to isolated osteoclasts, which contain carbonic anhydrase, and response to PTH and CT.(7.18,19) The approach allowed observation of osteoclasts in an environment close to in vivo conditions. By using acridine orange as an epifluorescent dye to delineate the shape of osteoclasts, the effects of PTH and calcitonin on osteoclast surface area could be determined. Acridine orange staining of cytoplasmic vacuoles provided an adequate means of visualizing osteoclasts, as others have shown.(6.7)However, it is clear that the total volume of the cell is not being visualized. A

+

f

f

218-2098 f 172** 655-3384 f 597*

2303 + 218-2105 f 223 1958 f 192-1859 f 189

2882 4970 0.91 0.95

0.72 0.68

0.51 0.54 0.82

1.44 1.70

1.24 1.85 1.12

-

0

30

-58

+38

+40

A verage Yo change in area

=

-

0

16

-22

+ 19

+ 18

Average Yo change in diameter

15). Treatments: 2.5 pg/ml of

-

54.2- 5 1.8 49.9 48.7

60.6- 51.7 79.5 -65.6

-

70.0-50.1 58.7 43.3 5 1.3-45.1

49.9-60.0 54.0-64.0

64.0-71.5 54.5-74.1 54.6-57.8

Average cell diameter (mm)b

aAreas are expressed as means f SEM. The * and ** indicate significance at p B 0.05 and 0.01, respectively. Walculated from average area. A total of 15 osteoclasts were measured for each experimental and each control value ( n PTH; 1 mU/ml of CT; M dbcAMP.

5b

5a

Untreated slices followed by dbcAMP

Calcitonin followed by CT dbcAMP 4a 4b

3848 f 378-1971 f 154** 2704 f 225-1471 f 158** 2065 f 196-1595 f 187

239-4015 f 363* 217-4313 f 302* 168-2629 f 197

Untreated slices followed by CT 3a 3b 3c

f

f

f

1959 f 191-2829 f 219** 2286 f 290-3214 f 227**

3223 2334 2339

Osteoclast areaa (pm2)

PTH followed by PTH 2a 2b

+ dbcAMP

Untreated slices followed by PTH treatment la lb lc

Experiment

Experimental/ control

TABLE1. CHANGES IN OSTEOCLAST AREAON THE ENDOSTEAL BONESURFACE FOLLOWING TREATMENT WITH PARATHYROID HORMONE AND CALCITONIN IN THE PRESENCE A N D ABSENCE OF DIBUTYRYL CYCLIC AMP

704

FIG. 1. Teflon slide for sequential treatment of bone slices and subsequent viewing by epifluorescence: (a) well for the specimen; 3 mm in diameter, 1 mm in height; (b) well for treatment solution (1 cm x 1 cm x 1 mm); (c) channel through which solution is removed and replaced, 0.5 cm in width and 2.5 cm in length from end to end; (d) stabilizing bar to hold tubing in place during injection of treatment solution. A coverslip is placed over the square well (b).

FIG. 2. Typical multinucleate osteoclasts (arrow) clustered on the endosteal surface of the tibia, stained with acridine orange. The large size of the cells is evident by comparison with mononuclear cells (arrowhead).

PANDALAI AND GAY

changes in area may be implemented by the cytoskeleton''') and/or by fusion with adjacent cells. Further study is needed to sort this out. It is important to note that whereas changes in area were up to 58%, the cell diameter only changed by around 20%, since area is a squared function (Table 1). Thus measurement of changes in cell area is easier to detect than measurement of cell diameter. Although cell area was quite variable among birds, cells selected for analysis were similar in size (f13% or less), making t-test comparisons possible. PTH is believed to utilize cAMP as a second messenger in triggering bone resorption indirectly through osteoblasts. Our results also indicate a role of cAMP as second messenger for PTH since the ratios of areas of cells treated by PTH versus PTH + dbcAMP showed enhanced spreading beyond the increase caused by PTH alone. The rapid action of PTH supports the concept that osteoblasts secrete a chemical signal to mediate rapid changes in osteoclast action, since cell secretory mechanisms are known to be rapid. The present study also employed osteoclasts from chickens raised on a low-calcium diet. The observation that osteoclasts treated with calcitonin + dbcAMP have a smaller area than when calcitonin alone is used is further evidence for a CAMP-mediated response to calcitonin in avian species. When bone slices were treated with dbcAMP alone, no changes in cell shape were observed. This may indicate that the osteoclasts were simultaneously receiving mixed signals, one through cyclic AMP-stimulated osteoblasts and one through direct action of dbcAMP on osteoclasts, and therefore could not respond. The enhanced response of both PTH and CT with dbcAMP, which act in opposing directions, may be explained by the action of the hormones on different cell types, that is, PTH on osteoblasts and CT on osteoclasts. It is of interest that all responses observed occurred within 2-4 minutes of hormone treatment. To our knowledge, this is the first report of such rapid responses to PTH. This study has demonstrated that both PTH and CT act in the cAMP second messenger system in avian osteoclasts and that responses to both hormones is rapid.

ACKNOWLEDGMENTS This work was supported by NIH grant DE 04345. We thank Nancy Kief and Virginia Gilman for able and much appreciated technical assistance.

large central region is detected that changes in shape as the osteoclast changes shape; therefore, measuring the surface area of acridine orange-stained osteoclasts approximates the extent of the cell apposed to the bone surface. Changes in surface area due to spreading out or rounding correlated well with changes in osteoclast activity and support current views of the behavior of osteoclasts in the presence of these hormones, namely that PTH activates osteoclasts and CT inactivates them. Specifically, we found that activated osteoclasts cover considerably more bone surface than nonactivated cells. The mechanism of

REFERENCES Akisaka T, Gay CV 1986 Ultracytochemical evidence for a proton-pump adenosine triphosphatase in chick osteoclasts. Cell Tissue Res 245:507-512. 2. Baron R, Neff L, Louvard D, Courtoy PJ 1985 Cell-mediated extracellular acidification and bone resorption: Evidence for a low pH in resorbing lacunae and localization of a 100-kDlysosomal membrane protein at the osteoclast ruffled 1.

border. J Cell Biol 101:2210-2222.

CHANGES IN OSTEOCLAST AREA 3. Blair HC, Teitelbaum SL, Ghiselli R, Cluck S 1989 Osteoclastic bone resorption by a polarized vacuolar proton pump. Science 245855-857. 4. Bekker PJ, Gay CV 1990 Biochemical characterization of an electrogenic vacuolar proton pump in purified chicken osteoclast plasma membrane vesicles. J Bone Min Res 5569-579. 5 . Anderson RE, Schraer H, Gay CV 1982 Ultrastructural immunocytochemical localization of carbonic anhydrase in normal and calcitonin-treated chick osteoclasts. Anat Rec 204: 9-20. 6. Anderson RE, Woodbury DM, Jee WSS 1986 Humoral and ionic regulation of osteoclast acidity. Calcif Tissue Int 39: 252-258. 7. Hunter SJ, Schraer H, Gay CV 1988 Characterization of isolated and cultured chick osteoclasts: The effects of acetazolamide, calcitonin and parathyroid hormone on acid production. J Bone Min Res 3:297-303. 8. Wong GL 1984 A comparison of the PTH-dependent cAMP responses in osteoclastic and osteoblastic bone cells. Min Electrolyte Metab 10:77-83. 9. Ito MB, Schraer H, Gay CV 1985 The effects of calcitonin, parathyroid hormone and prostaglandin E, on cyclic AMP levels of isolated osteoclasts. Comp Biochem Physiol 81A: 653-657. 10. Hall GE, Kenny AD 1986 Bone resorption induced by parathyroid hormone and dibutyryl cyclic AMP: Role of carbonic anhydrase. J Pharmacol Exp Ther 238:778-782. 11. Nicholson GC, Livesey SA, Moseley JM, Martin TJ 1986 Actions of calcitonin, parathyroid hormone, and prostaglandin E, on cyclic AMP formation in chick and rat osteoclasts. J Cell Biochem 31:229-241. 12. Miyaura C, Nagata N, Suda T 1981 Failure to demonstrate the stimulative effect of calcitonin on cyclic AMP accumulation in avian bone in vitro. Endocrinol Jpn 28:403-408.

705 13. Eliam MC, Bas16 M, Bouziar B, Bielakoff J, Moukhtar M, devernejoul MC 1988 Influence of blood calcium on calcitonin receptors in isolated chick osteoclasts. J Endocrinol 119243-248. 14. Rifkin BR, Auszmann JM, Kleckner AP, Vernillo AT, Fine AS 1988 Calcitonin stimulates cAMP accumulation in chicken osteoclasts. Life Sci 42:799-804. 15. Yamamoto T, Fukushima 0, Gay CV 1989 Ultrastructural and cytochemical localization of adenylate cyclase in chicken bone cells. J Histochem Cytochem 37:1705-1709. 16. Fukushima 0, Gay CV 1988 Ultrastructural localization of adenylate cyclase and guanylate cyclase in bone cells. J Cell Biol 107:494a. 17. Chambers TJ, Fuller K, Athanasou NA 1984 The effect of prostaglandins I,, E,, E,, and dibutyryl cyclic AMP on the cytoplasmic spreading of rat osteoclasts. Br J Exp Pathol65: 557-566. 18. Gay CV, Ito MB, Schraer H 1985 Carbonic anhydrase in isolated osteoclasts. Metabolic Bone Dis Relat Res 5:33-39. 19. Hunter SJ, Schraer H, Gay CV 1989 Characterization of the cytoskeleton of isolated chick osteoclasts: Effects of calcitonin. J Histochem Cytochem. 37:1529-1537. 20. Vaes G 1988 Cellular biology and biochemical mechanism of bone resorption. Clin Orthop Relat Res 23k239-271.

Address reprint requests to: Dr. Carol V. Gay 468A North Frear Laboratory The Pennsylvania State University University Park, PA 16802 Received for publication July 21, 1989; in revised form December 18, 1989; accepted February 10, 1990.