Journal of Clinical InvestigationVol. 44, No. 1, 1965

Bone Resorption in Tissue Culture. Factors Influencingthe Response to Parathyroid Hormone *

LAWRENCEG. RAIsz t(From the Departments of Pharmacology and Medicine, University of Rochester School of

Medicine and Dentistry, Rochester, N. Y.)

Direct stimulation of bone resorption by para-thyroid tissue extracts has been demonstrated intransplantation experiments (2, 3) and in tissueculture (4, 5). Since bone from rats previouslygiven parathyroid extract could maintain a highercalcium concentration in the medium than controlswhen incubated in vitro (6), it seemed likely thatdirect stimulation of bone resorption in culturewould also result in a movement of calcium frombone to medium. The present studies employ atechnique for measuring such resorption in termsof the release of radioactive calcium previouslyincorporated in embryonic bone. This techniquehas provided a sensitive bioassay for parathyroidhormone preparations (1). However, manyother factors can affect bone resorption in organcultures. In the present experiments the effectsof varying culture conditions, particularly pH,calcium and phosphate concentrations, and oxy-gen tension, were studied. Sera from patientswith abnormalities of parathyroid function wereadded to the culture medium to determine whetherthese had any direct effects or might alter theresponse to added parathyroid hormone. Theeffects of vitamin A, cortisol, triiodothyronine,and vitamin D were also studied alone and incombination with parathyroid hormone, since

* Submitted for publication June 18, 1964; acceptedSeptember 24, 1964.

This investigation was supported by a Clinical Phar-macology Award from the Burroughs-Wellcome Founda-tion and research grant AM-06205 from the NationalInstitute of Arthritis and Metabolic Diseases, Bethesda,Md.

Presented in part to the American Federation forClinical Research, Atlantic City, N. J., April 29, 1962,and to the American Society for Pharmacology andExperimental Therapeutics, Fall Meeting, San Fran-cisco, Calif., August 11-15, 1963. A preliminary reportof part of this work has been published (1).

t Work begun while a Special Research Fellow of theU. S. Public Health Service at the Strangeways ResearchLaboratory, Cambridge, England.

these agents have previously been shown to af-fect bone in vivo (7-14), by local application (15,16), or in tissue culture (5, 17-21).

Methods 1

Preparation of tissue. Dating of pregnancy was as-sured by mating large numbers of Rochester Wistarrats for only 16 hours. Seventeen days later pregnantanimals were injected with 300 to 500 ,uc of Ca'5 as CaClasolution subcutaneously. On the nineteenth day, the in-jected rats were sacrificed, the embryos removed, andthe forelimbs separated and kept in Eagle's medium (22)in a 150-mm Petri dish at 5 to 10° C until dissection.With a binocular dissecting microscope, the radius andulna were dissected free of muscle and the cartilaginousends cut off. The bone shaft was washed three timeswith Eagle's medium before transfer to a culture ves-sel. The bone explant consisted of a thin tube of cal-cified bone matrix surrounding a core of calcified cartilagewith discs of mature cartilage cells at each end (Figure1). There was often beginning marrow cavity forma-tion in 19-day embryos, but little hematopoietic tissueremained after washing.

Culture technique. Small embryo watch glasses wereused, eight of which fitted into a 150-mm glass Petridish. The bones were placed on squares of Milliporefilter resting on a metal screen so cut that 0.5 ml ofmedium filled the culture vessel to the surface of thescreen. The medium was a mixture of 50% serum and50% modified Eagle's medium. The final calcium con-centration in this medium was 7 to 9 mg per 100 ml.When human serum was used, the final phosphate con-centration was 3 to 4 mg per 100 ml; when rat serumwas used, the final phosphate concentration was 5 to 8 mgper 100 ml.

The culture vessels were filled with medium andtest substances and equilibrated with 5% C02, 20% 02,and 75%o N2 at 370 C in a plastic chamber before thebones were explanted. The gas was bubbled throughdemineralized distilled water at the bottom of thechamber. The bottom of each Petri dish was linedwith filter paper moistened with Eagle's medium con-taining phenol red. The change in color of this mediumprovided evidence that the cultures had equilibrated withCO2. After explantation the cultures were maintained

1 A more detailed description of methods and ma-terials is available on request.

103

LAWRENCEG. RAISZ

TABLE I

Effect of parathyroid extract (PTE) and purified parathyroid hormone (PTH)on bone resorption in organ culture*

Treated/Control ratio

No. Ca45 release Medium Ca concentrationexperi-

Material Dose ments Mean ± SE Range Mean 4 SE Range

U/miPTE 1.0 16 2.34 ± 0.16 1.30-3.80 1.12 i 0.02 1.01-1.22

0.3 2 2.57 ± 0.07 2.50-2.64 1.11 I 0.02 1.09-1.13

0.1 16 1.67 i 0.06 1.28-2.10 1.05 ± 0.01 0.92-1.16

0.01 5 1.10 ± 0.04 1.02-1.22 0.97 41 0.03 0.87-1.04

PTH 0.3 6 2.76 i 0.05 2.54-2.97 1.14 i 0.01 1.08-1.17

0.1 6 2.09 4 0.21 1.50-2.65 1.07 i 0.03 0.99-1.18

* Mean :i1 standard error and range of values are given for all multiple-dose experiments in medium containingnormal human serum. In each experiment four to six pairs of bones were used at each dose level. Treated/control ratiosunderlined once are significantly greater than 1.0 at p < 0.05; underlined twice, at p < 0.01.

under continuous gas flow in the closed chamber for 72hours.

Analytic methods. At the end of cultivation the fluidwas removed, and stable and radioactive calcium con-

centration in the medium of each culture was measured.Phosphate concentration (23) was measured in poolsof four samples from each treatment and control group.

pH was measured in pools of four samples after they hadbeen re-equilibrated with 5% CO2 at 370 C. The boneswere either fixed in Zenker's solution for histologicstudy or put in 0.5 ml of 1 N HCl to dissolve the bonemineral and measure its radioactivity. Stable calciumconcentration was measured by Edathamil (EDTA)titration (24). Radioactive calcium was measured in an

automatic liquid scintillation counter (25). Countingefficiency was 50 to 60% and did not vary by more than5%o in a given experiment for both samples of culturemedium and HCl solution of bone.

Materials. 1) Parathyroid extract2 (PTE, 100 U per

ml) was added to the medium directly or after dilutionin a solution containing 1.6% glycerol, 0.2% phenol,and 0.9%o NaCl. The diluent solution alone was addedto the medium cultivating control bones. 2) Parathyroidhormone (PTH). A single sample of purified para-

thyroid hormone 3 with an estimated potency of 2,500U per mg was diluted to 40 g per ml or 100 U per mlin distilled water and added directly. Samples of par-

tially purified hormone from phenol extracts of bovineparathyroids (26) were also examined in bioassay ex-

periments. 3) Vitamin A. Vitamin A alcohol was dis-solved in ethyl alcohol and diluted 1:100 in culture me-

dium. Ethyl alcohol alone was added to control cultures.4) Cortisol. Hydrocortisone sodium succinate 4 was

2 Eli Lilly, Indianapolis, Ind.3 Kindly provided by Dr. G. Aurbach.4 Solu-Cortef, Upjohn, Kalamazoo, Mich.

3LE II

Percentage of total Ca45 initially in embryonic bone that was released into medium after 72hours incubation under various conditions

No. of Ca45 released*Experi- bone Treated/ment Treatment pairs Treated Control Control

9 Heat killed(1 hr at 48 C) 6 29 ± lt 0.50

PTE, 0.1 U/ml 2 86 ± 4 52 + 8 1.65PTE, 0.5 U/ml 2 94 4± 2t 63 + 1 1.49

14 PTE, 0.01 U/ml 3 47 + 12 36 + 4 1.31PTE, 0.1 U/ml 5 51 :1 4t 32 i 1 1.59PTE, 1.0 U/ml 5 94 + 3t 34 + 4 2.76

* Values are mean + standard error in this and all subsequent Tables.t Significantly different from nonpaired controls from the same litter (p < 0.01).t Significantly different from paired control bones cultivated in normal human serum (p < 0.01).

104

BONERESORPTIONIN TISSUE CULTURE

dissolved in water and diluted 1:100 in culture medium.No control vehicle was used. 5) Trijodothyronine wasdissolved in 0.1 M NaHCOsand diluted 1: 100 or 1: 200in culture medium. 0.1 M NaHCO3alone was used asa control solution. 6) Thyroid cultivation fluid. Piecesof rat thyroid were cultured by the technique previouslydescribed for rat parathyroids (27) in media of high orlow calcium concentration. The media were then re-constituted by the addition of equal volumes of fresh

b. s#;0<.- '1 mow}>x-ofhhhhIE3*4'

medium, and the final medium calcium concentration wasadjusted to between 7 and 8 mg per 100 ml. The effectof reconstituted medium was compared with the effectof fresh medium alone on paired bones. 7) Vitamin D.Vitamin D3 was added either dissolved in ethyl alcoholas for vitamin A, or as Vi-De3-Hydrosol,5 a water solu-ble preparation. No control solution was available forVi-De3-Hydrosol. 8) Serum. The standard culture

5 Obtained from Dr. A. Wander, Berne, Switzerland.

.a ,

F

-.~-7K

FIG. 1. EFFECTS OF PARATHYROIDEXTRACT (PTE) ANDVITAMIN A WITH ANDWITHOUTCORTISOL, AND SERUMFROM

VITAMIN D-DEFICIENT RATS ON BONE RESORPTION. A to D are cultured in media made with normal human serum.

Upper bone is treated. lower is paired control. Magnification X 70, hematoxylin and eosin stain. A) Experiment50: radius, 1.0 U per ml PTE vs. control. B) Experiment 50: radius, 1.0 U per ml PTE plus 200 Ag per ml cor-

tisol vs. control. C) Experiment 62: radius, 3 Ag per ml vitamin A vs. control. D) Experiment 62: radius, 3 jzgper ml vitamin A and 200 A.tg per ml cortisol vs. control. E) Experiment 38: radius, 1.0 U per ml PTE vs. controlin medium made with serum from vitamin D-deficient rats given 60,000 U vitamin D (Vi-Des-Hydrosol) intrave-nously 24 hours before bleeding. The response is similar to that obtained in normal rat serum. F) Experiment 38:radius, 1.0 U per ml PTE vs. control in medium made with serum from vitamin D-deficient rats.

C,ct, .,

E

105

LAWRENCEG. RAISZ

3 3.0-

IIZZ 2.0

Kz

.0-

O-PTEA-PTH* -TCA-P

0033 0.1DOSE-UNITS /ml.

FIG. 2. DOSE RESPONSEDATA FOR THREE PARATHYROID

PREPARATIONS IN EXPERIMENT 82. The dose scale islogarithmic and is based on the original estimate of po-

tency of each test material. Vertical lines show thestandard error. The reference standard was commercialparathyroid extract (PTE) assayed by the manufacturerat 100 U per ml. The response to purified parathyroidhormone (PTH), originally estimated to have a po-

tency of 2,500 U per mg, was analyzed by a three-dosefactorial assay (29) against PTE. The potency esti-mated by this assay was 2,950 U per mg with 95% con-

fidence limits of 1,825 to 6,400 U per mg. A sample oftrichloroacetic acid precipitate (TCA-P) from a par-

tially purified phenol extract of bovine parathyroid glandsoriginally estimated to have a potency of 100 U per mg

was similarly analyzed. The potency by this analysiswas 110 U per mg with 95% confidence limits of 57 to260 U per mg.

medium was made with 50% normal human serum. Thedirect effects and the response to PTE or PTH were

also tested in media made with serum from patientswith parathyroid disorders or with hypercalcemia due tocancer, serum from normal and parathyroidectomizedrats, and serum from rats maintained on vitamin D-de-ficient diets (14, 28) or given various doses of vitamin D.

Design and calculations. In each experiment 32 to 72bones were cultivated from one or two large litters ofeight or more embryos each. Four or more pairs ofbones were used for each treatment, one being treatedand the other serving as the control. Each treatmentwas applied in at least two separate experiments. Theulna contained and released more Ca' than the radiusin some experiments, but this difference was sufficientlysmall so that by alternating treatments, responses infour bones from the same embryo could be compared.The results are expressed as 1) Ca' release in cpm ofCa' per 0.1 ml of medium, 2) medium Ca concentrationin mg per 100 ml, 3) treated/control ratio of Ca'5 re-

lease or medium Ca concentration in paired treated andcontrol cultures, and 4) percentage of Ca'5 release cal-culated from Ca"5 released and Ca" remaining in bone

where the latter was measured. Statistical significancewas estimated with Student's t test. For PTE and PTHresponses the significance of treated/control ratios greaterthan 1.0 was based on a one-tailed distribution; other-wise a two-tailed test was used. Dose-response experi-ments were analyzed, with a factorial design for 2 X 3and 2 X 2 assays (29).

Results

Effect of parathyroid extract and parathyroidhormone. The addition of PTE or PTH con-sistently stimulated bone resorption as indicatedby the greater release of Ca45 from the bone to themedium in treated cultures (Table I). Resorp-tion was maximal with doses of 0.3 to 1.0 U perml. These bones showed disappearance of osteo-blasts, complete resorption of the matrix, andproliferation of fibrous tissue and osteoclasts, witha disc of hypertrophied cartilage cells remainingat either end of the explant (Figure 1A). Morethan 90%7 of the Ca45 originally incorporated wasreleased to the medium by the end of cultivation(Table II). Some resorption also occurred incontrol bones. Ca45 release from living controlbones was greater than for bones in which thecells had been killed by heating to 45 to 50° C.This release was associated with the appearanceof scattered osteoclasts, although there was littlefibroblastic proliferation, and osteoblasts were stillpresent in the bony shaft. Variability in controlresorption accounted for much of the variabilityin the treated/control ratios for Ca45 release inresponse to large doses of PTE or PTH (TableII; see also Table V).

Doses of 0.1 U per ml of PTE or PTH usuallycaused partial resorption. Responses to purifiedPTH were not significantly different from theresponses to equivalent doses of PTE. Minimaleffects were obtained with doses of 0.033 to 0.01U per ml. The use of the treated/control ratio ofCa45 release to assay the relative potency of dif-ferent parathyroid preparations is illustrated inFigure 2.

Medium Ca concentration was significantly in-creased with 0.1 U per ml or more of PTE andPTH. The effect was small compared with Ca45release, since the medium initially contained about40 pug of Ca and the bones contained less than 10jig of Ca, some of which was released in controlcultures. Phosphate concentrations in the mediumalso increased, and the molar ratio of Ca/P re-

106

BONERESORPTIONIN TISSUE CULTURE

co: IRP q 0 In)0 . 0

o o 66o o o6-H ,H-H -H 41 -H -N U) m 0o)N_l -

i Ili-." !" 4 1 4 4 -

- m656

-H -4

o) o

0 0

-H

N o N4

41 41 -H

6H 6 6HN 0 Nlco r c

0o t-

0o , 0o I

-H -HN 0

_o o

--4 40N 0

0 to In I- c 4- 0%

0 N4 -4 t ll6; c 6 6;

-H - H-H -H -H -H -H -H

m - 0 N4 'tt -

N4 1- Nq q N4 -4

-H

-4.

0%co0E

U)-H

CS

U)

kn) 0

-H -H

O N

;5 0s-4

0

mUU)

-H -H -H

40

N4 -I

-E. E.- -£.o . -

p-i X4 i i

- _0% - 0a

-H

W, Id0%

0 - 40

-H -H -H

40 - m

- N

_

P4i

40

N N

440 +

o NSN

c0o

40 _NN No0, NS

-H -H

06-H0q

0

-H

N

6

-

oo

0%

-H

UiC40

40

-H

0

o,

-

::) ~ U) ;466 6.0I I m >,

930=

( E; Fe CoH4 H4 H14

-0 -~~~~0

E u E E q E

z ¢ z Z z z z

0 0In co 0

107

00

._

d.2440

4)0044u

0

0

10

4)444)

H~

I

0

-H ,,-

O It

4- 40

tho o.-H

440V 0

o.Ig

X.

0t

4..)

4.)

(-4

o- b

4)Nm4-

(.r

4.,

4.)

0

4.

4)

4)vI

0

0

u

44co

Cd

0

r.

E

H

E

0

k6coV

4

0

0*3

0o"

0

v

VV0 *

P4)o d

_ _

U) 0

0 U

o o)Cd

w

Hcd5

* I-+4

'U 0 bG

ZaQ

LAWRENCEG. RAISZ

leased in response to PTE was 1.4 to 1.8, whichis similar to the ratio in bone mineral.

Bone size was a critical factor in parathyroidresponse. The data of these experiments are forbone shafts whose length was 0.9 to 1.3 mm.Smaller bones showed the histologic effects ofparathyroid hormone, but contained too little cal-cium for accurate assay. With the larger boneshafts of older embryos or of 19-day embryos fromsmall litters, ranging from 1.4 to 1.9 mmin length,addition of 1.0 U per ml of PTE usually causedonly 10 to 20% more Ca45 release than for pairedcontrols. PTE and PTH regularly increased themedium Ca concentration and produced typicalhistologic changes, but there was no extensive re-sorption of the matrix.

Effect of changing calcium and phosphate con-centrations in the medium. When medium phos-phate concentration was increased to 12 mg per100 ml or 4 mM, there was a proportional de-crease in Ca45 release from both control andtreated bones (Table III), such that treated/con-trol ratios were the same as for cultures in normalmedia. In experiment 84 the ratio of Ca45 re-lease in normal medium to Ca45 release in highphosphate medium for bones from the same em-bryo was 1.80 +.24 for control cultures, 1.70 ±.21 when 0.3 U per ml PTH was added, and1.53 +.16 when 0.1 U per ml PTH was added.These ratios are significantly greater than 1.0(p < 0.05) but not significantly different fromeach other. Histologically, both control andPTH-treated bones cultivated in high phosphatemedium were less cellular and had fewer osteo-clasts. When calcium concentration in the me-dium was varied over a range of 5 to 12 mg per

TABLE IV

Effect of serum from hypercalcemic patients with hyperpara-thyroidism (H) or bronchogenic cancer (C) compared with

normal human serum (N) or serum obtained after suc-cessful parathyroidectomy (P)*

Patient's Type ofserum serum No. of

Ca con- Treated/ bone Ca45 releasePatient centration Control pairs Treated/Control

mg/100 mlUh 15.6 H/N 4 0.65 + 0.05tUh 15.6/9.4 H/P 7 1.00 ± 0.08Mc 18.9 H/N 4 0.53 4 O.lOtDo 15.5 C/N 3 0.76 :1: 0.03t

* Medium Ca adjusted to 7 to 9 mg per 100 ml.t Significantly less than 1.0 at p < 0.01.

100 ml, the final medium Ca concentration re-flected the initial value, but Ca45 release and his-tologic changes in response to PTH were similarto those observed in normal media (Table III).WhenEDTAwas added in sufficient concentration(2 mM) to chelate all the calcium initially presentin the medium, release of calcium was increasedin the control bones, but there was no response toPTH. These bones showed extensive cell ne-crosis.

Effects of changing pH. PTE could stimulateCa release and cause typical histologic changesover a range of pH in the medium of 6.78 to 7.45.The pH of medium from treated and control cul-tures was the same within 0.05 pH U in six ofseven experiments. Medium Ca concentrationwas usually higher for both control and treatedcultures in media of low pH. When the pH ofthe medium was increased to 7.6 by the additionof bicarbonate (Table III), the effect of PTE onCa45 release was inhibited, and Ca45 release andmedium Ca concentrations of control cultures weredecreased. Histologically, the PTE-treated cul-tures showed only slight fibroblastic proliferationand few osteoclasts.

Effects of hyperoxia. The response to para-thyroid hormone was inhibited when bones werecultivated in an atmosphere of 95%o oxygen, 5%oCO2 (Table III). Control bones showed similarvalues for Ca45 release and medium Ca concen-tration whether cultured in 95% or 207o% oxygen.Histologically, these bones showed a responsesimilar to that seen in alkalosis.

Effects of different sera. Ten samples of serumfrom patients with hypercalcemia due to para-thyroid adenoma or cancer and four samples ofhypocalcemic serum from patients with hypo-parathyroidism were compared with normal hu-man serum in paired bone cultures. In six in-stances the pre- and postoperative sera from thesame patients were available for comparison. Se-rum from patients with hypercalcemia did notincrease bone resorption, nor did sera from pa-tients with hypocalcemia decrease bone resorptionin tissue culture. There were no significant dif-ferences when pre- and postoperative sera fromthe same patients were compared directly. Inthree instances, bones cultured in serum from pa-tients with marked hypercalcemia actually inhibitedCa45 release when compared with bones cultured

108

BONERESORPTIONIN TISSUE CULTURE

in normal human sera, but there was no differ-ence in the effect of pre- and postoperative serumfrom one of these patients (Table IV). o .f°

There was no difference in resorption between E 8paired bones cultured in media made with serum a 'from normal or parathyroidectomized rats, and " "the response to parathyroid extract was similarin both sera. The treated/control ratio for Ca45release in response to 1.0 U per ml of PTE was O1.8 +.08 in six experiments in which the medium U cwas made with normal rat serum. Although this H i 4)

ratio is not significantly different from that in Xnormal human serum, the degree of resorptionseen histologically in response to PTE was usu-ally less in rat serum (Figure 1E). Some sam-ples of human serum appeared to stimulate Ca45 0 + -

release in both PTE-treated and control bones b . c O.(Table V; see also experiment 9, Table II). Re-sults with serum from vitamin D-deficient ratsare presented below. , ;E :

-H + -HInteractions of parathyroid extract, vntamin A, vand hydrocortisone (Table VI). Vitamin A (3 $0to 10 lcg per ml) stimulated the release of Ca45from embryonic bone and increased medium Caconcentration, but these effects were not so great a 00

7

as those obtained with 1.0 U per ml of PTE. vlHistologically, vitamin A-treated bones differed 0 :

from those treated with parathyroid extract. 0sThere was less proliferation of fibroblasts and v

osteoclasts despite considerable loss of matrix co

(Figure 1C). When a submaximal dose of PTE Wd

(0.1 U per ml) was added to vitamin A-treatedcultures, more extensive resorption resulted, leav- ving only a small remnant of fibroblasts and ->aC6osteoclasts. U

The large dose of cortisol employed in these t .|||studies had no effect on Ca45 release from bone' -H -H -H *|Cor on medium Ca concentration. Histologically, 4) > . .there were less connective tissue growth and more a Zr3pyknosis of cartilage cells in these cultures. Addi- OT 44A)tion of cortisol completely inhibited the effects -a

4)~ ~ ~ ~ >

of vitamin A, and the appearance of the bone after 6 d Ecombined treatment was like that of treatment EE

o0with cortisol alone (Figure iD). Cortisol dimin- Z -

.

ished the response to 1 U per ml of PTE to some e E .Y;< < W

extent; however, the treated bones still showed Z@ = X X Or.stimulation of Ca45 release and the typical histo- S 8

4044logic response to PTE (Figure 1B). W tX co

Effects of triiodothyronine and thyroid culti- *vation fluid. The addition of 0.5 to 1.0 jug per

109

LAWRENCEG. RAISZ

\0000 0 0o 0 0 0 O O 0 00 0 0

-H -H -H -H-H 4 -H -Hl 4-HH -H00 C' C-4 e-4 0' 0 - C4U--

Cd

0C)-- - - -

-C)

CN \0 ci - c ) CN ci c- ~~~~~ 66~C5C 6656 6 6; 66 6-C0 0 0

H -H -H -H + H -H 4i 4 4H1 0

'4s 40 0 0 O 00 if) c en en eqi CY)

~~~~~~006 0 6 rz X. . .

$ CU NS o o O o O o o-- Y Csed d q t oo oe

u 4e CN '0 CI4 UI) CN ci -4 Ci4 mY c --,T C ;C 3C666666 6 66 63 0

~~~~~~~~~~~ ~- H -H - H H41lf f 6H-O 'e0 O-ON '0 O 00 0 U

Cid

b U + + + ~~~~~~~~~~~~~~~~++ ++ + +0 C'-4 00 - 0'- 1 '0 '0 t-

u u O OX~~~~~~CI o enO enX o O8 Y m. c. 0 Y ci c--U O 0 ^SF t~~~~~~~~~~~~~~~~0

0 6 66 6666C;C;C; C 6 66 6~Q 41 H- 1 f H4 41 -H 4i -H

¢8 0 0 m ' ci 00C 00 '0 0CU f ' 0' 'd 0 - if°U') 0 0e

- c--ic-U _- - ci

OC-wXbNoOovoeoO+u) 0

r' U- -'_ 00 - 0'd - CU

t -H -H-H -H-H -H --H -H-H/ t o~ 00 - U00oC00e 0 0 t c =

O~~~~~~~~~~)U' ci -

C)~~~~~~~~~~~~~~~~~~~~~~~~~~1cd 00 f) f)00 '0 ci - 0'

0 Ut U- ri 00 0 '0 - 00 It zV - 0' - -i4 -c-ON 00 0'0~UCU~~~~~~~~~~~~~~~~

z r E r E r E r Y ~~~~~~~~~~~~~CZH

00 Ci4 t-~ -I 00 00 c-- 0 c- en m)H o0 0 \0 UI) CN CI4 CI4 - 1 '0 en I=c

W-- El)C\EO E UE E * { E U0

4-CU

0C)vc:.0.3 "t 4 C4' ' 0 ' 0 '0OCd 0

4- CU~~~~~~~~~~~~~~~~~~~C-. o-- -o -CUJ CU4 Cd U CUd2

E

o E 04 MhS 4-'

0 -~~~~~~~~~~~ ~ ~ ~ ~ ~~CUC~~~~CM3

--)

C>~ ~ ~ CE - CU -oC

4-i >~~~~~~~-a- a 4 a4 > 0- 91 a.-o4

0 Ciif) '0 * -I---+ e-o =

il1110

illBONERESORPTIONIN TISSUE CULTURE

0

00t:o m m )*

o 6 6 0 0 0 0 0 0 00DU -H -H i H- -H-Hi -H -H -HiV 00SfI) WI) -- (4 mU) V 000(

42%0ON Q 0 Q0O 0 0' 0

44. C

X- N0-

. oCa 41-H -H -H 41 -H 414 -H

0

b~~~$ .0%'1 0 3-t'. 0% O0%

te .t-2o o o o oo o6 o6 o6 o6t.S ts X m~~~~0 > .+ +='CC

>~~~~~~~~~~~~~~~~c ct; 81 c; C; 8s c; c;m~~~>OUU) U) - -) + U)t Ve° -

-H-H-H-H4i4i -H-H -H 0. 00 0 UO1 014 J E

z~~~~~~~~~~~~~~t- rZ %--aS (2 ci o6 o6

wS d + + + + + + + + ++wb $ F 0> s4-b ci<l+ec

0~ ~ ~ ~ ~ ~ 0

-H 4i -41 41 -H 41 - -H -H003 e' V 4k 0%-d4\0 -0 C3

rF- 00 to 0 U) 0% 00 '- It

o,> t ~~~~~~~~~~~~~~~~~~~~~~~~~~~~No " M.,

e) - -- - - ,0 0H9l|- - to |

ell U)f e- t- U- ("%0 n b4~

Cb~ ~-H -H-H 4i- 1- H -l-00 t- CdF

U) U -U~4 C

-It qt-t-I- k

&0 00F-0C'd "l- t- ~4U

0 0 0

E --4 4~~~~~~~~~~~~~~~~ ~~~~~~~~(

4-oCd 0

bS. h.

LAWRENCEG. RAISZ

ml of triiodothyronine did not stimulate bone re-sorption directly, alter the response to 0.1 U perml of PTE, or cause any consistent histologicchange. The fluid that had previously been usedto cultivate thyroid glands also had no effect onbone resorption whether the original culture hadbeen in medium of high or low Ca concentration.

Effect of vitamin D and serum from vitaminD-deficient animals (Table VII). Large dosesof vitamin D in aqueous or alcoholic solution didnot stimulate bone resorption. Histologically,there were some pyknosis of nuclei and death ofcells with the larger doses of vitamin D em-ployed. When both vitamin D and PTE wereadded together, the typical parathyroid responsewas obtained, apparently unaltered by the presenceof vitamin D. Since the direct addition of vitaminD might not have provided a physiologically ac-tive material, the effects of serum from vitaminD-deficient, vitamin D-treated, and normal ratson bone cultures were examined. In five of nineexperiments, the addition of 1.0 U per ml ofPTE did not increase Ca45 release significantlyand caused little histologic change (Figure iF)when the medium was made with serum fromvitamin D-deficient rats. When serum from ani-mals treated with various doses of vitamin Dwhile on a deficient diet was used, the response toPTE increased significantly in four experiments.However, the PTE response was usually less thanthat observed in cultures made with serum fromnormal rats. Serum from vitamin D-deficientanimals given a massive dose of 60,000 U of vita-min D intravenously 24 hours before sacrifice didnot stimulate bone resorption directly, but didrestore the response to PTE (Figure lE). Intwo of the four experiments in which PTE didstimulate bone resorption significantly in vita-min D-deficient serum, there was diminished Ca45release in both control and PTE-treated cultures(for example, experiment 67, Table VII). Thedifference was similar to that observed with ahigh phosphate concentration, but in these ex-periments phosphate concentration remained be-tween 5 and 7 mg per 100 ml. When the datawere pooled, the mean treated/control ratios forCa45 release in response to 1 U per ml of PTEwere 1.47 ± 0.09 for vitamin D-deficient mediumand 1.81 + 0.07 for normal medium (p < 0.01).

Discussion

The ability of parathyroid extract or hormoneto stimulate bone resorption in tissue culture canbe quantitated and, under suitable conditions, pro-vides a sensitive bioassay for parathyroid activity.However, many other variables can affect boneresorption in control cultures and hence can af-fect the response to added PTE or PTH. Para-thyroid responses were observed in culture mediawith pH ranging from 6.8 to 7.4, calcium concen-trations ranging from 5 to 12 mg per 100 ml, andphosphate concentrations ranging from 3 to 12mg per 100 ml, although there was a decrease inthe amount of calcium released and in the degreeof histologic change in response to PTH in highphosphate medium. The pH of the medium atthe end of cultivation was the same for treatedand control cultures. The experiments withEDTA show that removal of calcium from thebone by chelation did not produce a typical para-thyroid response, but rather caused marked celldamage. This damage may be due to the lowionized calcium concentration or to chelation ofsome other essential ion. If parathyroid hormoneacted by producing acids at the surface of thebone with the resulting dissolution of mineral anda movement of calcium and phosphate along theirconcentration gradients into the extracellular fluid,one might expect the response to depend moreupon the calcium, phosphate, and particularly thehydrogen ion concentrations in the medium (30).It is likely that the cells at the surface of bonecan overcome to some extent limitations imposedby changes in the concentration of ions in thebathing medium.

A marked diminution in parathyroid effect wasobserved when the medium pH was above 7.6or when the cultures were gassed with 957o oxy-gen. These effects might be due to interferencewith the production or release of some metabolicacid required for parathyroid effect, to inactiva-tion of PTH, or to nonspecific toxicity. Gold-haber (5) showed that high oxygen tension couldstimulate bone resorption directly. This was notobserved in the present study, but only a singlehigh concentration of 02 was tested.

The sensitivity of embryonic bone to parathyroidhormone is sufficient to detect 0.01 U per ml PTEon bioassay. The failure to detect any increase

112

BONERESORPTIONIN TISSUE CULTURE

in bone resorption in media containing serum

from patients with hyperparathyroidism indicatesthat the concentration of biologically active hor-mone is probably less than 0.02 U per ml (20mUper ml). Serum from patients with severe

hypercalcemia actually inhibited bone resorptioncompared with normal human serum. Berson,Yalow, Aurbach, and Potts (31) have developeda sensitive radioimmunoassay technique that can

detect as little as 1.5 mUper ml of bovine para-

thyroid hormone. They found concentrations on

the order of 3.0 mU per ml in patients withhyperparathyroidism and were able to detect im-munologically active material in some, but not all,normal subjects. Tashjian, Levine, and Munson(32) found no parathyroid hormone activity inhuman serum using a complement fixation im-munoassay technique that can detect 20 to 35 mUper ml.

The sensitivity of the present technique isgreater than that of the usual rat assay (33) andrequires less hormone, but is less precise andmore time consuming. The index of precision(A) of tissue culture assays has been 0.28 to 0.34compared to 0.18 to 0.28 for rat assays and 0.08to 0.17 for immunoassay by complement fixation.Nevertheless, the method has been useful in de-tecting the small amounts of bone-resorbing ac-

tivity secreted by single rat parathyroids in tissueculture (27) and in assaying other preparationswhere only a small amount of material is available.

The effects of parathyroid extract on bone intissue cultures were compared with the effects ofvitamin A, cortisol, triiodothyronine, and vitaminD. Vitamin A in large doses was found to stimu-late bone resorption as measured by release ofpreviously incorporated Ca45 and by the increasein medium Ca concentration. Vitamin A did not

produce the histologic picture characteristic of theresponse to PTE; moreover, a further effect couldbe produced by adding PTE. Bone explantsnearly disappeared when treated with 3 ug per mlof vitamin A plus 0.1 U per ml of PTE. Theeffect of vitamin A, but not that of PTE, couldbe completely inhibited by the addition of cortisolas previously reported by Fell and Thomas (18).The studies of Lucy, Dingle, and Fell (34) sug-

gest that the resorption of cartilage matrix in thepresence of vitamin A is the result of the releaseof proteolytic enzymes in the tissue. Dingle (35)

has shown that vitamin A can cause direct re-lease of proteolytic enzymes from a lysosomalpreparation of rat liver. Addition of PTE tosuch a preparation did not cause any release ofproteolytic enzyme (36). PTE does not causethe extensive resorption of cartilage that is seenwith vitamin A. On the other hand, vitamin Aintoxication usually does not lead to hypercal-cemia. These findings suggest that parathyroidhormone and vitamin A stimulate bone resorptionby quite different mechanisms. The possibilitythat a synergistic relationship between vitamin Aand PTE is of importance in physiologic or patho-logic bone resorption deserves further study.

In the present study, cortisol had no directeffect on bone resorption. Cortisol caused a par-tial inhibition of PTE response, but this mayrepresent a nonspecific toxic effect of the largedose employed. In whole animals, there is evi-dence that adrenal cortical steroids inhibit boneformation but have little effect on bone resorption(11). In rats, antagonism between the effects ofcortisone and PTE has been observed (10, 37, 38)particularly in renal toxicity; however, cortisoneonly occasionally reverses the hypercalcemia ofhyperparathyroidism in man (39).

In the present study, triiodothyronine did notaffect bone resorption directly or alter the responseto PTE. In hyperthyroidism, bone turnover isincreased and hypercalcemia may occur' (12).In tissue culture, triiodothyronine can cause ac-

celeration of differentiation in cartilaginous bonerudiments and has a variable effect on growth(19, 20). Gaillard (21) found that thyroxincould affect the cartilage and the osteoblasts incultures of 14- to 16-day mouse embryo radius,but did not cause an increase in osteoclasts or re-

sorption of bone matrix.Although the role of bone in the hypercalcemia

of vitamin D intoxication has been questioned(14), there is evidence for direct stimulation ofbone resorption by excessive amounts of vitaminD (5, 16). There is also evidence that the para-thyroid glands cannot maintain normal serumcalcium concentration in the absence of vitaminD. In vitamin D-deficient rats with hypocalcemiathere is little change in serum calcium concentra-tion in response to exogenous PTE (40, 41). Ap-parently increased endogenous hormone also failsto maintain serum calcium in vitamin D deficiency,

113

1LAWRENCEG. RAISZ

since hypocalcemia develops despite an increase inparathyroid activity (42).

In the present study, no direct effect of vitaminD on bone resorption was demonstrated. Somedependence of parathyroid hormone effect on vita-min D or its metabolic products was suggestedby the observation that the PTE response wasdecreased or absent in some experiments whenbones were cultured in media containing serumfrom vitamin D-deficient animals. The variableresults could reflect differences in the amount ofvitamin D in the embryos, which came from nor-mally fed mothers. Differences in PTE responsehave also been observed with different human andrat sera that could not be ascribed to differencesin vitamin D intake or in endogenous parathyroidactivity in the donor animals. Some samples ofhuman serum were found to stimulate bone re-sorption to a considerable degree without addedPTE.

In an attempt to identify the factors responsiblefor such resorption, serum fractions have beenseparated by gel filtration and assayed for bone-resorbing activity (43). Bone-resorbing activityhas been found in low molecular weight peptidefractions of normal serum and also in serum fromparathyroidectomized animals. Further studieson such materials may enable us to improve theprecision of bioassay in organ cultures of embry-onic bone by controlling their effects and to iden-tify factors other than parathyroid tissue that mayregulate bone resorption physiologically.

Summary

1. The addition of 0.01 to 1.0 U per ml ofparathyroid extract (PTE) or purified parathy-roid hormone (PTH) to the medium was foundto stimulate resorption in 72-hour cultures of boneshafts of radius and ulna from 19-day rat embryos.

2. Radioactive calcium injected into pregnantrats was incorporated in the embryonic bone. Re-sorption could then be measured in cultures asthe release of Ca45 and the effect of PTH or PTEtreatment as the ratio of Ca45 released from treatedbones to that released from paired controls(treated/control ratio).

3. Calcium concentration in the medium couldbe varied from 5 to 12 mg per 100 ml and pHfrom 6.8 to 7.4 without inhibiting parathyroid

response. In media of high phosphate concentra-tion (12 mg per 100 ml) absolute values for Carelease were decreased in both control and PTH-treated bones, but the treated/control ratio waslittle altered.

4. Hyperoxia (95 % °2, 5% CO2) or a highpH (7.60) markedly diminished the parathyroidresponse. A high pH also decreased Ca45 releasefrom control bones. The addition of 2 mMEdathamil (EDTA) increased Ca45 release fromcontrol bones and abolished the PTH effect; theseexplants were not well preserved.

5. Variations in resorption were encounteredwhen bones were cultivated in media made withdifferent samples of human and rat serum. Thesedifferences could not be ascribed to variations inparathyroid activity. There was no difference inbone resorption between paired bones cultured inmedia made from serum from patients with hyper-parathyroidism before and after parathyroidectomy.

6. Vitamin A (3 to 10 ug per ml) increasedrelease of Ca45 and stable calcium from embryonicbone. The effect was less than that of a maximaldose of PTE (1 U per ml). Addition of PTE(0.1 U per ml) to vitamin A further enhancedresorption.

7. A large dose of cortisol (200 MAg per ml) hadno apparent effect on cultures of embryonic bone.This dose inhibited the effect of vitamin A com-pletely but only partly inhibited the PTE effect.

8. Triiodothyronine (0.5 and 1.0 Mg per ml)had no effect on bone resorption directly or onthe response to PTE. Culture fluid from ratthyroids similarly had no effect on bone resorption.

9. The addition of vitamin D in alcoholic oraqueous solution or sera from animals treated withlarge doses of vitamin D did not stimulate boneresorption directly. The response to PTE wasdiminished when the culture medium was madewith the serum from vitamin D-deficient ratscompared with the culture media made with serafrom vitamin D-treated rats or rats on a normaldiet.

Acknowledgments

I wish to thank Dr. W. Y. W. Au and Dr. G. Eng-strom for providing some of the sera from vitaminD-deficient and vitamin D-treated animals, and Mr. J. E.O'Brien and Mrs. A. Doty for valuable technical as-sistance.

14

BONERESORPTIONIN TISSUE CULTURE

References1. Raisz, L. G. Stimulation of bone resorption by

parathyroid hormone in tissue culture. Nature(Lond.) 1963, 197, 1015.

2. Barnicot, N. A. The local action of the parathyroidand other tissues on bone in intracerebral grafts.J. Anat. (Lond.) 1948, 82, 233.

3. Chang, H. Y. Grafts of parathyroid and other tis-sues to bone. Anat. Rec. 1951, 111, 23.

4. Gaillard, P. Parathyroid and bone in tissue culturein The Parathyroids, R. T. Greep and R. V. Tal-mage, Eds. Springfield, Ill., Charles C Thomas,1961, p. 20.

5. Goldhaber, P. Some chemical factors influencingbone resorption in tissue culture in Mechanisms ofHard Tissue Destruction, R. F. Sognnaes, Ed.Washington, American Association for the Ad-vancement of Science (Publication no. 75), 1963,p. 609.

6. Raisz, L. G., W. Y. W. Au, and J. Tepperman. Effectof changes in parathyroid activity on bone metabo-lism in vitro. Endocrinology 1961, 68, 783.

7. Wolbach, S. B. Vitamin-A deficiency and excess inrelation to skeletal growth. J. Bone Jt Surg. 1947,29, 171.

8. Cohen, J., C. L. Maddock, and S. B. Wolbach. Theeffect of parathyroidectomy on the evolution ofhypervitaminosis A in rats. Arch. Path. 1955, 59,723.

9. Irving, J. T. The effects of avitaminosis and hyper-vitaminosis A upon the incisor teeth and incisalalveolar bone of rats. J. Physiol. (Lond.) 1949,108, 92.

10. Laron, Z., J. P. Muhlethaler, and R. Klein. Theinterrelationship between cortisone and parathyroidextract in rats. Arch. Path. 1958, 65, 125.

11. Clark, I., R. F. Geoffroy, and W. Bowers. Effectsof adrenal cortical steroids on calcium metabolism.Endocrinology 1959, 64, 849.

12. Krane, S. M., G. L. Brownell, J. B. Stanbury, andH. Corrigan. The effect of thyroid disease oncalcium metabolism in man. J. clin. Invest. 1956,35, 874.

13. Aubert, J.-P., A. G. Cherian, M. S. Moukhtar, andG. Milhaud. Rtude du metabolism du calciumchez le rat a l'aide de calcium 45. V. La thyro-parathyroidectomie et l'effet de la thyroxine et dela parathormone. Biochem. Pharmacol. 1964, 13,31.

14. Harris, L. J. Vitamin D and bone in The Biochem-istry and Physiology of Bone, G. H. Bourne, Ed.New York, Academic Press, 1956, p. 590.

15. Barnicot, N. A. The local action of vitamin A onbone. J. Anat. (Lond.) 1950, 84, 374.

16. Barnicot, N. A. The local action of calciferol andoestradiol on bone. J. Anat. (Lond.) 1951, 85,120.

17. Fell, H. B., and E. Mellanby. The effect of hyper-vitaminosis A on embryonic limb-bones cultivatedin vitro. J. Physiol. (Lond.) 1952, 116, 320.

18. Fell, H. B., and L. Thomas. The influence of hy-drocortisone on the action of excess vitamin A onlimb bone rudiments in culture. J. exp. Med. 1961,114, 343.

19. Fell, H. B., and E. Mellanby. The effect of L-triiodo-thyronine on the growth and development of em-bryonic chick limb-bones in tissue culture. J.Physiol. (Lond.) 1956, 133, 89.

20. Lawson, K. The differential growth response ofembryonic chick limb-bone rudiments to triiodo-thyronine in vitro. 1. Stage of development andorgan size. J. Embryol. exp. Morph. 1961, 9, 42.

21. Gaillard, P. J. Observations on the effect of thyroidand parathyroid secretions on explanted mouseradius rudiments. Develop. Biol. 1963, 7, 103.

22. Eagle, H. Nutrition needs of mammalian cells intissue culture. Science 1955, 122, 501.

23. Chen, P. S., Jr., T. Y. Toribara, and H. Warner.Microdetermination of phosphorus. Analyt. Chem.1956, 28, 1756.

24. Beale, R. N., and J. 0. Bostrom. Sensitive methodsfor the titrimetric microdetermination of biologi-cal calcium and magnesium. J. clin. Path. 1963,16, 252.

25. Chen, P. S., Jr. Liquid scintillation counting of C1'and H3 in plasma and serum. Proc. Soc. exp.biol. (N. Y.) 1958, 98, 546.

26. Aurbach, G. D. Isolation of parathyroid hormoneafter extraction with phenol. J. biol. Chem. 1959,234, 3179.

27. Raisz, L. G. Regulation by calcium of parathyroidgrowth and secretion in vitro. Nature (Lond.)1963, 197, 1115.

28. Steenbock, H., and D. C. Herting. Vitamin D andgrowth. J. Nutr. 1955, 57, 449.

29. Bliss, C. I. The Statistics of Bioassay. New York,Academic Press, 1952.

30. Schartum, S., and G. Nichols, Jr. Concerning pHgradients between the extracellular compartmentand fluids bathing the bone mineral surface andtheir relation to calcium ion distribution. J. clin.Invest. 1962, 41, 1163.

31. Berson, S. A., R. S. Yalow, G. D. Aurbach, andJ. R. Potts, Jr. Immunoassay of bovine and hu-man parathyroid hormone. Proc. nat. Acad. Sci.(Wash.) 1963, 49, 613.

32. Tashjian, A. H., Jr., L. Levine, and P. L. Munson.Immunoassay of parathyroid hormone by quan-titative complement fixation. Endocrinology 1964,74, 244.

33. Munson, P. L. Biological assay of parathyroidhormone in The Parathyroids, R. 0. Greep andR. V. Talmage, Eds. Springfield, Ill., Charles CThomas, 1961, p. 94.

34. Lucy, J. A., J. T. Dingle, and H. B. Fell. Studieson the mode of action of excess of vitamin A. 2.A possible role of intracellular proteases in the

115

LAWRENCEG. RAISZ

degradation of cartilage matrix. Biochem. J. 1961,79, 500.

35. Dingle, J. T. Studies on the mode of action of ex-

cess of vitamin A. 3. Release of a bound pro-tease by the action of vitamin A. Biochem. J.1961, 79, 509.

36. Dingle, J. T., and L. G. Raisz. Unpublished ob-servations.

37. Bradford, R. H., R. P. Howard, W. Joel, andM. R. Shetlar. Antagonistic effects of para-

thyroid extract and cortisone. Arch. Path. 1960,69, 382.

38. Stoerk, H. C., A. C. Peterson, and V. C. Jelinek.The blood calcium lowering effect of hydrocorti-sone in parathyroidectomized rats. Proc. Soc.exp. Biol. (N. Y.) 1963, 114, 690.

39. Dent, C. E. Some problems of hyperparathyroidism.Brit. Med. J. 1962, 2, 1419, 1495.

40. Harrison, H. C., H. E. Harrison, and E. A. Park.Vitamin D and citrate metabolism; effect of vita-min D in rats fed diets adequate in both calciumand phosphorus. Amer. J. Physiol. 1958, 192, 432.

41. Rasmussen, H., H. DeLuca, C. Arnaud, C. Hawker,and M. von Stedingk. The relationship betweenvitamin D and parathyroid hormone. J. din. In-vest. 1963, 42, 1940.

42. Au, W. Y. W., and L. G. Raisz. Parathyroid ac-

tivity in vitamin D-deficient rats (abstract). Clin.Res. 1964, 12, 261.

43. Stern, P. H., and L. G. Raisz. Unpublished ob-servations.

116