- Email: [email protected]
A suitable culture medium for ossification of embryonic chick femur in organ culture
89
A suitable culture medium for ossification of embryonic chick femur in organ culture
Toshiyuki Kaji, Rie Kawatani, Takumi Hoshino, Tatsuro Miyahara, Hiroshi Kozuka, Youichi Kurzshige’ and Fumitomo Koizumi’ Se&n
of Toxicology and ‘Departmenr of Pnfhology, Faculrv OJ Pharmacwicol
Sciences and
‘.Me.?icme, T’oyoma Medical and Phormoceurical Omversiry. 1630 Sugirani. Toyamo-rhr. Toyamo. 530-01. JaPon (Received 22 June 19SR) (Accepted 17January 1990)
To establish a culture medium which altnws owfication murswerecultured
invari,~u~lyrupplemented
in organ c&u:;.
Changes of Ca and P, concentrations in the BGJb-HW2 extract (CEE)
g-day-old embryonic chick fe-
BGJb-HW,media. medium or the LO% addition of chick embryo
did “(11 induce ossification. Furthermore. combinations
of the
10% CEE with ahigh Ca x
P, product or with 5 mM &glycemphorphare f&G?) or with LO% horse serum plus a high Ca x P, prodyc1 often caused pathological abnormalities in the pedonc..... On the other hand, BGJb-HW,
medium supplemented with 5 mM,&GP
indwddevetopmenr
ofoti-
fication. The Ca content of femurs and the diaphysial hydmxypmline ccntent were markedly increased. Histological obsrarion
showed a formation at a thick and active periosteum, numerom osteobtastic
cells, a sufficient amount of o ,teaid tissue and well dewlopzd calcified trabeculae without anypathotqcai changes. Thus, tb,c organ culture system using this medium was considered to be an appropriate one for studies an osteogsncsir ip. vile.
Key words: Bone formation: Calcification: Differentiahnn:
Fmb~onicchickbone:Tissue
culture:
Ossification
Introduction Organ culture systems have been used extensively for studying factors of bone resorption [1,2]. However, organ culture systems have not been widely applied to
016%6w91~~003.50
0 1990 Elscvier Sacnce Pub&hers B.V. (Bmmedral
Division)
90 studies on bone formation, particularly calcification. As pointed out by Gerstenfeld et al. [3], several factors appear to enhance calcification in cell culture systems. These factors include: (1) a highly enriched nutrient medium [4,5]: (2) the time period that cells are kept in culture [6]; (3) the use of ,%glycerophosphate @GP) or other organic phosphates [7-g]. This infurmation is suggestive for establishment of an organ culture system in which ossification occurs sufficiently. Such a system will be useful for studies on bone formation. Previously, a chemically defined cultxe m&urn BGJb-HW, [lo] was used in our studies [!I]. This medium was established by Endo et al. to induce active collagen synthesis of cultured bone on the basis of Bigger’s BGJb medium [12]. Although BGJb-HW* medium does not contain eirher serum or proteins such aa physiological growth factors, it is excellent for studying collagen metabolism, especially for that of chondrocytes. However, the medium did not induce a sufficient ossification, particularly calcification. Desiring ossification, modifications were made using calcium (Ca), inorganic phosphate (Pi), p-GP, chick embryo extract (CEE) and horse seT”rn (US). In the pxsent study, we have investigated in vitro ossification of embryonic chick femur in the modified BGJb-HW, media and discussed the effect of each supplement on ossification. Materials and Methods
Fertilized eggs from White Leghorn hens were incubated for 9 days in an electoric incubator and a whole femur was resected from an embryonic chicken under sterile conditions. The femur was added to 1.5 ml of the culture medium in a Leightontype culture tube. The tube was incubated at 38 ‘C in a rotating culture drum apparatus; the medium was changed every 2 days. The basal medium was BGlb-HW, [ll] (Table 1) which contained 1.26 mM Ca and 1.52 mM P,, while modifications were carried uut to examine their effect OR ossification. The employed modifications are shown in Fig. 1 and Table 2. Preporotion of chick embryo extract (CEE) Eleven-day-old chick embryos were crushed in a mechanical homogenizer under sterile conditions and the homogenate was centrifuged at 10 000 ‘pm for 15 min. The supematant fluid was frozen and thawed 7 times and centiifiiged at 10 000 rpm for 15 min again to obtain the supernatant (CEE). This CEE contained 0.49 mM Ca and 11.5 mM P,.
HS was purchased from Nakarai Kagakuyakuhin Inc., Kyoto, Japan. The HS had already been inactivated and contained 2.56 mM Ca and 1.54 mM P,. Anulytical
methods
A whole femur was extracted for 2 days with cold 0.1 M acetate buffer contaiuing
91 Table 1 The composition of chemically defined BGJb-HW,
medium
established
by Endo
[ill L-LyaiaeHC! !_-HistidineHCI~H,O L-ArginineHCI L-Tbreonine kValine L-Leucine L-lroleucine r-Mcthionine L-Phenykdaninc LTlyp’0pba”e r-Tyrosine 1,CysteincHCl.H,O ~Glutamine Glycine L-Serine l.4klli”~ Nicodnamide Thiamine.HCl Ca-pantothenate Riboflavine
11
n-.,A,
_.,“...~~.
15.0 7.5
7.5 6.5 5.0 3.0 5.0 5.0 4.0 4.0 9.0 20.0 15.0 10.5 11.5 2.0 0.4
0.02
DIrrirlnr., -,.---
-or
“hnsrhalp
0 m rnjdl
Fohc acid
0.02
Biotin p-‘minobrwo.c acid Na-n-tocopher>Iphosphate Cbob.wcbloride m-lnoskll Cyanocoba!amine Na-arcorbae
0.02
N&i KCI CZCI,.ZFI,O MgSO,.7H,O Na,HPO,.12H,O KH,PO, NaHCO, F&I, Glucose Phenolred
0.2 0.1 5.0 3.82 0.CC.l 5.0 800.0 40.0 18.5 20.0 12.1 16.0 140.0 Cl.047 MO.0 2.0
0.02
0.01 M EDTA (pH 5.5). After extraction, the epiphysial bulge was cut off from the diaphysis and each part was analysed for hydroxyproline (Hyp) content by the method of Huszar et al. [13]. The extract was analysed for Ca content by atozk absorption spectrophotometry. Since one femur of the same embrj<* vile n~ltured and the other was uncultured, increase in Ca and Hyp content was expressed as the percentage of Ca and Hyp of the corresponding uncultured bone.
Histological technique Femurs were fiied in 10% neutral formalin. They were embedded in paraffin in a conventional way without decalcification. Femurs were sectioned serially to 6.um thickness. These sections were stained by the method of van Kossa counterstained with Weiger’s hematoxylin and ponceau-acid titchsin [14] or stained with Mayer’s hematoxylin and eosin.
Results on Ca and Hyp contents were analysed for statistical significance by Student’s t-test. P values of less than 0.05 were considered to indicate statistically significant differences.
92 Results Ef&f n,fCg avd P, concentrations on Ca ond Hyp confents in fenws First, Ca and P, concentratiLo& i;: BGJb-KWz medium were cha :ged. Pre!iminary studies showed ectopic calcification in the periosteum of femurs cultured in BGJbHW, medium containing Ca x Pi products above 6 mM*. Based on the result, Ca x P, products at 5.0 mM2 and below were examined in this experiment. Figure 1 shows the effects of Ca and Pi concentrations on Ca and Hyp content of the cultured femurs. Although no marked increase in Ca content occurred as a whole, the highest increase in Ca content was seen when Ca and Pi concentrations were 2.0 and 2.5 mM, respectively. No marked increase in Hyp content occurred in the diaphysis by changing Ca and Pi concentrations. In the epiphysis, a tendency toward greater Hyp content was observed when Ca X Pi products were between 4.40 and 4.84, while no significant change was recognized statistically. Effecf of several modifications using a high Co X P,product, t¶-GP, CEE and HS on Ca and Hyp content in femurs As stated above, changes of Ca and P, concentrations could not induce a sufficient increase in Ca and diaphysial Hyp content in femurs by themselves. Therefore, modifications were made using fl-GP, CEE, HS and high Ca x P, products (5.0 mM2 and above). The experimental design is summarized in Table 2. Since either CEE or HS contained a significant amount of Ca and Pi (see Materials and Methods), Ca and Pi concentrations of the modified media were not always equal in spite of the same basal medium. In other words, CEE and HS were evaluated as supplements which rnntain a significant amount of Ca and Pv Each modified medium was given a provisional name (from A to F). Lengths of femurs before culture were 5.01
Fig. 1. Effect of Ca and pi c”ncentrstionson the increasein Ca contentin the whole femur, hydroxyproline (Hyp) contentin the diaphysisand in the epiphysis.The femurswere c”lt”red fcr 6 days.Values are means+ SE for four sampleswith percentageof eachcontentof the unculturedfemur. The Ca comem when ca and P, mncemrationsare 2.0 and 2.50 rnM, respectively,is significantlymare than th”sewhen Ca and P, concentrationsarc Z.0 and 2.00 mM (P 4 O.WN), 2.8 snd 1.52 mM (P c 0.05). 2.2 and 2.20 mM (P c 0.05)and 2.5 and2.W mM (P < 0.01).ThediaphysialHypcontent when CaandP,conce”tmlions xe 2.0 and 2.50 mM, respectively,is significantlym”re than thosewhen Ca and P, concentrations
93
Table 2 Summary of modifications of chemically defined BGJb-HW, medium Medium
Basal medium
Ca conccn-
P, eoncen-
trhm
tration
@M) ~-~
HS
Day, in
(%)
Nh”X
(mM)
A
BGJSW’,
2.0
2.5
Nmc
None
None
5
B
BGIb-HWz
2.0
1.5
s
None
None
5
C
BG;b-HW,
1.3
2.7
None
10
None
5
D
BGJb-HW,
2.1
3.7
None
1”
N.mC
5
E
BGJb-HW,
z.,
2.7
5
111
None
5
F
BGJb-HW,
2.3
3.8
None
10
LO
5
k 0.01 mm and the longitudinal growth of femurs by cultivation with medium A, B, C,D,EorFwas4.4+0.2,3.9+0.1,2.1+0.4,2.8+0.3,2.1~0.4or1.6+0.1 mm, respectively. The increase in Ca and Hyp contents of femurs cultured in each mod&d mediuii; in respect to uncultured femurs are shown in Fig. 2. The Ca content in femurs of both medium A and B increased by about 5 times but the values varied wideiy for medium A. The Ca content for medium E and F increased by 3 times. There was no sufficient change in Ca content in both medium C and D. The diaphysis consists of an intraperiosteal bone and a cartilage core, so that the diaphysial Hyp content reflects both the intraperiosteal bone matrix formation by
Ffg. 2. Effect of
modifications of BGJb-HW,
medium using a high Ca x F, product. jLglyempbcqkte.
chick embryo extract and horse serum on the increase in Ca content in the whole femur and hydmqproline (Hyp) content in tile diaphysis and in the epipbysir. The employed mo.iiEcaziom are *own i. Table 2. F.%nYn were E”,Nred for 5 clays. “?.,ueS are means f SE for far rampIes with perccmage of each content the ““cl&“red femur. As regards Ca content, B is rignincantly higherthan C (P< ll.00,). D (P< 0.05) and F(P < 0.05): C is rigniiicanrly lower than D (P< 0.0,). E (Pc0.M) and F(P< 0.05).
of
As regards diaphyaial Hyp content, A is significantly higher than C (P 4 0.05), E (PC
0.05); B issigniti-
cantly higher than C (P < 0.01). D (P 4 0.05). E (P < 0.001) and F (P < 0.031). As regards epiphysial Hyp content, C is significantly lower than A (P < 0.05) and F(P<
0.05).
94
.
.,I
y’
:;
..’
A‘:
..: :-
_:,
:
. ; ‘.‘.’
.*.:.
.
:
.;
‘*
. 1’. ..\-.
: . . . _ ,. -
.: . . . . ‘i._.I ‘.._..:*
1
:;*.j
. : .,, _. .
.,‘..
p,‘?; _*._-
,
;‘.,.*
;’
,..:.a:,
_, ,: ..,,..‘,
:j.:
:::.
:
;
:,
.’
1:‘..:.!:‘:...-.‘::~;. .-.,“_, :‘.
‘>.I
.::
..:
Pip. 1 Miausupic findings of the central pan of the diaphysis of the iettttu cultured with modified Btilb-HWt media for 5 days. The employed modifications are shown in Table 2. A. A femur cultured with medium A: a slightly active proliferation mesenchymal cellsin athinperiateum(PO), a thin layer of osteoid tissue (OS) and a large catdried eabecula (T). van Kossalhematorylinlponceau-acid hahsin stain (x100). B. Afemurcutturedwithmedium B: an acdvepratiferstionofmesenchymdceUs(MC) in a thick PO, a thick layer OSon the inside oftheP0 and alarge T. van KossalhematoxytinlponMauacidfuchsinstain (xltltt). C. AfemurculturedwithmediumC: a hypertropbkchange PO,B thinlayer of OS and a small T. van Korsalhelnatoxylinlponceau-a*d fuchsin stain (X100). D. A femur ctdtured with medium D: a rhrci P3 irritutir~g a tendency of degeneration (D), a %anty OS and a slightly developed T. “cl” KoJsamematolty,in/ponccsu-afid fvchrin stain (XNW). E. A fEm”r cultured WtIb medium E: a very thick PD. a thick layer of OS and a small T. “0” KosrYhema~~ytilllponEesu-add fucbsi” slain (x IW,. P. A femur cultured with medun F: a d@ophicchaoge (arrowheads) and an ectopic calcification (arrows) in a thick PO and a slightly developed T. won K~ralhematorylinlponeeau-acid fuchsin
of
of
of
95
osteoblasts and cartilage formation by chondrocytes. Unfortunately, it is impossible to measure the bone and cartilage Hyp contents separately. In addition, the Iongitudinal growth by cartilage formation may contribute to the diaphysial Hyp content. However, the diaphysial Hyp content w&s measured nevertheless, since it may reflect bone matrix formation. The high.% increases in diaphysial Hyp content were recognized in femurs cultured with medium B. Medium A increased diaphysial Hyp content by about 3 times, while the others only did by about 2 times. In the epiphysis, which consists of only cartilage, medium A markedly increased Hyp ccmtent by about 4.5 times. Hisrological investigahon of ossificalionin femurs cultured with each modified BGJb-HW, medium Figure 3 shows microscopic findiags of the central pan of the diaphysis of each ctdtured femur. In all conditions, calcification occurred in the intraperiostealbone matrix formed by osteob!asts, bzt cot in the cartilage core. ‘I?.:*~ observations are representative for four femurs in each group. In a femur cultured with medium A (Fig. 3A), a thin periosteum was observed. However, osteoblasts actively formed large trabeculae. There was a thin layer of osteoid tissue between the calcified tissue and the penosteum. In a femur cultured with medium B (Fig. 3B), the petiosteum was thick and there were numercms mesenchymal cells. A thick layer of osteoid tissue was formed by numerous osteoblasts differentiated from the mesenchymal cells in the periosteum. The calcified and uttcalcified osteoid tissue were well developed, forming large trabecuiae. In a femur cu!tured with medium C (Fig. 3C), a hypertrophic change of the periosteum due to a moderate proliferation of the mesenchymal celis was observed characteristically. The mitosis of the mesenchymal cells was seen in the periosteum. Both calcified and uncalcified tissue were developed moderately, while the ossification was less than that with medium A. In a femur cultured with medium D (Fig. 3D), a thick periosteum with a tendency to degeneration was observed. The osteoid tissue WBSscmtty and the celcified tissue slightly developed. In a femur cultured with medium E (Fig. 3E), the periosteum was the most thick and active. A marked proliferation of the mesenchymal cells and an active formation of the osteoid layer were observed. The calcified tissue was developed well. In a femur cultured with medium F (Fig. 3F), a degenerative change was seen in the outer layer of the petiosteum and an ectopic calcification was observed. The calcified tissue developed only slightly. Pathologicalchanges of periosteum of c&wed femurs Since an ectopic calcification was found in the periosteum of a femur cultured with medium F (Fig. 3F), further investigations were made on pathological changes of periwteum using cross-sections of the mid-diaphysis. In the petiosteum of a femur cultured with medium D. a formation of ectopic tissue, composed of cartilage tissue, was found (Fig. 4A). The formation of ectopic
96
Fig. 4. Pathological changes in the periosteum of femurs cultured in modified LtGJb-HW, media for 5 days. The employed modifications are shown inTablc2. A. A f~~~cultu~cdwithmediurn D: an ectopic fomration ofcartilaginous tissue in the paiasteum
(arrow). H&E
stain (X40). 8. A femurcuhuredwith
medium F: a dystmphicchan&e of meseochyms: calls (arrow) and an ectopic calcification (arrowhead) in the pariorteum. van Kossalhematoxylin/ponceau-acid
fuchsin stain (x40).
C. A femur cultured with me-
dium E: an ectopic calcification (arrows) in the petiosteu.:~ without a dystrophic change of mesenchymai cells (in the circle). van Kossalhematoxyli~ponceail-seid
fuchsin stain (x24M). D. A iemurculturedtitb
medium F: a necrotic change of mesenchymal cells around the ectopic calcification (in the circle). The section was obtained from the same femur shown in 8. H P; E stain (x2M)).
cartilaginous treated
femur
tissue
was
reproducibly
observed
in
the
periosteum
of
medium
C-
in this experiment.
In the periosteum of a femur cultured with medium F (Fig. 4B), both an ectopic calcification and a degenerative change were recognized. In a serial section obtained from the same femur, a necrotic change of mesenchymal cells was seen everywhere about the periosteum, especially around the ectopic calcification (Fig. 4D). An ectopic calcification in petiosteum was also found in a femur cultured with medium E (Fig. 4C). Near the area of the ectopic calcification, no necrotic change of cells was seen.
Discussion femur in this study included the followicg processes; mesenchymal cells in periosteum, differentiation of the mesencby-
Osteogenesis
of the cultured
proliferation
of
97
mal cells to osteoblasts, osteoid formation by the osteoblasts and calcification oi the osteoid tissue. In the culture of fetal rat long bones, calcification occurred in both the periosteai bone matrix formed by osteoblasts and the cartilage core (151. making it impossible to anaiyse cartilage and bone mineral separately. In contrast, in the embryotic chick femur used in this study, the cartilage core did not calcify. This is similar to :he situation in ova, where the diaphysial cartilage matrix is resorbed without calcification [l&17]. Ca content was increased with a chemically defined BGJb-HW, medium with a high Ca x Pi prod& where the concentra:ions of Ca and P, were 2.0 and 2.5 mM, respectively (Fig. 1). However, when the concentrations of Ca and P, were 2.5 and 2.0 mM, respectively, an effective increase in Ca content did not occur in spite of the same value of the Ca x Pi product. Ramp and Neuman [18] suggested that bone mineral is able to buffer the Ca concentration in its surxxmdingfluid, but not the P, concentration. Therefore, the effect of a high Ca x P, product on bone mineral deposition may be due to the high P, concentration rather than the high Ca concentration. However, a high Ca x Pi product was not always effective ia increasing the diaphysial Hyp content (Fig. 1). Histologically, proliferation of mesenchymal cells was not extensive, resulting in a thin periosteum even when the concentrations of Ca and Pi were 2.0 and 2.5 mM, respectively (Fig. 3A). Large but slim calcified and uncalcified osteoid tissues were observed. Bingham and Raisz [IS] showed that collagen content in cultured rat long bone was enhanced by increasing the P, concentration of the medium. On the contrary, Asher et al. [19] reported that increasing P, concentration did not alter the rate of collagen synthesis in cultured mandibles. Stimulatory effect of a high Ca x P, product on bone matrix formation was not conclusive, however, an assumption can be made that the main effect is astimulation of bone mineral deposition. A high increase in the diaphysial Hyp content by medium A (Fig. 2) may he due to the active longitudinal growth, in other words, due to the large amount of Hvp in cartilagenous tissue. CEE did not induce a considerabie increase in either Ca content or diaphysial Hyp content (Fig. 2). The data are consistent with the histological observation which revealed that the ossification developed only moderately by CEE (Fig. X). One of the typical histological changes of a femur treated with CEE was a hypertrophic one of the pedosteum. Evdo [ZO]demonstrated that CEE was indispensable for the maintenance of osteogenic cells in the periosteum using the same experimental technique. Ito et al. [21] showed that CEE ws not effective for a deposition of bone r~,ineral. Therefore, the effect of CEE on ossification in vitro must be a stimulation of the mesencbymal cell division in periosteum. The effect of serum on osteogenevis is not simple. It has been reported that bone collagen synthesis Possibly occurs without protein in organ culture [22], but calcification of ne-w matrix does not occur in the absence of serum [U]. On rhs othsr hand, it has also been reported that serum does not produce an appreciable increase in Ca content [24] and almost wtirely eliminates the stimuiatory effect of P, on calcification [18] in organ culture. In this study, HS did not stimulate a formation of OSsification by osteoblasts (Fig. 3F). This suggests that HS is not effective for activa-
98
tion of osteoblasts. Unfortunately, an cctopic calcification occurred in the periosteum (Figs. 3P. 4B) and a necrotic change of cells around the abnormal calcification was seen in a femur cultured with CEE- and HS-containing medium (Fig. 4D). The necrotic change of mesenchymal cells and the ectopic calcification in the periosteum may be caused by a dystrophy. An ectopic calcification occurred also in periosteum of a femur cultured with CEE- and &GP-containing medium (Fig. 4C). No necrotic change of cells, however, was observed in the periosteum (Fig. 4C). The 10% CEE contained inorganic and organic phosphate at concentrations of 1.15 and 1.34 mM, respectively, and a Ca concentration of 0.05 mM. In addition to the high concentration of P,, the CEE contained a high activity of alkaline phosphatase (about 63-times higher than that of the HS when p-nitrophenylphosphate was used as substrate). &GP is one of the suitable substrates for alkaline phosphatase. The ectopic calcification may be caused by increasing the concentration of P, liberated from @-GP by alkaline phosphatase of the CEE. This speculation is supported by the report of Bingham and Raisz [15] that an ectopic calcification in perlosteum was occasionally seen in cultured bone with high Ca x Pi products in the medium. However, the periosteum treated with CEE and ,!3-GP was active and the ossification was well developed (Fig. 3E). Such progressive changes were not seen in femurs treated with only CEE plus a high Ca x P, product (Fig. 3D). This suggests that the stimulatory effect ofS_GP on ossiftcatier. may be due to,!?-GP itself and not to the liberated Pi from@-GP. 1, is very interesting that a cartilaginous tissue was induced in the periosteum of the femur treated with CEE plus a high Ca x P, product (Fig. 4A). The formation of the ectopic tissue was caused by the combination of CEE and a high Ca x Pi product because each factor did not cause such an ectopic tissue (cross sections not shown). It is likely that the ectopic cartilage is formed by chondrocytes which had been differentiated from mesenchymal zells in the periosteum. In other words, CEE might contain some factor which could induce a differentiation of mesenchymal cells to chondrocytes under a high Ca x Pi product. Recently, /Y-GP has been used for studies on calcification in vitro [4,10,25,26]. These works indicate that the role of &GP in osteogenesis in vitro is as a stimulator of calcification but does not concern proliferation or differentiation of osteogenic mesenchymal cells, or orteoid formation by osteoblasts. Endogenous organic phosphates such as phosphoethanolamine and fructose 1,6-dipbosphate are also capable of inducing calcification in vi!ro [27]. The advantage of organic phosphates in calcification is that they can be a source of P, for calcification where alkaline phosphatase exists. Furthermore, as shown in this study, it is obvious that /3-GP induced a hyperplastic change of periosteum and a large osseous tissue formed by numerous osteoblasts (Fig. 3B). rhis active ossification was supported biochemically by the marked increase in Ca ‘zontent and diaphysial Hyp content (Fig. 2). Therefore, it is likely that ,&GP activates both the proliferation and differentiation of mesenchymal cells to osteoblasts in periosteum as well as calcification of osteoblast-formed osteoid tissue. Tenenbaum and Palangio [27] reported that ,9-GP decreased [‘HIthymidine uptake by the cultured periosteum of chick calvaria, thus partly supportin, the idea that @-GP may stimulate the differentiation of mesenchymal cells. Al-
99 though the role of an organic phosphate such as p-GF in ossification is not conslusive, it is likely that p-GP mimics the physiological situation as physiological organic phosphates do. The present results support the hypothesis that organic phoqhatm may play an important role in bone metabolism and calcification in vitro. It remains to be elucidated whether B-GP is only a source of phosphate ion for calcification or not. Although Gaillard [28] evaluated the effect of thyroid and pamthyroid secretions on bone. organ cultures, the discussion was made in the light of the complicated situation caused by Serum proteins which bind thyroxine in the culture medium. The present study showed that ,Y-GP in the chemically defined medium (BGJb-HW,) stimulated in vitro ossification of embryonic chick femur without any pathological changes. In addition, a large part of the thick layer of osteoid tissue formed by a Sday culture (Fig 3B) is Sure :o calcify at the sixth day of culture (not shown). AC though ,%GP-mediated calcification in vitro may not be always the same as c&i% cation in viva. the organ culture system of embryonic chick femur using p-GP-sup plemented BGJb-HW, medium is thought to be appropriate for studies on oaeogenesis because of the following adva?tages. (1) Only a S-day culture can induce a sufficient development of intraperiosteal ossification. (2) Since cartilage core does not calcify, the increase in Ca content of cultured femur reflects only the development of intraperiosteal calcified tissue quantitatively. (3) The components released into the medium from bone during cultivation can be analysed because the medium does not contain either serum or proteins. (4) The medium is easy to reproduce for the same reason. (5) It is possible by a histological technique to observe simultaneously a proliferation of mesenchymal cells in periosteum, a differentiation of the mesenchymal cells to osteoblasts, an osteoid formation by the osteoblasts and calcification of theosteoid tissue.
Ackuowletlgements We are grateful to Professor Hiroycshi Endo and Dr Kohtaro Kawashima of Teikyo University for their helpful advice on the organ culture technique.
1 Raisz LG. Bone resorption in tissue culture. Factors mfluencing the rqwnse mone. J Clin Invest 19Sl;hz:103-116,
10 paralhyroid
hor-
2 MahgoubA, SheppardH. Effects of hydroxyvitamin D derivatives on Ca release from ar fetal bolos in vitro. Endocrinology 3977;1(103629-634. 3 Gerstenfcld LC, Chipman SD, GlowackiJ, Lian JB. Expression of differentiation fuxtion by mineraliing cultures of chicken ~steoblasts. Dev Biol1987;122:49-60. 4 ad&y P, Caplan Al. ostcagenesi. in cldhueS of limb mesenchymal cells. De” Biel1979:TJ:S& *m. 5 Urist MB, DeLsngo RL, Finerman GAM. Bone cell differentiation and grow& factors. Science 1983;220:680-686.
loo 6 Sudo H, Kodama H, Amagi Y. Yamamoto S, Kasai S. In vitro differentiation and calcification in a new clonal osteogenic cell line derived from newborn mow
calvatia. I Cell Biol1983:96:191-198,
7 Bellows CC. Aubin JE, Heersche JNM, Antory ME. Mineralized bone nodules formed in vitro from cnzyr&ally
related rat calvatia cell populations. CalciFTissue Int 1986;38:143-154.
8 E&rot-Cha&
B. Glorieux F, van de;Rest M, Pereira G. Osteoblast isolated from mouse calvatia
initiate matrix mineralization in culture. J Czll Biol 1983;96:639-643. 3 Tenenbaum HC, Heersche JNM. Diffetentiation
of osteoblasts and ionnation of mineralized bone
in vitro. CalcifTissue Int 1982;33:76-79. LO Endo H. Cartilage matrix metabolism ofchondrocytes changing through their life span and its possible harmonalcontrol,
1
I Miyahata
ardemonstrated by tissue cultute studies. Connect Tissue 1974:6:139-348.
T, Kozuka. H. The direct effect of cadmium on bone metabolism and interaction between
cadmium andzincorcopperin 12 BiggwsJD,
tissuecuhure. Eisei Kagaku 1985;31:59-71.
Gwatkin RBL. Heynet S. The growth of avian and mammalian tibiae on a relatively sim-
ple chemically defined medium. Exp Cell Res 1961;25:41-58. 13 Huszar G, Maiocm J, Naftolin F. Monitotbtg of collagen and collagen fragments in chromatography
ofprotein mixtures.
Anal B&hem
1980:105:424-429.
14 Schenk RK, Olar Al, Heetmatm W. Preparation of calcified tissue for light micraropy. GR. ed Method of calcified tissue prep&ion.
15Bingham
In: Dickson
Amsterdam: Elsetier, l
DJ. Raisr LG. Bane growth in organ cultute: effect of phosphate and atber nutrients on
bone and cartilage. Calcif Tissue Res 1974;14:31-48. 16 Silvestrini G, Ricotdi ME,
Banuai
E. Resorption of uncalciticd cartilage in the diaphysis of the
chick embryo tibia. Cell Tissue Res 1979;1%:221-235. 17 Pechack DG, Kujawa MJ, Caplan Al. Motpho’ngy embryonicchick
of bone development and bone remodeling in
limbs. Bone 1986,7:459-472.
18 Ramp WK. Neuman WT. Some factors affecting mineralization iol1971;220:270-274. 19 Asher MA.
of bone
in tissue culture. Am J Phyr
Sledge CB, Glimcher MJ. The effect of inorganic anhophosphate
mt the rates of colla-
gm formation and degradation in bone and cartilage in tissue cultme. 3 Clin Endoctinol
Metab
197%3&376-389. 20 Endo H. Ossification in tissue culture. I. Histological development of the femur of chick embryo in various!iquidmedia.
ExpCell Res1959;21:151-163.
21 110 Y, Endo H, Enomoto H, Wakabayashi K, Takakura K. Ossification in tissue culture. II. Chemical development of the femur of chick embryo in vatious liquid media. Exp Cell Res 1%3;31: 119-127. 22 Past WL. The uptake of FE59 by ossew tissue arising in vitro. Am I Pathol lm,45:813-881. 23 Goldhaber P. Remodeling of bone in tissue culture. I Dent Res 1969;45:490-499. 24 Tteford ME, White AA.
Culture medium for suitable growth and development of embryonic chick
femora in vitro. Proc Sot Exp BLI Med 1964;117:536-541. 25 Tettenbaum HC. Role of or&c 1586-1589.
phosphate in mineralization of bane in vitro. I Dent Res 1981:MI:
26 Luria EA, Owen ME, Ftiedcnstein Al. Morris IF, Kttznehow SA. Bone fannation
in organ culture
of bone marrow. Cell Tissue Res 1987;248:449-454, 27 Tenenbaum HC, Palangio K. Phosphoethanolamitte and fructose 1.6-diphasphate-induced
calcium
uptake in bone formed in vitro. Bone Mineral 1987;2:201-210. 28 Gaillard PJ. The effect of thyroid and parathyroid sectetions on explanted mouse radius rudiments. Dev Biol 1963;7:103-116.
A suitable culture medium for ossification of embryonic chick femur in organ culture
Toshiyuki Kaji, Rie Kawatani, Takumi Hoshino, Tatsuro Miyahara, Hiroshi Kozuka, Youichi Kurzshige’ and Fumitomo Koizumi’ Se&n
of Toxicology and ‘Departmenr of Pnfhology, Faculrv OJ Pharmacwicol
Sciences and
‘.Me.?icme, T’oyoma Medical and Phormoceurical Omversiry. 1630 Sugirani. Toyamo-rhr. Toyamo. 530-01. JaPon (Received 22 June 19SR) (Accepted 17January 1990)
To establish a culture medium which altnws owfication murswerecultured
invari,~u~lyrupplemented
in organ c&u:;.
Changes of Ca and P, concentrations in the BGJb-HW2 extract (CEE)
g-day-old embryonic chick fe-
BGJb-HW,media. medium or the LO% addition of chick embryo
did “(11 induce ossification. Furthermore. combinations
of the
10% CEE with ahigh Ca x
P, product or with 5 mM &glycemphorphare f&G?) or with LO% horse serum plus a high Ca x P, prodyc1 often caused pathological abnormalities in the pedonc..... On the other hand, BGJb-HW,
medium supplemented with 5 mM,&GP
indwddevetopmenr
ofoti-
fication. The Ca content of femurs and the diaphysial hydmxypmline ccntent were markedly increased. Histological obsrarion
showed a formation at a thick and active periosteum, numerom osteobtastic
cells, a sufficient amount of o ,teaid tissue and well dewlopzd calcified trabeculae without anypathotqcai changes. Thus, tb,c organ culture system using this medium was considered to be an appropriate one for studies an osteogsncsir ip. vile.
Key words: Bone formation: Calcification: Differentiahnn:
Fmb~onicchickbone:Tissue
culture:
Ossification
Introduction Organ culture systems have been used extensively for studying factors of bone resorption [1,2]. However, organ culture systems have not been widely applied to
016%6w91~~003.50
0 1990 Elscvier Sacnce Pub&hers B.V. (Bmmedral
Division)
90 studies on bone formation, particularly calcification. As pointed out by Gerstenfeld et al. [3], several factors appear to enhance calcification in cell culture systems. These factors include: (1) a highly enriched nutrient medium [4,5]: (2) the time period that cells are kept in culture [6]; (3) the use of ,%glycerophosphate @GP) or other organic phosphates [7-g]. This infurmation is suggestive for establishment of an organ culture system in which ossification occurs sufficiently. Such a system will be useful for studies on bone formation. Previously, a chemically defined cultxe m&urn BGJb-HW, [lo] was used in our studies [!I]. This medium was established by Endo et al. to induce active collagen synthesis of cultured bone on the basis of Bigger’s BGJb medium [12]. Although BGJb-HW* medium does not contain eirher serum or proteins such aa physiological growth factors, it is excellent for studying collagen metabolism, especially for that of chondrocytes. However, the medium did not induce a sufficient ossification, particularly calcification. Desiring ossification, modifications were made using calcium (Ca), inorganic phosphate (Pi), p-GP, chick embryo extract (CEE) and horse seT”rn (US). In the pxsent study, we have investigated in vitro ossification of embryonic chick femur in the modified BGJb-HW, media and discussed the effect of each supplement on ossification. Materials and Methods
Fertilized eggs from White Leghorn hens were incubated for 9 days in an electoric incubator and a whole femur was resected from an embryonic chicken under sterile conditions. The femur was added to 1.5 ml of the culture medium in a Leightontype culture tube. The tube was incubated at 38 ‘C in a rotating culture drum apparatus; the medium was changed every 2 days. The basal medium was BGlb-HW, [ll] (Table 1) which contained 1.26 mM Ca and 1.52 mM P,, while modifications were carried uut to examine their effect OR ossification. The employed modifications are shown in Fig. 1 and Table 2. Preporotion of chick embryo extract (CEE) Eleven-day-old chick embryos were crushed in a mechanical homogenizer under sterile conditions and the homogenate was centrifuged at 10 000 ‘pm for 15 min. The supematant fluid was frozen and thawed 7 times and centiifiiged at 10 000 rpm for 15 min again to obtain the supernatant (CEE). This CEE contained 0.49 mM Ca and 11.5 mM P,.
HS was purchased from Nakarai Kagakuyakuhin Inc., Kyoto, Japan. The HS had already been inactivated and contained 2.56 mM Ca and 1.54 mM P,. Anulytical
methods
A whole femur was extracted for 2 days with cold 0.1 M acetate buffer contaiuing
91 Table 1 The composition of chemically defined BGJb-HW,
medium
established
by Endo
[ill L-LyaiaeHC! !_-HistidineHCI~H,O L-ArginineHCI L-Tbreonine kValine L-Leucine L-lroleucine r-Mcthionine L-Phenykdaninc LTlyp’0pba”e r-Tyrosine 1,CysteincHCl.H,O ~Glutamine Glycine L-Serine l.4klli”~ Nicodnamide Thiamine.HCl Ca-pantothenate Riboflavine
11
n-.,A,
_.,“...~~.
15.0 7.5
7.5 6.5 5.0 3.0 5.0 5.0 4.0 4.0 9.0 20.0 15.0 10.5 11.5 2.0 0.4
0.02
DIrrirlnr., -,.---
-or
“hnsrhalp
0 m rnjdl
Fohc acid
0.02
Biotin p-‘minobrwo.c acid Na-n-tocopher>Iphosphate Cbob.wcbloride m-lnoskll Cyanocoba!amine Na-arcorbae
0.02
N&i KCI CZCI,.ZFI,O MgSO,.7H,O Na,HPO,.12H,O KH,PO, NaHCO, F&I, Glucose Phenolred
0.2 0.1 5.0 3.82 0.CC.l 5.0 800.0 40.0 18.5 20.0 12.1 16.0 140.0 Cl.047 MO.0 2.0
0.02
0.01 M EDTA (pH 5.5). After extraction, the epiphysial bulge was cut off from the diaphysis and each part was analysed for hydroxyproline (Hyp) content by the method of Huszar et al. [13]. The extract was analysed for Ca content by atozk absorption spectrophotometry. Since one femur of the same embrj<* vile n~ltured and the other was uncultured, increase in Ca and Hyp content was expressed as the percentage of Ca and Hyp of the corresponding uncultured bone.
Histological technique Femurs were fiied in 10% neutral formalin. They were embedded in paraffin in a conventional way without decalcification. Femurs were sectioned serially to 6.um thickness. These sections were stained by the method of van Kossa counterstained with Weiger’s hematoxylin and ponceau-acid titchsin [14] or stained with Mayer’s hematoxylin and eosin.
Results on Ca and Hyp contents were analysed for statistical significance by Student’s t-test. P values of less than 0.05 were considered to indicate statistically significant differences.
92 Results Ef&f n,fCg avd P, concentrations on Ca ond Hyp confents in fenws First, Ca and P, concentratiLo& i;: BGJb-KWz medium were cha :ged. Pre!iminary studies showed ectopic calcification in the periosteum of femurs cultured in BGJbHW, medium containing Ca x Pi products above 6 mM*. Based on the result, Ca x P, products at 5.0 mM2 and below were examined in this experiment. Figure 1 shows the effects of Ca and Pi concentrations on Ca and Hyp content of the cultured femurs. Although no marked increase in Ca content occurred as a whole, the highest increase in Ca content was seen when Ca and Pi concentrations were 2.0 and 2.5 mM, respectively. No marked increase in Hyp content occurred in the diaphysis by changing Ca and Pi concentrations. In the epiphysis, a tendency toward greater Hyp content was observed when Ca X Pi products were between 4.40 and 4.84, while no significant change was recognized statistically. Effecf of several modifications using a high Co X P,product, t¶-GP, CEE and HS on Ca and Hyp content in femurs As stated above, changes of Ca and P, concentrations could not induce a sufficient increase in Ca and diaphysial Hyp content in femurs by themselves. Therefore, modifications were made using fl-GP, CEE, HS and high Ca x P, products (5.0 mM2 and above). The experimental design is summarized in Table 2. Since either CEE or HS contained a significant amount of Ca and Pi (see Materials and Methods), Ca and Pi concentrations of the modified media were not always equal in spite of the same basal medium. In other words, CEE and HS were evaluated as supplements which rnntain a significant amount of Ca and Pv Each modified medium was given a provisional name (from A to F). Lengths of femurs before culture were 5.01
Fig. 1. Effect of Ca and pi c”ncentrstionson the increasein Ca contentin the whole femur, hydroxyproline (Hyp) contentin the diaphysisand in the epiphysis.The femurswere c”lt”red fcr 6 days.Values are means+ SE for four sampleswith percentageof eachcontentof the unculturedfemur. The Ca comem when ca and P, mncemrationsare 2.0 and 2.50 rnM, respectively,is significantlymare than th”sewhen Ca and P, concentrationsarc Z.0 and 2.00 mM (P 4 O.WN), 2.8 snd 1.52 mM (P c 0.05). 2.2 and 2.20 mM (P c 0.05)and 2.5 and2.W mM (P < 0.01).ThediaphysialHypcontent when CaandP,conce”tmlions xe 2.0 and 2.50 mM, respectively,is significantlym”re than thosewhen Ca and P, concentrations
93
Table 2 Summary of modifications of chemically defined BGJb-HW, medium Medium
Basal medium
Ca conccn-
P, eoncen-
trhm
tration
@M) ~-~
HS
Day, in
(%)
Nh”X
(mM)
A
BGJSW’,
2.0
2.5
Nmc
None
None
5
B
BGIb-HWz
2.0
1.5
s
None
None
5
C
BG;b-HW,
1.3
2.7
None
10
None
5
D
BGJb-HW,
2.1
3.7
None
1”
N.mC
5
E
BGJb-HW,
z.,
2.7
5
111
None
5
F
BGJb-HW,
2.3
3.8
None
10
LO
5
k 0.01 mm and the longitudinal growth of femurs by cultivation with medium A, B, C,D,EorFwas4.4+0.2,3.9+0.1,2.1+0.4,2.8+0.3,2.1~0.4or1.6+0.1 mm, respectively. The increase in Ca and Hyp contents of femurs cultured in each mod&d mediuii; in respect to uncultured femurs are shown in Fig. 2. The Ca content in femurs of both medium A and B increased by about 5 times but the values varied wideiy for medium A. The Ca content for medium E and F increased by 3 times. There was no sufficient change in Ca content in both medium C and D. The diaphysis consists of an intraperiosteal bone and a cartilage core, so that the diaphysial Hyp content reflects both the intraperiosteal bone matrix formation by
Ffg. 2. Effect of
modifications of BGJb-HW,
medium using a high Ca x F, product. jLglyempbcqkte.
chick embryo extract and horse serum on the increase in Ca content in the whole femur and hydmqproline (Hyp) content in tile diaphysis and in the epipbysir. The employed mo.iiEcaziom are *own i. Table 2. F.%nYn were E”,Nred for 5 clays. “?.,ueS are means f SE for far rampIes with perccmage of each content the ““cl&“red femur. As regards Ca content, B is rignincantly higherthan C (P< ll.00,). D (P< 0.05) and F(P < 0.05): C is rigniiicanrly lower than D (P< 0.0,). E (Pc0.M) and F(P< 0.05).
of
As regards diaphyaial Hyp content, A is significantly higher than C (P 4 0.05), E (PC
0.05); B issigniti-
cantly higher than C (P < 0.01). D (P 4 0.05). E (P < 0.001) and F (P < 0.031). As regards epiphysial Hyp content, C is significantly lower than A (P < 0.05) and F(P<
0.05).
94
.
.,I
y’
:;
..’
A‘:
..: :-
_:,
:
. ; ‘.‘.’
.*.:.
.
:
.;
‘*
. 1’. ..\-.
: . . . _ ,. -
.: . . . . ‘i._.I ‘.._..:*
1
:;*.j
. : .,, _. .
.,‘..
p,‘?; _*._-
,
;‘.,.*
;’
,..:.a:,
_, ,: ..,,..‘,
:j.:
:::.
:
;
:,
.’
1:‘..:.!:‘:...-.‘::~;. .-.,“_, :‘.
‘>.I
.::
..:
Pip. 1 Miausupic findings of the central pan of the diaphysis of the iettttu cultured with modified Btilb-HWt media for 5 days. The employed modifications are shown in Table 2. A. A femur cultured with medium A: a slightly active proliferation mesenchymal cellsin athinperiateum(PO), a thin layer of osteoid tissue (OS) and a large catdried eabecula (T). van Kossalhematorylinlponceau-acid hahsin stain (x100). B. Afemurcutturedwithmedium B: an acdvepratiferstionofmesenchymdceUs(MC) in a thick PO, a thick layer OSon the inside oftheP0 and alarge T. van KossalhematoxytinlponMauacidfuchsinstain (xltltt). C. AfemurculturedwithmediumC: a hypertropbkchange PO,B thinlayer of OS and a small T. van Korsalhelnatoxylinlponceau-a*d fuchsin stain (X100). D. A femur ctdtured with medium D: a rhrci P3 irritutir~g a tendency of degeneration (D), a %anty OS and a slightly developed T. “cl” KoJsamematolty,in/ponccsu-afid fvchrin stain (XNW). E. A fEm”r cultured WtIb medium E: a very thick PD. a thick layer of OS and a small T. “0” KosrYhema~~ytilllponEesu-add fucbsi” slain (x IW,. P. A femur cultured with medun F: a d@ophicchaoge (arrowheads) and an ectopic calcification (arrows) in a thick PO and a slightly developed T. won K~ralhematorylinlponeeau-acid fuchsin
of
of
of
95
osteoblasts and cartilage formation by chondrocytes. Unfortunately, it is impossible to measure the bone and cartilage Hyp contents separately. In addition, the Iongitudinal growth by cartilage formation may contribute to the diaphysial Hyp content. However, the diaphysial Hyp content w&s measured nevertheless, since it may reflect bone matrix formation. The high.% increases in diaphysial Hyp content were recognized in femurs cultured with medium B. Medium A increased diaphysial Hyp content by about 3 times, while the others only did by about 2 times. In the epiphysis, which consists of only cartilage, medium A markedly increased Hyp ccmtent by about 4.5 times. Hisrological investigahon of ossificalionin femurs cultured with each modified BGJb-HW, medium Figure 3 shows microscopic findiags of the central pan of the diaphysis of each ctdtured femur. In all conditions, calcification occurred in the intraperiostealbone matrix formed by osteob!asts, bzt cot in the cartilage core. ‘I?.:*~ observations are representative for four femurs in each group. In a femur cultured with medium A (Fig. 3A), a thin periosteum was observed. However, osteoblasts actively formed large trabeculae. There was a thin layer of osteoid tissue between the calcified tissue and the penosteum. In a femur cultured with medium B (Fig. 3B), the petiosteum was thick and there were numercms mesenchymal cells. A thick layer of osteoid tissue was formed by numerous osteoblasts differentiated from the mesenchymal cells in the periosteum. The calcified and uttcalcified osteoid tissue were well developed, forming large trabecuiae. In a femur cu!tured with medium C (Fig. 3C), a hypertrophic change of the periosteum due to a moderate proliferation of the mesenchymal celis was observed characteristically. The mitosis of the mesenchymal cells was seen in the periosteum. Both calcified and uncalcified tissue were developed moderately, while the ossification was less than that with medium A. In a femur cultured with medium D (Fig. 3D), a thick periosteum with a tendency to degeneration was observed. The osteoid tissue WBSscmtty and the celcified tissue slightly developed. In a femur cultured with medium E (Fig. 3E), the periosteum was the most thick and active. A marked proliferation of the mesenchymal cells and an active formation of the osteoid layer were observed. The calcified tissue was developed well. In a femur cultured with medium F (Fig. 3F), a degenerative change was seen in the outer layer of the petiosteum and an ectopic calcification was observed. The calcified tissue developed only slightly. Pathologicalchanges of periosteum of c&wed femurs Since an ectopic calcification was found in the periosteum of a femur cultured with medium F (Fig. 3F), further investigations were made on pathological changes of periwteum using cross-sections of the mid-diaphysis. In the petiosteum of a femur cultured with medium D. a formation of ectopic tissue, composed of cartilage tissue, was found (Fig. 4A). The formation of ectopic
96
Fig. 4. Pathological changes in the periosteum of femurs cultured in modified LtGJb-HW, media for 5 days. The employed modifications are shown inTablc2. A. A f~~~cultu~cdwithmediurn D: an ectopic fomration ofcartilaginous tissue in the paiasteum
(arrow). H&E
stain (X40). 8. A femurcuhuredwith
medium F: a dystmphicchan&e of meseochyms: calls (arrow) and an ectopic calcification (arrowhead) in the pariorteum. van Kossalhematoxylin/ponceau-acid
fuchsin stain (x40).
C. A femur cultured with me-
dium E: an ectopic calcification (arrows) in the petiosteu.:~ without a dystrophic change of mesenchymai cells (in the circle). van Kossalhematoxyli~ponceail-seid
fuchsin stain (x24M). D. A iemurculturedtitb
medium F: a necrotic change of mesenchymal cells around the ectopic calcification (in the circle). The section was obtained from the same femur shown in 8. H P; E stain (x2M)).
cartilaginous treated
femur
tissue
was
reproducibly
observed
in
the
periosteum
of
medium
C-
in this experiment.
In the periosteum of a femur cultured with medium F (Fig. 4B), both an ectopic calcification and a degenerative change were recognized. In a serial section obtained from the same femur, a necrotic change of mesenchymal cells was seen everywhere about the periosteum, especially around the ectopic calcification (Fig. 4D). An ectopic calcification in petiosteum was also found in a femur cultured with medium E (Fig. 4C). Near the area of the ectopic calcification, no necrotic change of cells was seen.
Discussion femur in this study included the followicg processes; mesenchymal cells in periosteum, differentiation of the mesencby-
Osteogenesis
of the cultured
proliferation
of
97
mal cells to osteoblasts, osteoid formation by the osteoblasts and calcification oi the osteoid tissue. In the culture of fetal rat long bones, calcification occurred in both the periosteai bone matrix formed by osteoblasts and the cartilage core (151. making it impossible to anaiyse cartilage and bone mineral separately. In contrast, in the embryotic chick femur used in this study, the cartilage core did not calcify. This is similar to :he situation in ova, where the diaphysial cartilage matrix is resorbed without calcification [l&17]. Ca content was increased with a chemically defined BGJb-HW, medium with a high Ca x Pi prod& where the concentra:ions of Ca and P, were 2.0 and 2.5 mM, respectively (Fig. 1). However, when the concentrations of Ca and P, were 2.5 and 2.0 mM, respectively, an effective increase in Ca content did not occur in spite of the same value of the Ca x Pi product. Ramp and Neuman [18] suggested that bone mineral is able to buffer the Ca concentration in its surxxmdingfluid, but not the P, concentration. Therefore, the effect of a high Ca x P, product on bone mineral deposition may be due to the high P, concentration rather than the high Ca concentration. However, a high Ca x Pi product was not always effective ia increasing the diaphysial Hyp content (Fig. 1). Histologically, proliferation of mesenchymal cells was not extensive, resulting in a thin periosteum even when the concentrations of Ca and Pi were 2.0 and 2.5 mM, respectively (Fig. 3A). Large but slim calcified and uncalcified osteoid tissues were observed. Bingham and Raisz [IS] showed that collagen content in cultured rat long bone was enhanced by increasing the P, concentration of the medium. On the contrary, Asher et al. [19] reported that increasing P, concentration did not alter the rate of collagen synthesis in cultured mandibles. Stimulatory effect of a high Ca x P, product on bone matrix formation was not conclusive, however, an assumption can be made that the main effect is astimulation of bone mineral deposition. A high increase in the diaphysial Hyp content by medium A (Fig. 2) may he due to the active longitudinal growth, in other words, due to the large amount of Hvp in cartilagenous tissue. CEE did not induce a considerabie increase in either Ca content or diaphysial Hyp content (Fig. 2). The data are consistent with the histological observation which revealed that the ossification developed only moderately by CEE (Fig. X). One of the typical histological changes of a femur treated with CEE was a hypertrophic one of the pedosteum. Evdo [ZO]demonstrated that CEE was indispensable for the maintenance of osteogenic cells in the periosteum using the same experimental technique. Ito et al. [21] showed that CEE ws not effective for a deposition of bone r~,ineral. Therefore, the effect of CEE on ossification in vitro must be a stimulation of the mesencbymal cell division in periosteum. The effect of serum on osteogenevis is not simple. It has been reported that bone collagen synthesis Possibly occurs without protein in organ culture [22], but calcification of ne-w matrix does not occur in the absence of serum [U]. On rhs othsr hand, it has also been reported that serum does not produce an appreciable increase in Ca content [24] and almost wtirely eliminates the stimuiatory effect of P, on calcification [18] in organ culture. In this study, HS did not stimulate a formation of OSsification by osteoblasts (Fig. 3F). This suggests that HS is not effective for activa-
98
tion of osteoblasts. Unfortunately, an cctopic calcification occurred in the periosteum (Figs. 3P. 4B) and a necrotic change of cells around the abnormal calcification was seen in a femur cultured with CEE- and HS-containing medium (Fig. 4D). The necrotic change of mesenchymal cells and the ectopic calcification in the periosteum may be caused by a dystrophy. An ectopic calcification occurred also in periosteum of a femur cultured with CEE- and &GP-containing medium (Fig. 4C). No necrotic change of cells, however, was observed in the periosteum (Fig. 4C). The 10% CEE contained inorganic and organic phosphate at concentrations of 1.15 and 1.34 mM, respectively, and a Ca concentration of 0.05 mM. In addition to the high concentration of P,, the CEE contained a high activity of alkaline phosphatase (about 63-times higher than that of the HS when p-nitrophenylphosphate was used as substrate). &GP is one of the suitable substrates for alkaline phosphatase. The ectopic calcification may be caused by increasing the concentration of P, liberated from @-GP by alkaline phosphatase of the CEE. This speculation is supported by the report of Bingham and Raisz [15] that an ectopic calcification in perlosteum was occasionally seen in cultured bone with high Ca x Pi products in the medium. However, the periosteum treated with CEE and ,!3-GP was active and the ossification was well developed (Fig. 3E). Such progressive changes were not seen in femurs treated with only CEE plus a high Ca x P, product (Fig. 3D). This suggests that the stimulatory effect ofS_GP on ossiftcatier. may be due to,!?-GP itself and not to the liberated Pi from@-GP. 1, is very interesting that a cartilaginous tissue was induced in the periosteum of the femur treated with CEE plus a high Ca x P, product (Fig. 4A). The formation of the ectopic tissue was caused by the combination of CEE and a high Ca x Pi product because each factor did not cause such an ectopic tissue (cross sections not shown). It is likely that the ectopic cartilage is formed by chondrocytes which had been differentiated from mesenchymal zells in the periosteum. In other words, CEE might contain some factor which could induce a differentiation of mesenchymal cells to chondrocytes under a high Ca x Pi product. Recently, /Y-GP has been used for studies on calcification in vitro [4,10,25,26]. These works indicate that the role of &GP in osteogenesis in vitro is as a stimulator of calcification but does not concern proliferation or differentiation of osteogenic mesenchymal cells, or orteoid formation by osteoblasts. Endogenous organic phosphates such as phosphoethanolamine and fructose 1,6-dipbosphate are also capable of inducing calcification in vi!ro [27]. The advantage of organic phosphates in calcification is that they can be a source of P, for calcification where alkaline phosphatase exists. Furthermore, as shown in this study, it is obvious that /3-GP induced a hyperplastic change of periosteum and a large osseous tissue formed by numerous osteoblasts (Fig. 3B). rhis active ossification was supported biochemically by the marked increase in Ca ‘zontent and diaphysial Hyp content (Fig. 2). Therefore, it is likely that ,&GP activates both the proliferation and differentiation of mesenchymal cells to osteoblasts in periosteum as well as calcification of osteoblast-formed osteoid tissue. Tenenbaum and Palangio [27] reported that ,9-GP decreased [‘HIthymidine uptake by the cultured periosteum of chick calvaria, thus partly supportin, the idea that @-GP may stimulate the differentiation of mesenchymal cells. Al-
99 though the role of an organic phosphate such as p-GF in ossification is not conslusive, it is likely that p-GP mimics the physiological situation as physiological organic phosphates do. The present results support the hypothesis that organic phoqhatm may play an important role in bone metabolism and calcification in vitro. It remains to be elucidated whether B-GP is only a source of phosphate ion for calcification or not. Although Gaillard [28] evaluated the effect of thyroid and pamthyroid secretions on bone. organ cultures, the discussion was made in the light of the complicated situation caused by Serum proteins which bind thyroxine in the culture medium. The present study showed that ,Y-GP in the chemically defined medium (BGJb-HW,) stimulated in vitro ossification of embryonic chick femur without any pathological changes. In addition, a large part of the thick layer of osteoid tissue formed by a Sday culture (Fig 3B) is Sure :o calcify at the sixth day of culture (not shown). AC though ,%GP-mediated calcification in vitro may not be always the same as c&i% cation in viva. the organ culture system of embryonic chick femur using p-GP-sup plemented BGJb-HW, medium is thought to be appropriate for studies on oaeogenesis because of the following adva?tages. (1) Only a S-day culture can induce a sufficient development of intraperiosteal ossification. (2) Since cartilage core does not calcify, the increase in Ca content of cultured femur reflects only the development of intraperiosteal calcified tissue quantitatively. (3) The components released into the medium from bone during cultivation can be analysed because the medium does not contain either serum or proteins. (4) The medium is easy to reproduce for the same reason. (5) It is possible by a histological technique to observe simultaneously a proliferation of mesenchymal cells in periosteum, a differentiation of the mesenchymal cells to osteoblasts, an osteoid formation by the osteoblasts and calcification of theosteoid tissue.
Ackuowletlgements We are grateful to Professor Hiroycshi Endo and Dr Kohtaro Kawashima of Teikyo University for their helpful advice on the organ culture technique.
1 Raisz LG. Bone resorption in tissue culture. Factors mfluencing the rqwnse mone. J Clin Invest 19Sl;hz:103-116,
10 paralhyroid
hor-
2 MahgoubA, SheppardH. Effects of hydroxyvitamin D derivatives on Ca release from ar fetal bolos in vitro. Endocrinology 3977;1(103629-634. 3 Gerstenfcld LC, Chipman SD, GlowackiJ, Lian JB. Expression of differentiation fuxtion by mineraliing cultures of chicken ~steoblasts. Dev Biol1987;122:49-60. 4 ad&y P, Caplan Al. ostcagenesi. in cldhueS of limb mesenchymal cells. De” Biel1979:TJ:S& *m. 5 Urist MB, DeLsngo RL, Finerman GAM. Bone cell differentiation and grow& factors. Science 1983;220:680-686.
loo 6 Sudo H, Kodama H, Amagi Y. Yamamoto S, Kasai S. In vitro differentiation and calcification in a new clonal osteogenic cell line derived from newborn mow
calvatia. I Cell Biol1983:96:191-198,
7 Bellows CC. Aubin JE, Heersche JNM, Antory ME. Mineralized bone nodules formed in vitro from cnzyr&ally
related rat calvatia cell populations. CalciFTissue Int 1986;38:143-154.
8 E&rot-Cha&
B. Glorieux F, van de;Rest M, Pereira G. Osteoblast isolated from mouse calvatia
initiate matrix mineralization in culture. J Czll Biol 1983;96:639-643. 3 Tenenbaum HC, Heersche JNM. Diffetentiation
of osteoblasts and ionnation of mineralized bone
in vitro. CalcifTissue Int 1982;33:76-79. LO Endo H. Cartilage matrix metabolism ofchondrocytes changing through their life span and its possible harmonalcontrol,
1
I Miyahata
ardemonstrated by tissue cultute studies. Connect Tissue 1974:6:139-348.
T, Kozuka. H. The direct effect of cadmium on bone metabolism and interaction between
cadmium andzincorcopperin 12 BiggwsJD,
tissuecuhure. Eisei Kagaku 1985;31:59-71.
Gwatkin RBL. Heynet S. The growth of avian and mammalian tibiae on a relatively sim-
ple chemically defined medium. Exp Cell Res 1961;25:41-58. 13 Huszar G, Maiocm J, Naftolin F. Monitotbtg of collagen and collagen fragments in chromatography
ofprotein mixtures.
Anal B&hem
1980:105:424-429.
14 Schenk RK, Olar Al, Heetmatm W. Preparation of calcified tissue for light micraropy. GR. ed Method of calcified tissue prep&ion.
15Bingham
In: Dickson
Amsterdam: Elsetier, l
DJ. Raisr LG. Bane growth in organ cultute: effect of phosphate and atber nutrients on
bone and cartilage. Calcif Tissue Res 1974;14:31-48. 16 Silvestrini G, Ricotdi ME,
Banuai
E. Resorption of uncalciticd cartilage in the diaphysis of the
chick embryo tibia. Cell Tissue Res 1979;1%:221-235. 17 Pechack DG, Kujawa MJ, Caplan Al. Motpho’ngy embryonicchick
of bone development and bone remodeling in
limbs. Bone 1986,7:459-472.
18 Ramp WK. Neuman WT. Some factors affecting mineralization iol1971;220:270-274. 19 Asher MA.
of bone
in tissue culture. Am J Phyr
Sledge CB, Glimcher MJ. The effect of inorganic anhophosphate
mt the rates of colla-
gm formation and degradation in bone and cartilage in tissue cultme. 3 Clin Endoctinol
Metab
197%3&376-389. 20 Endo H. Ossification in tissue culture. I. Histological development of the femur of chick embryo in various!iquidmedia.
ExpCell Res1959;21:151-163.
21 110 Y, Endo H, Enomoto H, Wakabayashi K, Takakura K. Ossification in tissue culture. II. Chemical development of the femur of chick embryo in vatious liquid media. Exp Cell Res 1%3;31: 119-127. 22 Past WL. The uptake of FE59 by ossew tissue arising in vitro. Am I Pathol lm,45:813-881. 23 Goldhaber P. Remodeling of bone in tissue culture. I Dent Res 1969;45:490-499. 24 Tteford ME, White AA.
Culture medium for suitable growth and development of embryonic chick
femora in vitro. Proc Sot Exp BLI Med 1964;117:536-541. 25 Tettenbaum HC. Role of or&c 1586-1589.
phosphate in mineralization of bane in vitro. I Dent Res 1981:MI:
26 Luria EA, Owen ME, Ftiedcnstein Al. Morris IF, Kttznehow SA. Bone fannation
in organ culture
of bone marrow. Cell Tissue Res 1987;248:449-454, 27 Tenenbaum HC, Palangio K. Phosphoethanolamitte and fructose 1.6-diphasphate-induced
calcium
uptake in bone formed in vitro. Bone Mineral 1987;2:201-210. 28 Gaillard PJ. The effect of thyroid and parathyroid sectetions on explanted mouse radius rudiments. Dev Biol 1963;7:103-116.