Osteogenic differentiation of cultured rat and human bone

marrow cells on the surface of zinc-releasing calcium

phosphate ceramics

Masako Ikeuchi,1,2 Atsuo Ito,1 Yoshiko Dohi,3 Hajime Ohgushi,1 Hideki Shimaoka,3 Kunio Yonemasu,3
Tetsuya Tateishi1
1
National Institute of Advanced Industrial Science and Technology (AIST), Tissue Engineering Research Center
(TERC), 3-11-46 Nakouji, Amagasaki, Hyougo 661-0974, Japan
2
Department of Oral and Maxillofacial Surgery, Nara Medical University, 840 Shijo-cho, Kashihara,
Nara 634-8522, Japan
3
Department of Public Health, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan

Received 17 September 2002; accepted 9 January 2003

Abstract: Rat and human bone marrow cells (BMCs) were surface, and the addition of 100 ␮M of free ZnCl2 (6.5
cultured on a composite ceramic of zinc-containing ␤-trical- ␮g/mL) to the culture medium significantly increased the
cium phosphate and hydroxyapatite (ZnTCP/HAP) with a ALP activity of the BMCs relative to the culture medium
(Ca⫹Zn)/P molar ratio of 1.60 and varying zinc contents. without the ZnCl2 addition. The maximum zinc concentra-
After a 2-week culture of the BMCs in the presence of tion required to enhance mineralization was higher in hu-
␤-glycerophosphate and dexamethasone, many macroscopic man BMCs than in rat BMCs. The present study demon-
mineralized areas with high alkaline phosphatase (ALP) strates the superiority of ZnTCP/HAP ceramics over TCP/
activity were seen on the ZnTCP/HAP ceramic disks. The HAP in supporting the osteogenic differentiation of BMCs,
ALP activity increased with increasing zinc content in the and thus these ceramics are safe and useful in clinical set-
ceramics. The highest ALP activity was observed when the tings, such as for bone reconstructive surgery. © 2003 Wiley
BMCs were cultured on the ceramics with 1.26 wt % zinc, Periodicals, Inc. J Biomed Mater Res 67A: 1115–1122, 2003
and the ceramics released zinc ions at concentrations from
2.2 to 7.2 ␮g/mL into the culture medium. Zinc ions were
incorporated into mineralized areas produced by BMCs. Key words: zinc; TCP/HAP; bone marrow cell; alkaline
BMCs also were directly cultured onto the culture dish phosphatase; osteogenesis

INTRODUCTION molar ratio of 1.60, and the other is a monophasic

ZnTCP ceramic with a (Ca⫹Zn)/P molar ratio of 1.50.
Zinc-containing ␤-tricalcium phosphate (ZnTCP) ZnTCP/HAP significantly promoted mouse osteo-
functions as a carrier of zinc, an essential element that blastic MC3T3-E1 cell proliferation in vitro at a zinc
has stimulatory effects on bone formation in vitro and content of 1.20 wt %.4 Cortical bone apposition in
in vivo.1–3 Using ZnTCP, two types of zinc-releasing rabbit femora to the ZnTCP/HAP implant with a zinc
calcium phosphate ceramics, useful for promoting content of 0.316 wt % increased significantly even at 60
bone formation, were developed: one is a composite of weeks after implantation compared with that of a
ZnTCP and hydroxyapatite (HAP) with a (Ca⫹Zn)/P TCP/HAP implant with a Ca/P molar ratio of 1.60. 5
It is still unclear whether the mechanism underlying
Correspondence to: H. Ohgushi; e-mail: hajime-ohgushi@ increased bone formation is associated with the supe-
aist.go.jp riority of ZnTCP/HAP over TCP/HAP in supporting
Contract grant sponsor: Three-Dimensional Tissue Mod-
ule Project, METI (A Millennium Project) osteogenic differentiation. The indices of osteogenic
Contract grant sponsor: New Energy and Industrial Tech- differentiation include biochemical parameters, such
nology Development Organization (NEDO); R&D projects as alkaline phosphatase (ALP) activity. Zinc at concen-
in ‘Advanced Support System for Endoscopic and Other trations several times higher than the physiologic level
Minimally Invasive Surgery,‘ entrusted to the Japan Fine
Ceramics Center
resulted in an increase in ALP activity of human os-
teoblastic cell lines and rat calvaria tissue culture.2,6
© 2003 Wiley Periodicals, Inc. Therefore ZnTCP/HAP may promote osteogenic dif-
1116 IKEUCHI ET AL.

ferentiation. However, these data were obtained using was mixed with TCP and HAP powders, maintaining a
cells that cannot be applied in a clinical situation. (Ca⫹Zn)/P molar ratio of 1.60. The mixed powders were
From a clinical viewpoint of the application of ground, passed through a 75-␮m sieve, supplemented with
ZnTCP/HAP as bone graft substitutes, it is essential to 3 wt % poly(vinyl alcohol) and 1 wt % poly(ethylene glycol),
pressed at 98 MPa, and, finally, sintered at 1100°C for 1 h to
clarify the effect of ZnTCP/HAP on bone formation in
obtain sintered disks of ZnTCP/HAP composite ceramics.
humans. To assess the optimum zinc content, we eval- The size of the disks was 14 mm in diameter and 1 mm in
uated the effect of ZnTCP/HAP on osteogenic differ- thickness. Disks of six different zinc contents (0, 0.32, 0.63,
entiation using in vitro mouse osteoblastic cell cultures 0.88, and 1.26 wt %) were prepared.
and in vivo implantation in rabbits.4,5 However, it is
still unclear whether or not the optimum zinc content
of ZnTCP/HAP depends on animal species. Plasma
zinc concentrations are roughly the same between hu- Rat bmc preparation and culture
man and rat. Thus a comparative study of osteogenic
differentiation in human and rat cell cultures on BMCs were obtained from the bone shafts of femora of
ZnTCP and ZnTCP/HAP would provide information Fischer 344, male, 7-week-old rats. Both ends of the femora
that could verify whether optimum zinc content is were cut away from the epiphysis, and the marrow was
determined by plasma zinc concentration or by spe- flushed out using 10 mL of the culture medium with a
cies-specific factors. If the comparative study could be 21-gauge needle syringe. The released cells were collected in
performed using clinically available cells, such as hu- two T-75 flasks (Costar Co., Cambridge, MA) containing 15
man bone marrow cells, it also would determine the mL of an Eagle’s Minimal Essential Medium (MEM; Nakalai
optimum zinc content for clinical applications. Tesque, Inc. Kyoto) containing 15% fetal bovine serum (JRH
Biosciences, Lenexa, KS) and antibiotics (100 units/mL of
There are two stem cell lineages in marrow cell
penicillin, 100 ␮g/mL of streptomycin, and 0.25 ␮g/mL of
populations, namely, the hematopoietic and the stro-
amphotericin B, Sigma Chemicals Co., St. Louis, MO). Cul-
mal, which give rise to committed progenitors of dif- tures were maintained for 1 week until confluence in a
ferent cell lines, including osteogenic cells.7 Stromal humidified atmosphere of 95% air and 5% CO2 at 37°C.
stem cells, later termed mesenchymal stem cells, are The confluent cells were released from their culture sub-
part of cultured cells adherent to the culture dish strata using 0.25% trypsin. After trypsinization, the cells
surface.8,9 Treated with dexamethasone (Dex), these were seeded at 20 ⫻ 103 per 16-mm-␾ Falcon tissue well
adherent cells can show osteoblastic differentiation with and without ceramic disks for subculture. Just before
that results in mineralization (in vitro bone forma- cell seeding, five different disks (ZnTCP/HAP; TCP/HAP
tion).10 The mineral consists of poorly crystallized hy- containing 0, 0.32, 0.63, 0.88, or 1.26 wt % Zn) were placed in
droxyapatite that is similar to bone mineral. the wells.
The subcultures were performed with 1 mL of the stan-
This culture technique is useful for determining the
dard medium supplemented with 10 mM of ␤-glycerophos-
effects of various factors on osteogenesis; it is a means
phate (Merck Japan Co., Tokyo) and 82 ␮g/mL of vitamin C
of observing biomaterials/bone tissue interactions.11 phosphate (Wako Pure Chemical Industries, Osaka) with or
And when human marrow cells are used, the culture without 10 nM of Dex (Sigma Chemicals Co.) for 2 weeks.
can be used in clinical situations.9 In the present study, The culture medium was changed three times a week. The
Dex-treated rat and human bone marrow cells (BMCs) culture conditions essentially were the same as those re-
were cultured on ZnTCP/HAP disks with a ported by Maniatopoulos et al.,10 with minor modifications.
(Ca⫹Zn)/P ratio of 1.60 and varying zinc contents in The 2-week subculture with Dex showed extensive miner-
order to clarify the effect of ZnTCP/HAP on their alization (in vitro bone formation).11
osteogenic differentiation.

Human bmc preparation and culture

MATERIALS AND METHODS
Bone marrow was obtained from five human donors of
various ages (mean age, 49.4 years; range 11–70 years) with
Fabrication of zntcp/hap composite ceramics informed consent. Three milliliters of bone marrow were
harvested from the ilium of each donor by needle aspira-
ZnTCP with 10 mol % zinc [Ca2.7Zn0.3(PO4)2] was precip- tion12 and added to 3 mL of heparinized phosphate-buffered
itated by an aqueous method using calcium hydroxide ob- saline (PBS) in a tube. The mixture then was centrifuged at
tained from high-purity calcium carbonate (99.99 wt %, Ube 900 rpm for 5 min. About 3 mL of the supernatant with fat
Materials, Ube, Japan), reagent-grade phosphoric acid (85 wt layer was discarded, and then the remaining supernatant
%, Nakalai Tesque, Kyoto, Japan), and reagent-grade zinc (buffy coat and red blood cell layer) was poured into T-75
nitrate hexahydrate (98 wt %, Kanto Chemical Co., Tokyo, flasks containing 15 mL of the above-mentioned culture
Japan). The precipitate was obtained, dried at 100°C and medium and incubated for 10 days to 2 weeks until conflu-
calcined at 850°C for 3 h. The ZnTCP powder thus obtained ence.13
BONE FORMATION ON ZnTCP/HAP CERAMICS 1117

The confluent cells were released from their culture sub- antibiotics) supplemented with 10 mM of ␤-glycerophos-
strata using 0.25% trypsin. After trypsinization, the cells phate and 82 ␮g/mL of vitamin C phosphate. The medium
were seeded at 10 ⫻103 per 16-mm-␾ Falcon tissue well. was changed to a fresh one on days 2, 4, 7, 9, 11, and 14. At
Some experiments also were performed using a cell density each medium change, the supernatant of the old culture
of 10 ⫻103 per 16-mm-␾ well. To determine the optimum medium was obtained, mixed with 12N of hydrochloric acid
concentration of Dex for osteogenesis in human BMC cul- solution at a volume ratio of 2% and analyzed for zinc
ture, subcultures were performed with 1 mL of the standard concentration using an atomic absorption spectrometer (AA-
medium supplemented with 10 mM of ␤-glycerophosphate, 610S; Shimadzu Co., Kyoto).
82 ␮g/mL of vitamin C phosphate, and 10 or 100 nM of Dex
for 2 weeks. As controls, cultures without Dex also were
performed.
To determine the effect of the ZnTCP/HAP ceramics, Effect of zinc on in vitro bone formation by rat and
subcultures at a cell density of 10 ⫻ 103 per 16-mm-␾ well human bmcs
were performed with 1 mL of the standard medium supple-
mented with 10 mM of ␤-glycerophosphate and 82 ␮g/mL The rat and human primary cultured cells were seeded at
of vitamin C phosphate with or without 100 nM of Dex for 20 ⫻ 103 and 10 ⫻ 103 cells per 16-mm-␾ Falcon tissue well,
2 weeks. The 2-week subculture showed obvious mineral- respectively, for subculture without ZnTCP/HAP disks. The
ization (in vitro bone formation).13 The ceramics were placed subculture was performed with 1 mL of the standard me-
in the medium as described in the above-mentioned rat dium supplemented with 10 mM of ␤-glycerophosphate, 82
BMC culture. ␮g/mL of vitamin C phosphate, 10 nM (in rat subculture) or
100 nM (in human subculture) of Dex, and 100 or 500 ␮M of
ZnCl2 (a zinc concentration of 6.5 or 32.7 ␮g/mL, respec-
tively).
Assessment of alkaline phosphatase (alp) activity After 2 weeks of the subculture, ALP activity was mea-
in cultures of rat and human bmcs on zntcp/hap sured as described above. Then calcium was extracted from
ceramics
the mineralized areas (in vitro bone) with 0.5 mL of 20%
formic acid for 1 week at 4°C. An aliquot of the formic acid
Alp staining of bmcs extract was diluted with strontium solution. Calcium con-
centration was determined using an atomic absorption spec-
trometer.
Subcultured rat and human cell layers on the disks were
washed twice with PBS, then fixed with 0.4% paraformalde-
hyde/PBS for 10 min. They then were rinsed with 56 mM of
2-amino-2-methylpropandiol (AMP) buffer (pH 9.9) and
stained with 0.5 mg/mL of naphthol-AS-MX phosphate so- RESULTS
dium salt and 0.5 mg/mL of fast red violet B in 56 mM of
AMP buffer for 10 min. After staining, the disks were rinsed Bmc cultures (primary culture)
with tap water.
Consistent cell growth was observed in the rat BMC
Measurement of alp activity primary culture, which reached confluence at 1 week.
In contrast, the extent of cell growth varied in human
Subcultured rat and human cell layers on the disks were
BMC primary culture and 10 days to 2 weeks of cul-
washed twice with PBS. A small amount (0.2 mL) of 0.2% ture was needed to reach confluence.12 Under the
Nonidet P-40 (NP-40) containing 1.0 mM of MgCl2 was primary culture conditions, the medium (standard
added to the culture well, and the mixture was incubated at medium) did not contain Dex and ␤-glycerophos-
4°C for 1 h. The supernatant was assayed for ALP activity phate, which were added in the subsequent subcul-
using p-nitrophenylphosphate as the substrate in 56 mM of ture conditions, under which the rat/human BMCs
AMP buffer (pH 9.9) containing 1.0 mM of MgCl2, as de- showed a fibroblast-like shape and did not show min-
scribed previously.11 ALP activity was defined as the eralized areas even after confluence.
amount in ␮mol of p-nitrophenol released after a 30-min
incubation at 37°C.

In vitro bone formation by subcultured bmcs

Elution test of zinc ions from zntcp/hap with 1.26
wt % zn Rat BMCs subcultured at a cell density of 20 ⫻ 103
per 16-mm-␾ well in the presence of 10 nM of Dex
The elution test of zinc ions was performed under subcul- showed many macroscopic mineralized areas (nod-
ture condition without cells. ZnTCP/HAP disks with 0 and ules). Human BMC subcultured at the same density
1.26 wt % Zn were placed in the wells and immersed in 1 mL grew rapidly and frequently peeled off from the cul-
of the standard medium (MEM containing 15% FBS and ture dish; therefore we utilized a cell density of 10 ⫻
1118 IKEUCHI ET AL.

Alp staining of subcultured bmcs on culture dishes

and zntcp/hap ceramics

After 2 weeks of rat BMC subculture, many macro-

scopic ALP-positive areas were observed in the entire
area of the culture dishes or ZnTCP/HAP disks in the
presence of Dex (Fig. 2). The intensity of staining for
ALP increased as zinc content of the disk increased.
No cytotoxic effects were found up to a zinc content of
1.26 wt %. However, in the absence of Dex, the cells on
the culture dish or ceramics showed weak staining for
ALP.
After 2 weeks of human BMC subculture, many
macroscopic areas stained for ALP were likewise ob-
served on each culture substratum in the presence of
Dex (Fig. 3). The ALP-positive area increased relative
to the zinc content of the disk, and the most expansive
area indicative of ALP positivity was observed on the
ZnTCP/HAP disk of 1.26 wt % Zn. Almost the entire
surface of the disk was covered with ALP-positive
areas.

Alp activity of the subcultured bmcs on zntcp/hap

ceramics

To quantify osteoblastic differentiation, ALP activi-

ties of the rat and human cell layers were measured. In
the absence of Dex, ALP activities on the ZnTCP/HAP
disks remained very low. Figure 4 shows the effect of
the zinc content of the disk on ALP activities in the
presence of Dex.
By regression analysis, the ALP activities were
found to increase with increasing zinc content of the
disk in a linear correlation. The data demonstrate that
Figure 1. Phase contrast microscopy of human BMCs after ZnTCP/HAP influences cell differentiation in rat and
a 2-week culture (A) in the absence of Dex, and (B) in the human BMC cultures, resulting in high osteoblastic
presence of 10 nM of Dex or (C) 100 nM of Dex. Arrows activity, and that ZnTCP/HAP with 1.26 wt % Zn
indicate the mineralized areas (original magnification ⫻40). stimulates osteoblastic differentiation at the highest
level. It is important to note that the ALP activity of
103 per 16-mm-␾ well for human BMC subculture on both rat and human cells showed dependence on Zn
ZnTCP/HAP ceramics. content, as seen in this quantitative analysis as well as
After 2 weeks of human BMC subculture (at a cell in the ALP stain analysis.
density of 10 ⫻ 103 per 16-mm-␾ well), phase contrast
microscopy of the culture showed more mineralized
areas in the presence of 100 nM of Dex than in the
presence of 10 nM of Dex (Fig. 1). Also, ALP activity
was significantly higher in the presence of 100 nM of Release of zn ions from ceramic disks
Dex than in the presence of 10 nM of Dex (data not
shown). These results indicate that 100 nM of Dex is As shown in Figure 5, ZnTCP/HAP with 1.26 wt %
optimal for osteogenic differentiation leading to in Zn consecutively released zinc ions in the medium
vitro bone formation in human BMC culture. In all without the cells. The initial release of zinc ions from
human BMC cultures, bone mineralization was ob- the ceramic disks during the 2 days after soaking the
served in the presence of 100 nM of Dex. Therefore 100 disk in the medium was considerable, and zinc con-
nM of Dex were used as an osteogenic factor in human centration increased up to about 7.2 ␮g/mL. The
BMC culture on the ZnTCP/HAP disks. amount of zinc released gradually decreased; how-
BONE FORMATION ON ZnTCP/HAP CERAMICS 1119

Figure 2. (A) Alkaline phosphatase (ALP) staining of rat BMCs on Falcon tissue well after a 2-week culture in the presence
(⫹) or absence (⫺) of Dex. (B) ALP staining of rat BMCs on ZnTCP/HAP disks after a 2-week culture in the presence of Dex.
Zn: 0, Zn: 0.32, Zn: 0.63, Zn: 0.88, and Zn: 1.26 indicate the cell cultures on TCP/HAP, ZnTCP/HAP with 0.32 wt % Zn;
ZnTCP/HAP with 0.63 wt % Zn; ZnTCP/HAP with 0.88 wt % Zn; and ZnTCP/HAP with 1.26 wt % Zn, respectively.

ever, zinc concentration was maintained at 2.2 ␮g/mL creased ALP activity, the calcium content was signif-
for 14 days. icantly lower in the rat BMC culture in the presence of
100 ␮M of ZnCl2 than in the absence of ZnCl2. There-
fore, 100 ␮M of ZnCl2 inhibited bone matrix formation
Effect of the addition of zn ions in the rat BMC culture.
on subcultured bmcs In contrast, no inhibition was observed in the hu-
man BMC culture at 100 ␮M of ZnCl2 (Fig. 7). In the
In both rat and human BMCs, the ALP activities presence of 500 ␮M of ZnCl2, the ALP activities and
were significantly higher in the presence of 100 ␮M of calcium contents in both rat and human BMCs were
ZnCl2 than in the absence of ZnCl2 (Fig. 6). This zinc extremely low. Microscopic examination of the culture
concentration corresponds to that of the culture me- revealed a very low number of cells, probably due to
dium that contained the disk of ZnTCP/HAP with the cytotoxic effect of the high zinc concentration. The
1.26 wt % Zn in the initial 2 days. Despite the in- cytotoxicity suppressed cell division, resulting in low

Figure 3. (A) ALP staining of human BMCs on Falcon tissue wells after a 2-week culture in the presence (⫹) or absence (⫺)
of Dex. (B) ALP staining of human BMCs on ZnTCP/HAP disks after a 2-week culture in the presence of Dex. Zn: 0, Zn: 0.32,
Zn: 0.63, Zn: 0.88, and Zn: 1.26 indicate the cell cultures on TCP/HAP; ZnTCP/HAP with 0.32 wt % Zn; ZnTCP/HAP with
0.63 wt % Zn; ZnTCP/HAP with 0.88 wt % Zn; and ZnTCP/HAP with 1.26 wt % Zn, respectively.
1120 IKEUCHI ET AL.

content of 1.2 wt %, which is the same zinc content

required for the maximum proliferation of MC3T3-E1
cells, as reported in a previous study.4 ZnTCP/HAP
with 1.26 wt % Zn released zinc into the medium at
concentrations between 2.2 and 7.2 ␮g/mL for 14 days
(Fig. 5). The addition of 100 ␮M of ZnCl2 (6.5 ␮g/mL)
to the culture medium (instead of placing the ZnTCP/
HAP disk with 1.26 wt % Zn in the medium) also
significantly increased the ALP activity of the BMCs
relative to those cultured without ZnCl2.
Even if the addition of free zinc ions causes an effect
similar to the addition of ZnTCP/HAP, it is possible
that the zinc in ZnTCP/HAP is transported directly to
cells through the cell membrane attached to the
ZnTCP/HAP disk. Whatever the route of the zinc
ions, the superiority of ZnTCP/HAP over TCP/HAP
in supporting osteogenic differentiation is because of
the presence of zinc in the ceramic. In this regard, zinc
is known to enhance protein synthesis14 and to in-
crease the activity of 1␣,25-dihydroxyvitamin D3-de-
pendent promoters.15
Zinc has also been reported to have a stimulatory
effect on bone formation and mineralization in osteo-
blastic cells because of activation of aminoacyl-tRNA
synthetase and stimulation of cellular protein synthe-
sis.16 This suggests that Zn affects various protein
functions, which results in stimulatory effects on os-
teogenic differentiation of BMCs.
Although 100 ␮M of zinc in the culture medium
caused the increase in ALP activity, it also caused the
significant decrease in the calcium content of the min-
eralized areas produced by rat BMCs [Figs. 6(A) and
7(A)]. However, 10 ␮M of zinc caused an increase in
both ALP activity and calcium content (data not
shown).
Figure 4. Dose-response effect of zinc content of ZnTCP/
HAP disks on ALP activities. (A) Rat and (B) human BMCs These findings indicate that 100 ␮M of zinc are not
were cultured on disks in the presence of Dex for 2 weeks
and used for ALP measurements. The data showed good
correlation between ALP activity and zinc content of the
ZnTCP/HAP disks for both rat and human cell layers.

osteoblastic activity, and thus there was no obvious

mineralization.

DISCUSSION

Both rat and human BMCs cultured on ZnTCP/

HAP showed dependence on Zn content for the in-
crease of ALP activity, as seen in the quantitative
analyses as well as in the ALP stain analyses (Figs.
2– 4). These findings demonstrate the superiority of Figure 5. Temporal elution profile of Zn released from the
ZnTCP/HAP over TCP/HAP in supporting osteo- ZnTCP/HAP disks with 1.26 wt % Zn immersed in the
standard medium (MEM ⫹ 15% FBS ⫹ antibiotics) supple-
genic differentiation of rat and human BMCs in the mented with 10 mM of ␤-glycerophosphate and 82 ␮g/mL
presence of Dex. of vitamin C phosphate. Each value is the mean ⫾ SEM of
The maximum ALP activity was observed at a zinc four disks.
BONE FORMATION ON ZnTCP/HAP CERAMICS 1121

mechanism of this stability against zinc remains to be

clarified.
We have reported that the mineral composition of
the in vitro bone fabricated by BMCs is poorly crystal-
lized hydroxyapatite.11 Poorly crystallized hydroxy-
apatite also can be constructed in chemically defined
solutions, and thus apatite formed at physiologic min-
eral concentration absorbs or incorporates zinc to a
concentration as high as 0.4 wt %.17,18
It also is known that calcium in bone is exchanged
for zinc.19 Therefore it was confirmed that zinc re-
leased from ZnTCP/HAP is consumed during osteo-
genic differentiation and zinc can be incorporated in
the surface layer or in the crystal lattice of hydroxy-
apatite of the in vitro bone produced by BMCs. Al-
though zinc can be incorporated into the bone mineral
phase, as described above, zinc is known as a potent

Figure 6. Effect of zinc supplementation on ALP activity of

(A) rat and (B) human BMC cultures. Rat and human BMCs
were cultured in the presence of Dex for 2 weeks and used
for ALP measurements. ZnCl2 at 100 or 500 ␮M (6.5 or 32.7
␮g/mL, respectively) was added to the medium. For the
control, ZnCl2 was not added to the medium. Each value is
the mean ⫾ SEM of four wells. Significance levels are rela-
tive to the control [multiple comparison test, (Fisher’s
PLSD); *p ⬍ 0.001].

suitable for inducing bone mineralization for rat

BMCs. Thus there is a possibility that the extent of
mineralization was reduced in rat BMC culture on
ZnTCP/HAP with 1.26 wt % Zn. In contrast, 100 ␮M
of zinc caused a significant increase in both the ALP
activity and calcium content of human BMCs and,
consequently, stimulated in vitro bone formation [Figs.
6(B) and 7(B)].
Although the plasma zinc concentrations in both rat
Figure 7. Effect of zinc supplementation on calcium con-
and human are almost the same, the zinc concentra- tent in (A) rat and (B) human BMC cultures. The rat and
tion for enhancing osteogenic response differs among human BMCs were cultured in the presence of Dex for 2
species. Moreover, it is noted that human BMCs tol- weeks and calcium was extracted from the cell layers using
erate the inhibitory effect of zinc on mineralization formic acid solution. ZnCl2 at 100 or 500 ␮M (6.5 or 32.7
even at a concentration much higher than that for rat ␮g/mL, respectively) was added to the medium. For the
control, ZnCl2 was not added to the medium. Each value is
BMCs. The species-dependent factor would function the mean ⫾ SEM of four wells. Significance levels are rela-
effectively in clinical applications of ZnTCP/HAP be- tive to the control [multiple comparison test, (Fisher’s
cause of the stability of human BMCs against zinc. The PLSD); *p ⬍ 0.001].
1122 IKEUCHI ET AL.

inhibitor for matrix vesicle formation during the 5. Kawamura H, Ito A, Muramatsu T, Miyakawa S, Ochiai N,
growth of plate cartilage.20 –22 Tateishi T. Long-term implantation of zinc-releasing calcium
phosphate ceramics in rabbit femora. J Biomed Mater Res
Moreover, it has been reported that zinc can inhibit 2003;65A:468 – 474.
the crystal growth of apatite in vitro.23–25 These reports 6. Hall SL, Dimai HP, Farley JR. Effects of zinc on human skeletal
may explain the present results that 100 ␮M of zinc alkaline phosphatase activity in vitro. Calcif Tissue Int 1999;
caused the significant decrease in calcium content in 64:163–172.
rat BMC culture while it increased ALP activity. How- 7. Owen M. Lineage of osteogenic cells and their relationship to
the stromal system. In: Peck WA, editor. Bone and mineral
ever, 100 ␮M of zinc increased the calcium content as research, Vol. 3. Amsterdam: Elsevier Science; 1985; p 1–25.
well as ALP activity in the culture of human BMCs. 8. Caplan AI. Mesenchymal stem cells. J Orthop Res 1991;9:641–
All the above data showed the stimulatory effect of 650.
ZnTCP/HAP ceramics on osteogenic differentiation of 9. Ohgushi H, Caplan AI. Stem cell technology and bioceramics:
BMCs, resulting in in vitro bone formation on the From cell to gene engineering. J Biomed Mater Res 1999;48:
913–927.
ceramics. It is significant that physicochemical and 10. Maniatopoulos C, Sodek J, Melcher AH. Bone formation in
biochemical analyses have shown that the bone vitro by stromal cells obtained from bone marrow of young
formed in vitro is the counterpart of bone formed in adult rats. Tissue Res 1988;254:317–330.
vivo. 11. Ohgushi H, Dohi Y, Katuda T, Tamai S, Tabata S, Suwa Y. In
We also have reported that the bone formed in vitro vitro bone formation by rat marrow cell culture. J Biomed
Mater Res 1996;32:333–340.
on calcium phosphate ceramics exhibits continuous 12. Ohgushi H, Okumura M. Osteogenic ability of rat and human
osteogenic activity after in vivo implantation of the marrow cells in porous ceramics. Acta Orthop Scand 1990;61:
ceramics.9,26 Furthermore, not only rat BMCs but also 431– 434.
human BMCs when combined with ceramics exhibit 13. Yoshikawa T, Ohgushi H, Uemura T, Nakajima H, Ichijima K,
an osteogenic response.12,13 Therefore, the in vitro Tamai S, Tateishi T. Human marrow cells-derived cultured
bone in porous ceramics. Biomed Mater Eng 1998;8:311–320.
bone derived from rat BMCs as well as from human 14. Yamaguchi M, Oishi H, Suketa Y. Zinc stimulation of bone
BMCs on ZnTCP/HAP ceramics could survive after in protein synthesis in tissue culture. Activation of aminoacyl-
vivo implantation and has clinical significance for ap- tRNA synthetase. Biochem Pharmacol 1988;37:4075– 4080.
plication as an osteogenic bone graft substitute. 15. Lutz W, Burritt MF, Nixon DE, Kao PC, Kumar R. Zinc in-
creases the activity of vitamin D-dependent promoters in os-
teoblasts. Biochem Biophys Res Commun 2000;271:1–7.
16. Yamaguchi M. Role of zinc in bone formation and bone resorp-
tion. J Trace Elem Exp Med 1998;11:119 –135.
CONCLUSIONS 17. Smeyers-Verbeke J, Michotte Y, Pelsmaeckers J, Lowenthal A,
Massart DL, Dekegel D, Karcher D. The chemical composition
of idiopathic nonarteriosclerotic cerebral calcifications. Neurol-
This study reveals that the surface of ZnTCP/HAP
ogy 1975;25:48 –57.
ceramics stimulates osteogenic differentiation in both 18. Hamilton EI, Minski MJ, Cleary JJ. The concentration and
rat and human BMCs. Because of the broad range of distribution of some stable elements in healthy human tissues
zinc concentrations able to induce osteogenic stimula- from the United Kingdom: An environmental study. Sci Total
tion in human BMCs relative to that in rat BMCs, Environ 1972/1973;1:341–374.
19. Samachson J. Mechanism for the exchange of the calcium in
ZnTCP/HAP ceramics are expected to be safe and
bone mineral. Nature 1967;216:193–194.
useful materials for bone reconstructive surgery. 20. Sauer GR, Wu LN, Iijima M, Wuthier RE. The influence of trace
elements on calcium phosphate formation by matrix vesicles.
J Inorg Biochem 1997;65:57– 65.
21. Litchfield TM, Ishikawa Y, Wu LNY, Wuthier RE, Sauer GR.
References Effect of metal ions on calcifying growth plate cartilage chon-
drocytes. Calcif Tissue Int 1998;62:341–349.
1. Yamaguchi M, Inamoto K, Suketa Y. Effect of essential trace 22. Kirsch T, Harrison G, Worch KP, Golub EE. Regulatory roles of
metals on bone metabolism in weanling rats: Comparison with zinc in matrix vesicle-mediated mineralization of growth plate
zinc and other metals’ actions. Res Exp Med 1986;186:337–342. cartilage. J Bone Miner Res 2000;15:261–270.
2. Yamaguchi M, Oishi H, Suketa Y. Stimulatory effect of zinc on 23. Kanzaki N, Onuma K, Treboux G, Tustsumi S, Ito A. Inhibitory
bone formation in tissue culture. Biochem Pharmacol 1987;36: effect of magnesium and zinc on crystallization kinetics of
4007– 4012. hydroxyapatite (0001) face. J Phys Chem B 2000;104:4189 – 4194.
3. Hashizume M, Yamaguchi M. Stimulatory effect of ␤-alanyl-l- 24. LeGeros RZ. Calcium phosphates in oral biology and medi-
histidinato zinc on cell proliferation is dependent on protein cine. New York: Karger; 1991.
synthesis in osteoblastic MC3T3-E1 cells. Mol Cell Biochem 25. De Groot K. Bioceramics of calcium phosphate. Boca Raton,
1993;122:59 – 64. Florida: CRC Press; 1983.
4. Ito A, Ojima K, Naito H, Ichinose N, Tateishi T. Preparation, 26. Yoshikawa T, Ohgushi H, Tamai S. Immediate bone forming
solubility, and cytocompatibility of zinc-releasing calcium capability of prefabricated osteogenic hydroxyapatite. J
phosphate ceramics. J Biomed Mater Res 2000;50:178 –183. Biomed Mater Res 1996;32:481– 492.