You are on page 1of 7

Calcif Tissue Int (1983) 35:755-761

Calcified Tissue
,~ 1983by Springer-Verlag

Influence of Magnesium Supplementation on Bone Turnover in the

Normal Young Mouse

Pierre J. Marie, Rose Travers, and Edgard E. Delvin

Genetics Unit, Shriners Hospital and the Department of Experimental Surgery, McGillUniversity, Montreal, Canada

Summary. The effect of magnesium (Mg) sup- sorption in the young mouse, and that the stimu-
plementation on bone metabolism has been studied lation of bone mineralization was the result of in-
in the normal young mouse. Weanling male mice creased extracellular mineral availability. On the
were given Mg-supplemented drinking water (5 mM other hand, stimulation of osteoclastic bone resorp-
or 32 mM Mg) for 4 weeks. Mineral and skeletal tion appeared to occur independently of PTH or of
changes were assessed by biochemical methods and increased 1,25(OH)2D production. Therefore, this
by histomorphometric analysis of endosteal bone study suggests that Mg may control in v i v o bone
formation and resorption parameters evaluated on metabolism by directly influencing bone resorbing
tetracycline double-labeled, undecalcified caudal cells activity.
vertebrae. Magnesium supplementation increased Key words: Magnesium - - Mouse - - Mineralization
serum and urinary Mg concentrations and bone Mg --Resorption - - Osteoclasts - - Bone turnover.
content. Both the calcification rate and the extent
of tetracycline double-labeled osteoid surface in-
creased progressively in Mg-treated mice, whereas
the mineralization lag time was shortened. The os- Several abnormalities in bone structure and bone
teoblastic surface was reduced, leading to a fall in turnover have been described in Mg-deficient ani-
osteoid surface. Stimulation of bone mineralization mals [ 1 - 8 ] . M g - d e p l e t e d rats show r e t a r d e d
was associated with a rise in extracellular calcium skeletal growth [1], decreased bone formation [2]
(Ca) and phosphorus (P) concentrations whereas and mineralization [3, 4], and reduced bone resorp-
serum 25-OHD and 1,25(OH).,D levels remained nor- tion [4-6]. Though these observations suggest that
mal. The Mg supplementation increased the number the presence of Mg affects bone formation and re-
of acid phosphatase stained chondroclasts and osteo- sorption in rodents, its role in the regulation of bone
clasts and the extent of resorbing surface showing metabolism remains uncertain. The decreased bone
histochemically stained osteoclasts. Although uri- turnover observed in Mg-deficient animals has been
nary OH-proline increased above normal, Ca, P, and ascribed to reduced responsiveness of the skeleton
cyclic adenylic acid (cAMP) excretion and phosphate to PTH [7-9]. On the other hand, several studies
concentration (TmP/GFR) remained normal, sug- have shown that dietary Mg supplementation may
gesting that parathyroid hormone (PTH) secretion produce significant changes in Ca and P metabolism
was not altered. The trabecular bone volume re- in the rat in the absence of parathyroid glands [1, 10,
mained normal, showing that the increased bone re- 11]. In addition, indirect evidences have been pre-
sorption was balanced by the stimulated bone sented showing that Mg supplementation can in-
mineralization. The results show that Mg supple- crease bone resorption in v i v o [12] independently of
mentation influenced both bone formation and re- PTH secretion [10, 13]. These observations sustain
the contention that Mg can directly influence bone
metabolism. However, no attempt has been made to
Send offprint requests to P. J. Marie, Shriners Hospital. 1529 verify this hypothesis. Therefore, we have exam-
Cedar Avenue, Montreal, P.Q., Canada H3G 1A6 ined, by quantitative histological methods, the ef-
756 P.J. Marie et al.: Effect of Magnesium Supplementation on Bone

fects of Mg supplementation on bone resorption and rapid response to biological stimuli [19]. The seventh caudal
f o r m a t i o n in t h e n o r m a l y o u n g m o u s e . vertebra was embedded in methylmethacrylate for routine bone
morphometry [19]. Five micron thick sections of the central por-
tion of the vertebra were stained with toluidine blue and 15/zm
thick sections were mounted unstained for fluorescence micros-
Materials and Methods copy. The eight caudal vertebra was embedded in glycolmethac-
rylate at 4~ and sectioned longitudinally for enzyme histochem-
Animals istry. Five micron thick sections were stained histochemically
for acid phosphatase using naphtol ASTR as substrate [20] and
Normal male C57BL/6J mice were randomly divided at weaning counterstained with hemalum. Active osteoclasts and chondro-
(21 days of age) into three groups. The first (n = 16) and second clasts were identified as multinucleated cells showing a strong
(n = 15) groups were given deionized water, ad libitum, contain- acid phosphatase activity characterized by a bright red cytoplasm.
ing 5 and 32 mM MgClz, respectively. The third (n = 17) received The histologic sections were quantitatively analyzed using a
the same water without additional MgCL_. The animals, housed in Zeiss 100 point integrating eyepiece and a semiautomatic image
plastic cages, were maintained throughout the experiment on a analyzer (MOP 3, Carl Zeiss Inc., W. Germany). Bone growth
Purina Mouse Chow containing 0.50% Ca, 0.74% P, and 0.15% was assessed by measuring the total vertebral length [19] on the
Mg. given ad libitum. The food consumption was similar in all methylmethacrylate-embedded sections. The trabecular bone
three groups and the Mg intake in the food was about 3 mg volume was calculated as the percentage of endosteal bone and
Mg/day/mouse. Since the daily water intake averaged 4 ml/ marrow tissue occupied by osteoid, calcified cartilage, and
mouse, the total Mg ingested by each mouse amounted approxi- mineralized bone [2l]. The osteoid surface Ipercentage of total
mately to 3.5 and 6.1 mg/day for the first and second groups, endosteal surface covered by an osteoid seam) and the osteo-
respectively, and 3.0 rng/day for the third. All animals were in- blastic surface (percentage of total endosteal surface showing
jected intraperitoneally on the 45th and 48th days of age with plump osteoblasts) were measured on the endosteal surface of
tetracycline (Achromicin, 30 mg/kg body weight) and were sac- the vertebral diaphysis. The endosteal mean osteoid seam thick-
rificed on the 50th. ness (MOST) was determined by measuring the average width of
at least 30 randomly selected intercepts along the osteoid sur-
face. The endosteal calcification rate (CR) was determined as the
Biochemical Analysis mean distance (in microns) between the midpoints of the two
fluorescent tetracycline labels in all the double-labeled zones di-
On the day preceding sacrifice, mice were housed individually in vided by the interval of time (3 days) between the two labelings.
metabolic cages under fasting conditions and urine was collected The percentage of endosteal osteoid surface exhibiting a double
for 16 h overnight. At sacrifice, blood was drawn by cardiac fluorescent label (double labeled surface) was also determined.
puncture. Serum and urine samples were frozen at -20~ until The mineralization lag time (MLTI was evaluated as the ratio of
use. Calcium phosphorus, magnesium, alkaline phosphatase and the MOST over the CR [22]. The endosteal bone formation rate
creatinine concentrations were determined using modified Auto (BFR) was calculated by multiplying the double labeled surface
Analyzer II R procedures (Technicon Instruments, Tarrytown, by the calcification rate [22].
NY). Urinary cAMP was determined by a competitive binding On the glycolmethacrylate-embedded sections, the following
assay (Amersham Searle, Arlington Heights, I11.). Urinary hy- variables were measured separately on the metaphyseal region
droxyproline (OH-pro) was measured colorimetrically by the extending from the primary to the secondary spongiosa, and on
method of Ramamurthy et al. [14]. The renal threshold for the diaphyseal area [23]: the percentage of total surface uncov-
TmP/GFR was evaluated using the nomogram of Walton and ered by osteoid (calcified surface), the percentage of calcified
Bijvoet [15]. surface showing resorption lacunae filled with characteristic
Serum vitamin D metabolite concentrations were determined acid-phsophatase stained osteoclasts (osteoclastic surface), and
in pooled samples (n = 3-5) from control and treated mice. the number of osteoclasts per mm 2 of bone section. The number
Serum 25-hydroxyvitamin D (25-OHD) levels were measured by of h i s t o c h e m i c a l l y stained c h o n d r o c l a s t s p r e s e n t at the
a modified radioligand assay [16]; those of 1,25-dihydroxyvita- growthplate-primary spongiosa junction was determined and ex-
rain D [I,25(OH)2D] were determined using a modified com- pressed as number of cells per mm z of area.
petitive binding assay [17, 18] with inter- and intraassay varia- Individual histological values are the mean of measurements
tions not exceeding 10%. made on 3 - 4 nonadjacent sections. Normal biochemical, his-
tomorphometric, and histochemical data were obtained under
identical conditions to the control group.
The results were expressed as mean • SD. and probabilities of
Skeletal Analysis
difference between means of normal and treated animals were
Tibias. which are mainly composed of cortical bone in the calculated by Student's t test.
mouse, were freed of bone marrow, dried to constant weight at
100~ and then ashed at 600~ for 8 h. Bone ashes were dis-
solved in 3 N hydrochloric acid. Calcium and phosphorus con- Results
centrations were measured colorimetrically, using an Auto
Analyzer II after dilution in deionized water. Magnesium was
Biochemical Studies
measured by atomic absorption spectrophotometry in aliquots
diluted in a lanthanum oxide solution.
The proximal half of the tail was fixed in cold neutral formal- Serum calcium levels, compared with controls,
dehyde, dehydrated in alcohol, and embedded, undecalcified, in w e r e s l i g h t l y i n c r e a s e d o n l y i n t h e 32 m M M g
plastic material. The tail vertebra has been chosen because of its s u p p l e m e n t e d g r o u p ( T a b l e 1). H o w e v e r , u r i n a r y
R J. Marie et al.: Effect of M a g n e s i u m Supplementation on Bone 757

Table 1. Effects of Mg s u p p l e m e n t a t i o n on biochemical vari a bl e s in the normal mous e

Controls 5 mM Mg 32 mM Mg
(n = 17) (n = 16) (n = 15)

Serum calciu m (mg/dl) 9.3 • 0.4 a 9.5 • 0.3 9.6 • 0.4 b

Serum p h o s p h o r u s (mg/dl) 7.5 • 0,9 8.5 • 1.2 b 8.3 • 1.2 h
Serum m a g n e s i u m (mg/dt) 2.8 • 0.1 3.0 • 0.2 b 2.9 • 0.3
Serum alkaline p h o s p h a t a s e (IU/I) 150 • 26 173 • 28 176 +- 38
Urinary calciu m tmg/mg creatinine) 0.13 • 0.05 0.16 -+ 0.05 0.11 • 0.03
Urinary p h o s p h o r u s (mg/mg creatinine) 6.1 • 0.8 6.1 • 1.8 7.0 +- 1.9
T m P / G F R (mg/dl) 5.7 • 1.3 6.4 • 0.9 5.8 • 1.1
Uriuary m a g n e s i u m (mg/mg creatinine) 0.81 • 0.20 1.13 • 0.10 h 1,47 • 0.25 b'~
Urinary OH-proline (/zg/mg creatinine) 153 • 35 217 • 55 b 206 • 65 b
Urinary c A M P (nM/mg/creatinine) 74.7 • 16.3 63.6 • 8.9 81.7 • 11.1

a Mean • SD
Significantly different from controls (P < 0.05 or better level of significance by t test)
c Significantly different from 5 mM Mg-treated mice (P < 0.001 by t test)

Table 2. Serum v itamin D m e t a b o l i t e c o n c e n t r a t i o n s in Mg- Table 3. Bone mineral c ont e nt in the n o r m a l mous e given Mg
s u p p l e m e n t e d normal mice supplementation

25-OHD 1,25(OH)zD a Controls 5 mM Mg 32 mM Mg

ng/ml pg/ml (n = 13) (n = 13) (n = 13)

Controls 34 • 9 a 36 ~ 1 Bone ash

5 mMMg 31 • 5l • 12 (% dry bone) 66.6 _+ 1.5 a 67.7 • 1.5 67.0 • 1.2
32mMMg 27• 32 • 2 Bone calcium
(mg/gash) 365 • 364 • 375 •
Mean • S D o f n = 3 5 pooled samples Bone phos phorus
(rag/gash) 190 - 3 182 • 189 •
Bone m a g n e s i u m
calcium excretion remained unchanged. Both the
(mg/g ash) 7.4 -+ 0.2 7.4 *_ 0.3 7.9 • 0.4 b,c
low (3.5 mM) and high (32 mM) Mg supplementa-
tion resulted in increased serum phosphorus levels. a Me a n +_ SD
There was, however, no modification of either the Significantly different from controls (P < 0.0l by t test)
urinary phosphorus excretion or the renal reab- Significantly different from 5 mM Mg-treated mice (P < 0.01
by t test)
sorption. Serum Mg rose in the 5 mM supplemented
animals without further increase for the 32 mM
group. However, the magnesuria rose progressively
according to the level of supplementation. Serum treated mice (Fig. 1). Both the extent of osteoid and
the osteoblastic surface were p r o g r e s s i v e l y de-
alkaline phosphatase was not modified by Mg
creased (Table 4). Accordingly, the osteoblastic
supplementation whereas urinary OH-Pro was in-
surface, expressed in percentage of osteoid surface,
creased at the two Mg dosages. No significant
remained unchanged and not different from controls
change in cAMP excretion was observed.
Serum 25-OHD levels and 1,25(OH)2D concen- (52.3 • 8.2% and 58.3 • 16.6% in the 5 and 32 mM
Mg-treated mice, respectively compared with 53.6
trations were not significantly different in Mg-
• 8.0% in controls, NS). The mean osteoid seam
supplemented mice compared with controls (Table 2).
thickness was also progressively decreased (Fig. 2)
as a result of increasing calcification rate (Table 4).
Skeletal Analysis The combination of these two factors resulted in a
marked and progressive shortening of the minerali-
At either level, Mg supplementation had no effect zation lag time (Fig. 2). The percentage of total en-
on bone ash content and on bone Ca and P contents dosteal surface exhibiting a double tetracycline
measured in whole bone ash of tibias (Table 3). On labeling remained unchanged in the 5 and 32 mM
the other hand, bone Mg content in whole tibias was Mg-treated mice compared with controls (20.5 --+
increased above normal at the highest Mg dosage. 6.1% and 17.2 • 3.3%, respectively vs 19.6 _ 1.3%,
Neither the vertebral length, nor the trabecular NS). H o w e v e r , because of the fall in osteoid sur-
bone volume were altered by Mg supplementation face, the extent of osteoid with double label in-
(Table 4). On the other hand, marked changes in creased progressively with Mg dosage (Fig. 2). The
bone formation parameters were observed in Mg- combined rise in CR and in the double labeled os-
758 P . J . Marie et al.: Effect of Magnesium Supplementation on Bone

Fig. 1. Histological appearance of the diaphyseal area in the seventh caudal vertebra of a control mouse (A) and a 32 mM Mg
supplemented age-matched mouse (B). Note the reduced osteoid surface (arrowheads) accompanied with a decreased extent of bone
surface lined by osteoblasts (arrows) in the treated mouse. Undecalcified, toluidine blue stained 5 ~ m section, • 100.

Table 4. Effects of Mg supplementation on histological parameters of bone formation in the normal mouse

Controls 5 mM Mg 32 mM Mg
(n = 14) (n = 13) (n = 14)

Vertebral length (mm) 3.60 • 0.15 a 3.62 • 0.09 3.63 • 0.08

Trabecular bone volume
(% bone tissue) 13.7 • 2.2 13.0 _+ 1.8 13.1 • 1.7
Osteoid surface
(% endosteal surface) 46.3 _+ 4.0 29.3 _+ 8.6 b 21.0 • 6.0 b,~
Osteoblastic surface
(% endosteal surface) 24.8 • 4.4 15.3 -+ 5.2 b 12.0 -+ 4.4 b
Calcification rate (~m/day) 2.07 • 0,13 2.19 • 0.11 2.29 • 0.09 b'~

Mean _+ SD
b Significantly different from controls (P < 0.01 or better level of significance by t test)
Significantly different from 5 mM Mg-treated mice (P < 0.05 or better level of significance by t test)

teoid surface resulted in a marked elevation of the

endosteal bone formation rate.
[ ] Controls
< Magnesium supplementation produced significant
[ ] 5 mM Mg %~ changes in bone resorption parameters (Fig. 3). At
t ~ 3 2 mM Mg .~ . ~ --100 ~
the m e t a p h y s e a l level, the two doses of Mg
~lO-- .-rL markedly increased the number of acid-phosphatase
0 Z
v _~o5 stained chondroclasts and osteoclasts (Table 5).
Only the highest dose of Mg increased the osteo-
clastic surface in both the metaphyseal and diaph-
JJifi o
yseal regions. In the latter area, the osteoclast
number was increased above normal at the largest
%% ae Mg dosage.

,iiiii: E


.~5 2- u_ z~
The results of this study show that Mg supplemen-
tation influences both bone formation and resorp-
tion in the young mouse. The slight rise in serum Mg
Fig. 2. Histological parameters of bone formation in controls and and the elevation of urinary Mg demonstrates that
Mg supplemented mice. Data are expressed as mean • SD. Sig-
nificant differences with controls are indicated by one star above
Mg intake was increased in the Mg-treated mice. In
the bars (P < 0.001). The two stars denote a significant difference contrast to urinary Mg, however, serum Mg con-
with the group of 5 mM Mg-treated mice (P < 0.01) and with centrations remained relatively stable in face of in-
controls (P < 0.001). creased Mg intake. Therefore, as previously ob-
E J. Marie et al.: Effect of Magnesium Supplementation on Bone 759

Fig. 3. Histological appearance of the metaphyseal area in the eight caudal vertebra of a control mouse (A) and a 32 mM Mg
supplemented age-matched mouse (B). Note the higher n u m b e r of acid phosphatase stained chondroclasts (arrowheads) resorbing
cartilage (C) and the increased osteoclast n u m b e r (arrows) along the calcified bone surface (CB) in the treated mouse. Undecalcified, acid
phosphatase stained and hemalum counterstained 5/xm section, • 158.

Table 5. Effects of Mg supplementation on histological parameters of bone resorption in the normal mouse

Metaphysis Diaphysis

Calcified Osteoclastic Calcified Osteoclastic

Chondroclast surface surface Osteoclast surface surface Osteoclast
no/mm 2 % total % calcified n o / m m2 % total % calcified no/mm 2
bone section surface surface bone section surface surface bone section
(n = 7) 42.1 _+ 3.7 a 39.2 --_ 3.6 22.0 • 3.3 53.4 • 9.5 37.8 _+ 5.5 12.4 • 1.8 10.8 • 1.6
5 mM Mg
(n = 13) 74.1 • 10.6 b 47.9 • 6.3 b 22.2 • 3.3 75.5 • 21.4 b 35.4 • 5.0 13.6 • 3.7 9.2 -+ 2.0
32 mM Mg
(n = 14) 83.0 -+ 6.5 b,c 43.4 --_ 4.3 b 26.4 • 2.7 b,c 72.7 _+ 12.3 b 37.2 • 3.2 20.6 • 4.5 b,c 13.5 • 3.2 b,~

a Mean • SD
b Significantly different from controls (P < 0.05 or better level of significance by t test)
~ Significantly different from 5 mM Mg-treated animals (P < 0.02 or better level of significance by t test)

served [23], urinary Mg excretion appears to be a length and may not reflect overall changes in long
better index of increased Mg loading than serum Mg bone or skeletal growth in Mg-treated animals.
concentrations in our animals. Our finding that Nevertheless, it must be noticed that the vertebral
bone Mg content was unchanged or increased in length has been shown to be an adequate and sensi-
whole long bones after Mg supplementation corro- tive parameter of bone growth in the mouse, since it
borates previous observations in Mg-supplemented responds rapidly to mineral and hormonal stimuli
rats [1, 24]. In contrast to Mg-supplemented rats, [19, 21, 23]. On the other hand, there are several
however [1], we have not observed significant histological evidences that endosteal bone minerali-
changes in bone Ca and P content in tibias of Mg- zation was stimulated in Mg-treated mice. Both the
treated mice. It must be pointed out, however, that calcification rate and the extent of osteoid with
we have used the whole tibia to determine bone double tetracycline labeling were increased, show-
mineral content. It is possible that changes similar ing that the rate and activity of individual
to those reported [1] occurred in cancellous bone mineralizing sites were s t i m u l a t e d after Mg
but not in cortical bone, since the former responds supplementation. The reduction of osteoid thick-
more rapidly to Mg supplementation than the ness and the shortening of MLT indicate that the
latter [6]. calcification rate exceeded the rate of matrix syn-
As judged by the lack of change in vertebral thesis in Mg~tveated animals. On the other hand, the
length, Mg supplementation did not appear to influ- decreased osteoid surface appears to result, in part,
ence bone growth of the tail in our mice. However, from the concomitant reduction in osteoblastic
this observation pertains to the caudal vertebra population. The combination of increased rate of
760 P.J. Marie et al.: Effect of MagnesiumSupplementationon Bone

calcification and the presumed reduced bone matrix diaphysis, possibly because Mg was more rapidly
synthesis can account for the observed decline in incorporated at the epiphyseal than at the diaph-
amount of osteoid. Interestingly, Mg supplementa- yseal level [6, 31 ]. It appears, therefore, that the rise
tion stimulated bone mineralization without in- in serum Ca and P concentrations resulted from
creasing the percentage of osteoid with plump os- stimulated bone resorption. The unchanged trabec-
teoblasts and serum alkaline phosphatase levels. On ular bone volume indicates, however, that the in-
the other hand, stimulation of bone mineralization creased bone resorption was balanced by the ob-
under 1,25(OH)2D administration in young mice of served stimulated new bone mineralization. The
the same strain is associated with an elevation of the lack of urinary Ca and P wasting in face of stimu-
fractional osteoblastic surface and a concomitant lated bone resorption also indicates that the min-
rise in alkaline phosphatase activity [21 ]. It appears, erals resorbed were recycled for new bone miner-
therefore, that in contrast to 1,25(OH)2D, Mg alization.
supplementation influenced bone mineralization in- Our findings that Mg supplementation produced a
dependently of changes in osteoblastic recruitment. marked stimulation of bone resorption in the mouse
The lack of change in serum alkaline phosphatase corroborates previous indirect observations in the
concentration also indicates that, although Mg is rat [12, 13, 32] and in man [33]. The mechanism by
required for normal alkaline phosphatase activity which Mg supplementation stimulated bone resorp-
[25], Mg supplementation does not stimulate this tion, however, remains unclear. In our Mg-treated
enzymatic function above normal. mice, the changes observed cannot be explained by
Although Mg is known to influence in vivo bone an effect of 1,25(OH).,D on bone, since circulating
mineralization [1, 26], this cation appears to inhibit levels of this metabolite remained unchanged. Al-
rather than stimulate the initiation of calcification though serum PTH levels could not be measured in
[27, 28], presumably because of competition be- the present study, several indications tend to show
tween Mg and Ca for molecular sites [29]. There- that PTH secretion has not been increased. Neither
fore, differences in tetracycline labeling in Mg- cAMP excretion nor TmP/GFR were significantly
treated mice is probably not related to Mg-tetra- altered after Mg supplementation, indicating that
cycline complexing at the mineralization front, and PTH effect on kidney r e m a i n e d unchanged.
the observed stimulation of the calcification rate is Moreover, the observed rise in serum Ca would
unlikely to result from a direct effect of Mg on bone tend to suppress rather than stimulate PTH secre-
mineralization. Alternatively, increased bone min- tion.
eralization may be the consequence of the ob- Finally, elevated serum Mg concentrations have
served rise in serum mineral concentrations. been reported to suppress PTH secretion in vitro
Increased Ca and P extracellular levels can result [34] and in vivo [35-37]. Therefore, increased bone
from stimulated intestinal absorption, reduced renal resorption is unlikely to result from stimulated PTH
excretion, or increased bone mobilization of miner- secretion. On the other hand, it is possible that Mg
als. The first hypothesis appears unlikely, since acted by promoting the skeletal responsiveness to
high dietary Mg has been reported to inhibit the PTH, as previously observed in Mg-deficient ani-
intestinal absorption of Ca and P [13, 30]. The lack mals [38]. However. this is not supported by the
of changes in serum 1,25(O H)2D levels also suggests previously reported observation that high dietary
that the intestinal Ca and P absorption was not Mg can increase serum Ca in absence of parathy-
stimulated. Possible improvement of renal Ca and P roid glands [10]. We then postulate that the ob-
conservation in Mg-treated mice is not supported by served stimulation of bone resorption resulted from
the lack of changes in urinary Ca and P excretion a direct stimulatory effect of Mg on bone resorbing
and by the stable TmP/GFR. On the other hand, cells. Since Mg is a cofactor of several enzymatic
both biochemical and histological data indicate that reactions, its influence on bone cells may not be
bone resorption has been stimulated by Mg specific. The precise mechanism by which Mg acted
supplementation. on chondroclast and osteoclast number and activity
Urinary OH-proline was increased parallel with remains to be determined.
urinary concentrations of Mg. In addition, the In summary, this study shows that Mg supple-
number of chondroclasts and osteoclasts was in- mentation produces significant changes in bone
creased, suggesting that Mg supplementation stim- formation and resorption in the normal young mouse.
ulated the recruitment of new bone resorbing cells. Bone mineralization was increased as a result of the
The extent of bone surface exhibiting acid- rise in serum concentrations of Ca and P. The latter
phosphatase stained osteoclasts was also increased, finding seems to result from stimulated osteoclastic
indicating that Mg supplementation stimulated os- bone resorption which appears to occur indepen-
teoclastic activity. Bone resorption parameters dently of PTH or of increased 1,25(OH).,D produc-
were more increased in the metaphysis than in the tion. The results indicate that Mg can directly in-
E J. Marie et al.: Effect of Magnesium Supplementation on Bone 761

fluence bone resorbing cells activity and suggest of 1.25-dihydroxyvitamin D in human plasma. A r c h
that this cation may be an important factor con- Biochein Biophys 176:235-243
18. Delvin EE, Glorieux FH (1981) Serum 1,25-dihydroxyvita-
t r o l l i n g in vivo b o n e m e t a b o l i s m .
rain D concentration in hypophosphatemic vitamin D-resis-
tant rickets. Calcif Tissue Int 33:173-175
19. Marie PJ, Travers R, Glorieux FH (1982) Bone response to
Acknowledgments. We would like to express our thanks to Mrs. phosphate and vitamin D metabolites in the hypophos-
Johanne Bibeau and Rejeanne Desjardins for their excellent phatemic male mouse. Calcif Tissue Int 34:158-164
technical assistance and to Denyse Lavall6e for her expert se- 20. Evans RA, Dunstan CR, Baylink DJ (1979) Histochemical
cretarial work. This investigation was supported by the Shriners identification of osteoclasts in undecalcified sections of
of North America. human bone. Mineral Electrolyte Metab 2:179 185
21. Marie PJ, Travers R (in press) Continuous infusion of 1,25-
dihydroxyvitamin D3 stimulates bone turnover in the normal
young mouse. Calcif Tissue Int
References 22. Baylink D, Stauffer M, Wergedal J, Rich C (1970) Forma-
tion, mineralization, and resorption of bone in vitamin D-
1. Clark I, Belanger L (1967) The effects of alterations in di- deficient rats. J Clin Invest 49:1122-1134
etary magnesium on calcium, phosphate and skeletal metab- 23. Marie PJ, Travers R (in press) Effects of magnesium and
olism. Calcif Tissue Res 1:204-218 lactose supplementation on bone metabolism in the X-linked
2. Jones JE, Schwartz R, Krook L (1978) Bone changes in hypophosphatemic mouse. Metabolism
magnesium deficiency. Fed Proc 37:667A 24. Kreusser WJ, Kurokawa K, Aznar E, Sachtjen E, Massry
3. Schwartz R, Reddi AH (1979) Influence of magnesium de- SG (1978) Effect of phosphate depletion on magnesium
pletion on matrix-induced endochondral bone formation. homeostasis in rats. J Clin Invest 61:573-58l
Calcif Tissue Int 29:15-20 25. Heaton FW (1965) Effect of magnesium deficiency on
4. Jones JE, Schwartz R. Krook L (1980) Calcium homeostasis plasma alkaline phosphatase activity. Nature 18:1292-1293
and bone pathology in magnesium deficient rats. Calcif Tis- 26. Boskey AL (1978) The role of c a l c i u m - p h o s p h o l i p i d -
sue Int 3l:231-238 phosphate complexes in tissue mineralization. Metab Bone
5. MacManus J, Heaton F (1969) The effect of magnesium defi- Dis Rel Res 1:137-142
ciency on calcium h o m e o s t a s i s in the rat. Clin Sci 27. Nielsen SP (1973) Effects of magnesium on calcification of
36:297-306 young bone in tissue culture. Calcif Tissue Res 11:78-94
6. Rayssiguier Y, Larvor P (1978) Mineral bone composition 28. Tomazic B, Thomson M, Nancollas GH (1975) Growth of
and some elements of calcium metabolism in magnesium- calcium phosphates on hydroxyapatite crystals: the effect of
deficient growing rats. Ann Biol Anim Biochim Biophys magnesium. Arch Oral Biol 20:803-808
18:157-166 29. Boskey AL, Posner AS (1980) Effect of magnesium on
7. Levi J, Massry S. Coburn J, Llach F, Kleeman C (1974) lipid-induced calcification: an in vitro model for bone
Hypocalcemia in magnesium depleted dogs. Evidence for mineralization. Calcif Tissue Int 32:139 - 143
reduced responsiveness to PTH and relative failure of 30. Hodgkinson A, Marshall DH, Nordin BEC (1978) Vitamin D
parathyroid gland function. Metabolism 23:323-345 and magnesium absorption in man. Clin Sci 57:121-123
8. Freitag J J, Martin K J, Conrades MB, Bellorin-Font E, 31. Field AC (1961) Studies on magnesium in ruminant nutrition.
Teitelbaum S, Klahr S, Slatopolski E (1979) Evidence for 3. Distribution of 28Mg in the gastrointestinal tract and tis-
skeletal resistance to parathyroid hormone in magnesium sues of sheep. Brit J Nutr 15:349-359
deficiency. J Clin Invest 64:1238- 1244 32. MacManus J, Heaton FW (1970) The influence of mag-
9. Forbes RM, Parker HM (1980) Effect of magnesium defi- nesium on calcium release from bone in vitro. Biochim
ciency on rat bone and kidney sensitivity to parathyroid Biophys Acta 215:360-367
hormone. J Nutr 110:1610-1617 33. Briscoe AM, Ragan C (1966) Effect of magnesium on cal-
10. Clark I (1969) Effects of magnesium on calcium and phos- cium metabolism in man. Amer J Clin Nutr 19:295- 306
phate metabolism in parathyroidectomized rats. Endocri- 34. Targovnik JH, Rodman JS, Sherwood LM (1971) Regulation
nology 85:1103 - l 113 of parathyroid hormone secretion in vitro: quantitative as-
11. Clark I, Rivera-Cordero F (1973) Effect of endogenous pects of calcium and magnesium ion control. Endocrinology
parathyroid hormone on calcium, magnesium and phosphate 88:1477-1482
metabolism in rats. Endocrinology 92:62-71 35. Buckle RM, Care AD, Cooper CW, Gitelman HT (1968) The
12. Clark I, Geoffroy R (1958) Studies in calcium metabolism. J influence of plasma magnesium c o n c e n t r a t i o n on
Biol Chem 233:203-205 parathyroid hormone secretion. J Endocrinol 42:429-431
13. Levine BS, Walling MW. Coburn JW (1981) Effect of vita- 36. Mayer GP (1975) Effect of calcium and magnesium on
min D sterols and dietary magnesium on calcium and phos- parathyroid secretion rate in calves. In: Talmage RV. Owen
phorus homeostasis. Am J Physiol 241:E35-E41 M, Parsons JA (eds) Proceedings of the 5th Parathyroid
14. Ramamurthy NS, Zebrowski EJ, Golub LM: A modified Hormone Conference. Calcium Regulating Hormones. Ex-
procedure for determining hydroxyproline in normal and cerpta Medica, Amsterdam, pp 122-124
diabetic urine. Biochem Med 5:555-558 37. Massry SG, Coburn JW, Kleeman CR (1970) Evidence for
15. Walton RJ, Bijvoet OLM (1975) Nomogram for derivation of suppression of parathyroid gland activity by hypermag-
renal threshold phosphate concentration. Lancet 2:309-310 nesemia. J Clin Invest 49:1619-1629
16. Delvin EE, Dussault M, Glorieux FH (1980) A simplified 38. Estep H, Shaw WA, Watlington C, Hobe R, Holland W,
assay for 25-hydroxycalciferol. Clin Biochem 13:106-108 Tucker St. G (1969) Hypocalcemia due to hypomagnesemia
17. Eisman JA, Hamstra AJ, Kream BE, DeLuca HF (1976) A and reversible parathyroid hormone unresponsiveness. J
sensitive, precise and convenient method for determination Clin Endocrin Metab 29:842-848