JOURNAL OF BONE AND MINERAL RESEARCH

Volume 6, Number 4,1991

Mary Ann Liebert, Inc., Publishers

A Systemic Acceleratory Phenomenon (SAP) Accompanies

the Regional Acceleratory Phenomenon (RAP) During
Healing of a Bone Defect in the R.at

MARTINA MUELLER, TOBIAS SCHILLING, HELMUT W. F4INNE,

and REINHARD ZIEGLER

ABSTRACT
The rate of remodeling in the region of a bone defect exceeds normal tissue activity. It was Frost who de-
scribed this reaction as a regional acceleratory phenomenon (RAP). We investigated the local healing process
with rats with a burr hole defect (1.2 mm in diameter) in the left tibia. We differentiated an initial phase of
bone formation followed by a phase of predominant resorption. To determine whether this regional en-
hancement of bone formation would result in a systemic impact on bone metabolism, we analyzed both
tibiae and femora and the fourth lumbar vertebra. On day 7 both femora of rats with the tibial defect
showed a significant increase in computerized x-ray density, dry weight, ash weight, and Ca2+content. Both
tibiae and the fourth lumbar vertebra showed a significant increase in mineralizing surface, mineral apposi-
tion rate, and bone formation rate. Because of these results we conclude that a systemic acceleratory phe-
nomenon (SAP) accompanies the RAP.
SAP affects only the cancellous, but not the cortical bone compartment. SAP is associated closely with
the occurrence of woven bone during the formation phase of the healing process. Thus we assume that
woven bone formation plays a pivotal role in the mediation of SAP.

INTRODUCTION the rate of remodeling in the region of a bone defect ex-

ceeds the ordinary tissue activity to potentiate tissue heal-

B ECAUSE they may reveal insights in the understanding

of bone growth, repair, and remodeling processes, the
control and stimulation of bone formation and bone re-
sorption have received a great deal of attention.
There is ample evidence that both bone formation and
resorption are regulated by a complex interplay between
systemic hormones and an increasing number of local fac-
tors,“) but the effects of systemic and local growth factors
on bone formation and resorption have been examined al-
most exclusively by use of in vitro systems.(z-6)Very little
is known about this interplay between hormones and sys-
temic and local growth factors in the in vivo regulation of
bone formation and resorption.
We applied and further characterized an animal model
of maximally stimulated local bone formation and resorp-
tion under in vivo c~nditions.‘’-~) According to Frost,(1o)
ing and local tissue defense reactions. This “reaction of
mammalian tissues to diverse noxious stimuli” was termed
the regional acceleratory phencimenon (RAP) by Frost. In
the present study we examined whether acceleration of re-
gional remodeling activities leads to an impact on systemic
bone remodeling. We induced RAP by a burr hole defect
in the rat tibia and characterized the local healing process.
Associated changes in growth and remodeling dynamics at
distant skeletal sites were analyzed as well, since our previ-
ous results pointed toward syslemic effects of local repair
mechanisms.(L1)There have also been reports of an en-
hancement of osteogenesis in rat mandibular condyles fol-
lowing tibial marrow ablation.(”’ Our study comprises the
analyses of weight-bearing (femora and tibiae) as well as of
so-called non-weight-bearing structures (vertebrae).

Department of Medicine I, Endocrinology, University of Heidelberg, Bergheimerstrasse 58, D-6900 Heidelberg, Germany.

401
402 MUELLER ET AL.

MATERIALS AND METHODS of this method is that the density of the x-rayed object is
Animals directly proportional to the intensity of the x-ray shadow.
For densitometric measurement the x-ray film is positioned
Female rats (strain Chbb: THOM, Thomae GmbH, under a photomicroscope equipped with a high-resolution
Biberach, Germany) weighing 180-200 g were maintained video camera that displays the x-ray shadow of the bone
at random in cages (Ehret, Inc., D7830 Emmendingen, on a video monitor. This video image is then digitized by
Germany) and received food and water ad libitum. The means of threshold discriminators. Approximately 60 gray-
diet was a standard laboratory chow [Atromin standard scale values can be discriminated. The instrument trans-
324 (0.9% calcium; 0.7% phosphorus; 0.2070 magnesium; forms the gray-scale value of each picture point (pixels) be-
0.2% sodium; 1.0% potassium; 600 IU/kg of vitamin D3) longing to the bone image into a brightness value. The sum
Altromin Co., Lage, Germany]. of the brightness values (termed the hydroxylapatite area
coefficient, ha/mmz) in a defined area of measurement is
Experimental design directly proportional to the bone density in this region of
Two experiments were carried out. The first studied the i n t e r e ~ t . " ~ . ' ~ )
healing pattern of an experimentally induced defect in the
rat tibia. We investigated whether this local healing would Physical Analyses: Bone volume: Both femora were re-
follow a predictable series of events that could be quanti- moved and completely cleaned of adherent soft tissue. The
fied and thus be used as an appropriate in vivo model to bones were weighed and their volumes determined accord-
study bone formation under defined conditions. A hole 1.2 ing to Archimedes.('5)
mm in diameter was drilled in the diaphysis of the left tibia
Dry weight: For determination of their dry weight the
of female rats. There were six groups of 10 animals each.
femora were dried to constant weight at 37°C.
The animals were sacrificed on days 3, 7, 10, 14, 21, and
28 under ether anesthesia by aortal bleeding and cervical Mean specific density: The ratio of bone dry weight to
dislocation. We used a triple-fluorochrome labeling the total bone volume was determined.
method to demonstrate the dynamics of healing.
Ash weight: The femora were ashed in a muffle furnace
The second studied whether this local enhancement of
(Heraeus MR 170, W.C. Heraeus Co., Hanau, Germany)
bone formation by the experimental defect would affect
at 600°C for 24 h.
other sites of the skeleton or even lead to a systemic bone
formation stimulus. There were three groups of animals. Calcium content: Ash specimens were dissolved with
One group ( n = 12) was designated as the control group. 32% (1.16 M) HCI and diluted 1:40,0oO for A A S (atomic
The second group underwent a sham operation ( n = 12). absorption spectroscopy, Perkin-Elmer 4ooo).
A hole was drilled in the diaphysis of the left tibia of the
rats of the third and experimental group (n = 14). The Triple-Fluorochrome Labeling: In experiments 1 and 2
animals received injections of calcein before the operation the animals received intraperitoneal (IP) calcein (6 mg/kg
and of tetracycline before sacrifice on day 7. This date was body weight per day) 24 h before surgery and IP tetracy-
chosen after the results of our experiment 1 (data shown cline (30 mg/kg body weight per day) 24 h before sacrifice.
below) as the date on which local enhancement of bone In addition IP xylenol orange (90mg/kg body weight per
formation was at its peak. Both femora, both tibiae, and day) was injected 1 week before sacrifice in the 14, 21, and
the fourth lumbar vertebra were removed for analyses. All 28 day groups.'16-'9)
the bones examined were x-rayed to perform computerized
x-ray densitometry. The femora were then used for physi- Histomorphometry: The tibiae and fourth lumbar verte-
cal analysis, the tibiae and vertebrae for quantitative histo- bra were removed from each animal and fixed in 70% eth-
morphometric analysis. anol. The bones were dehydrated in successive concentra-
tions of ethanol and embedded in methyl methacrylate.(zO)
Experimental defect
For visual morphometric analysis [Merz grid'z'.zz)]of
Rats were anesthesized with ether, and a skin incision the tibiae and vertebrae, 3 pm sections were cut with a
was made parallel to the long axis of the medial tibial di- Jung microtome (Jung Co., Heidelberg, Germany) and
aphysis. A defect 1.2 mm in diameter was drilled from the stained according to van K o s ~ a . ' ~In~addition
) to the Merz
medial to the lateral side of the midshaft cortex (the defect grid method we applied a computerized histomorphometry
extending from one cortical side to the other throughout method for the van Kossa stains.
the marrow cavity). A sodium chloride solution (0.9%) To prepare sections of the defect-bearing tibia we chose
was used to clean the area of debris. The incision was a standardized cutting plane parallel to the long axis of the
closed with two sutures. In the sham procedure the bone bone and to that of the defect.
was exposed and was left intact. The fluorescent labels were estimated by means of hand-
ground sections (approximate thickness 50 pm) using a
Zeiss Morphomat. The following measurements were car-
Analyses
ried out at x 50 in vertebral cancellous bone and in tibial
Computerized X-ray Densitometry: X-ray films were metaphyseal area of secondary spongiosa, 1 mm distal to
taken of all the examined bones. The underlying principle the epiphyseal growth plate: bone volume (Vo), bone sur-
SYSTEMIC ACCELERATORY PHENOMENON
face (mm2/mm3), mineralizing surface (MS: single-labeled
surface = sL, double-labeled surface = dl, Vo), mineral
apposition rate (MAR, pm/day), bone formation rate
(BFR, tissue level, pm3/pm2/day).
The histomorphometric parameters were determined ac-
cording to the ASBMR Histomorphometry Nomenclature
Committee.
Tibia1 growth was determined by measuring the distance
of the fluorescent band from the epiphyseal line.'25)
Statistical Analysis: We applied MANOVA (multivari-
ate analysis of variance) to calculate statistical
significance.'26) We used the SPSS-X software system
(SPSS Inc., 44 N. Michigan Avenue, Chicago, IL 60611)
to analyze the data obtained for femora, tibiae, and verte-
brae.
The precondition for the application of the MANOVA
was nonsignificance of the Box's M test. As the Box's M
test for the vertebrae showed a significance of 0.001, we
had to exclude the vertebrae from the MANOVA. Thus
statistical significance for the vertebrae was calculated by
means of Wilcoxon's test.
We applied an a adjustment to integrate the statistical
tests for femora, tibiae, and vertebrae. The significance
level a, of the individual tests must be <0.017 (a,= 1 -
(1 - p)l/n; n = number of individual tests) to obtain an
overall significance level p < 0.05.
RESULTS
There was no difference in body weight between control
and defect-bearing rats at the end of the experiment. There
was also no enhancement of tibial growth in rats with a
burr hole defect. The data obtained for our sham group
corresponded with those of the control group, suggesting
that the sham procedure did not provoke alterations of
bone metabolism (data therefore not shown).
Experiment 1
The healing pattern of a local defect in the rat tibial mid-
diaphysis followed a predictable series of events. Determi-
nation of the x-ray density in the defect region by means of
computerized x-ray analysis showed a first phase of activa-
tion of bone formation followed by a phase of predomi-
nant resorption (data not shown).
Our histologic examinations (50 pm sections with triple-
fluorochrome labeling) revealed that the initial formation
phase consists of
1. A complete filling of the defect with woven bone
throughout the marrow cavity (medullary callus forma-
tion; Color Plate l).
2. An additional enhancement of bone formation not
only at this site of local defect but also at the lateral peri-
osteal envelope of the tibia within a certain distance of the
burr hole defect (periosteal bone neoformation; Color
Plate 2).
On day 14 bridging of the interrupted cortical ends was
achieved by newly formed woven bone, whereas the med-
403
ullary callus underwent almost complete resorption (Color
Plate 3). On day 21 lamellar bone began to replace the
bridging woven bone (Color Plates 4 and 5). On day 28
lamellar bone formation in the restored cortical zone con-
tinued and led to an increase in cortical bone density at the
defect area (Color Plate 6 ; for quantitative data see later).
Computerized histomorphometry of the defect area (by
means of 3 pm sections stained according to van Kossa)
confirmed the time-dependent formation and resorption
phase described earlier.
In our histomorphometric analyses we additionally dif-
ferentiated between the medullary callus and the bridging
bone at the cortical site, which 'we quantitated separately.
Figure 1 gives an overview of our measurement areas in the
defect-bearing tibia. Figure 2 shows the bone volume in the
medullary cavity (area 1 in Fig. l), which reached a peak
of 40% at day 7 (also compare Color Plate l), and then the
resorption of this medullary callus subject to temporal lim-
itations. The volume of the cortical bridging bone (area 2
in Fig. 1) increased to reach 850,'o on day 28 after surgery,
along with its transformation into lamellar bone (Fig. 3).
Bone formation at the periosteal site revealed a peak at
day 10 postsurgery and was also followed by a phase of
predominant resorption (area 3a in Fig. 1). Thus forma-
tion and resorption of the bone at the periosteal site paral-
leled the formation and resorption phase of woven bone in
the medullary cavity (Fig. 2 shows the development of
medullary woven and periosteal bone).
However, this newly formed periosteal bone was not re-
sorbed entirely: approximately 20% became an integral
part of the cortical envelope. The mineralized volume of
the remaining 20% of this periosteal callus rose continu-
ously and thus showed a dynamic profile similar to that of
the cortical bridging bone (Fig. 3).
Experiment 2
Femora: X-ray Density + Physical Analyses: On day 7,
when local repair achieved a maximum of newly formed
bone, both femora of rats with the tibial defect showed an
increase in computerized x-ray density, dry weight, ash
weight, and Ca'+ content (see Table 1).
The multivariate test, which is the overall test for all re-
peated measurements (i.e., computerized x-ray density, dry
weight, ash weight, and Ca2+content) showed a significant
increase between the experimental and the control group
(p < 0.015). The individual univariate tests, which focus
on the difference between the control and the experimental
group with respect to the single repeated measurement,
showed a significant increase in the experimental group
(see Table 1). There was no within-subject difference; that
is, there was no difference in tht: reaction of right and left
femora of defect-bearing rats.
Tibiae: Histomorphometric Data: Mineral apposition
rate, bone formation rate, and (double- and single-labeled
mineralizing surface of both tibiae (area 4, Fig. 1) of de-
fect-bearing rats were increased (Fig. 4). The effects as
measured by the dynamic parameters, single- and double-
labeled MS, MAR, and BFR were more marked in the de-
404 MUELLER ET AL.

fect-bearing left tibia, but the contralateral tibia responded
in the same way.
The multivariate test showed a significant increase in
both experimental tibiae with respect to all repeated mea-
surements, that is, MS sL, MS dL, and MAR, that were
integrated by MANOVA (p < 0.001). Each of these indi-
vidual repeated measurements showed a significant in-
crease in the experimental group in the univariate test.
There was no within-subject effect in the multivariate test:
the right and left tibiae of defect-bearing rats reacted in the
same way. Only the individual test for MS SL showed a sig-
nificant increase in the left defect-bearing tibia compared
to the right tibia (see Fig. 4).
Vertebrae: Histomorphometry: The mineral apposition
rate, bone formation rate, and single- and double-labeled
mineralizing surface in vertebral cancellous bone of rats
with the tibia1 defect were increased (Fig. 5). The Wilcoxon
test for the BFR of the fourth lumbar vertebra of defect-
bearing rats was significant, with p < 0.005.
a-Adjustment: The significance levels for the femora
(multivariate test, p < 0.015), tibiae (multivariate test, p
< 0.001), and vertebrae (Wilcoxon test, p < 0.005) were
all below the a-adjusted level of a0 = 0.017.Thus there is
a significant systemic osteogenic stimulus in the repair pro-
cess of a local bone defect.

a r e a of histomorphometric results
measurement parameters shown

1 bone volume fig.3

(%)

2 bone volunie sig.4

( 010 I

3a mean width of callus fig.3

(pm)

3b bone volume of frg.4

mineralized callus
(%)

4 mineralizing surface sL fig.5

(oh)
4 mineralizing surface dL r1g.5
(%)

4 mineral apposition r a t e S1g.S

(um/d)
4
5 Y
bone To m a t ' n r a t e
(urn /urn /d)
fig.5

5 = position of the pretreatment fluoro-

chrome label a t the time of
microscopic evaluation

FIG. 1. Measurement areas in the defect-bearing tibia: 1, medullary area; 2, cortical area; 3, periosteal area; 4, subepi-
physeal area; 5 , pretreatment fluorochrome label, 1 mm proximal to area 4.

COLOR PLATE 1. Day 7: Longitudjnal section of the defect-bearing tibia parallel to the long axis of the defect.
COLOR PLATE 2. Day 7: Note the woven bone filling the defect and the periosteal proliferation on the lateral side of
the tibia.
COLOR PLATE 3. Day 14: Note the resorption of the medullary callus and the bone formation at the cortical site of
the defect.
COLOR PLATE 4. Day 21: Transformation of the cortical bridging bone into lamellar bone.
COLOR PLATE 5. Day 21: Detail of the occurrence of lamellar bone (double labels) at the cortical site.
COLOR PLATE 6. Day 28: Note the increase in density of the cortical briding bone and the double labels indicating
lamellar bone formation.
COLOR PLATE 1

COLOR PLATE 2

COLOR PLATE 3

COLOR PLATE 4

COLOR PLATE 5 COLOR PLATE 6

405
406 MUELLER ET AL.

bone volume

i
OIO
YO bone volume
50
80
40
60
30
10
20

10 20

h
3. 7 10. 11. 21. 28. day 3. 7 10. 14. 21. 28. day

I
qm mean thickness Oh bone volume

"1
T

i
300 I T

LL
200

100

3. 7
14 14
FIG. 2. (Top) Bone volume in the medullary cavity (area FIG. 3. (Top) Bone volume of the cortical bridging bone
1 in Fig. l), SD. (Bottom) Quantitation of the perios- at the defect site (area 2 in Fig. I), f SD. (Bottom) Bone
teal bone neo-formation (area 3a in Fig. l), f SD. volume of the periosteal bone neo-formation at the lateral
side of the tibia (area 3b in Fig. 2), * SD.

Our histomorphometric data show that the secondary DISCUSSION

spongiosa in the tibia and the vertebral cancellous bone re-
sponded to the observed systemic stimulus toward bone
formation. We examined the cortical bone compartment to
further investigate if both fractions of bone (cancellous
and cortical) are affected.
We found no significant difference in the vertebral corti-
cal bone volume (cortical BV/TV, (70) between experimen-
tal and control groups. The endocortical mineral apposi-
tion rate, measured in the tibiae as well as in the vertebrae,
also did not show a significant difference (see Table 2). All
histomorphometric changes observed thus far must be at-
tributed to the cancellous bone compartment.
By means of polarized light microscopy the newly
formed bone in the cancellous compartment turned out to
be lamellar bone.
Bab et aI.('2) proposed a generalized enhancement of
bone formation, since they observed increased formation
in rat mandibular condyles during postablation regenera-
tion of tibial bone marrow. They suggest that this increase
in osteogenesis is preferentially associated with postabla-
tion marrow regeneration, but they also reported an in-
crease fourfold lower than after marrow injury after a
periosteal injury and after drilling a hole in the tibial cor-
t e ~ . ( ~ ~A. *similar
') reaction has also been described in hu-
mans, whose serum parameters of bone formation (osteo-
calcin and alkaline phosphatase) were elevated after aspira-
tion of large amounts of marrow.(28)
In our study we focused on the local healing site of a
bone defect, on one hand, and on the other its possible im-
SYSTEMIC ACCELERATORY PHENOMENON 407

TABLE1. PHYSICAL
AND X-RAYDATAOF FEMORA^

Defect-bearing rats
Control rats
Left side
(defect side) Right side Left side Right side

X ray density
Subepiphyseal zone
(distal growth plate), pixels 5130 (457) 5202 (411) b 4648 (407) 4657 (456)
Physical data
Dry weight, mg 377 (12.8) 380 (13.5) b 357 (17.3) 359 (18.5)
Ash weight, mg 249 (9.4) 246 (10.6) c 237 (13.5) 236 (14.5)
CaZ+content, mg 82 (3.5) 83 (5.0) d 78 (3.6) 79 (7.4)
Volume, mm3 403 (18.7) 404 (15.0) 384 (15.5) 398 (26.1)
Density, mg/cm3 935 (46.8) 943 (37.2) 931 (25.0) 906 (75.7)
~~~

aValues in parentheses are standard deviations; defect = drill hole of 1.2 mm diameter in the left tibia1 midshaft
on day 0; femora examined on day 7. There were no significant differences in the influence on the left compared to
the right side of experimental animals (within-subject effect).
bp < 0.005 experimental compared to control group (between-subject effect).

pact on distant skeletal sites. Our results indicate that such
detailed analyses of the local site in relation to the systemic
changes is a precondition for the further characterization
of the osteogenic stimulus. Thus it is also a precondition to
draw further conclusions regarding the (patho)physiologic
regulation of bone formation in vivo.
We characterized the healing of a local bone defect as a
model of accelerated modeling in the sense of a RAP.('O)
The RAP was described by Frost as a phenomenon that in-
volves acceleration of normal activities in both hard and
soft tissues, as an SOS mechanism to potentiate tissue
healing and local tissue defensive reactions.
We distinguished between an initial phase of maximally
stimulated bone formation followed by a phase of predom-
inant resorption in the repair of a local bone defect. This
initial osteogenic phase was characterized by true bone
n e ~ f o r m a t i o n ' ~woven
~): bone not only filled the defect
but also stabilized the lateral periosteal side of the defec-
tive tibia.'3D)Most of the medullary callus(31)then under-
went resorption, whereas the woven bone bridging the cor-
tical ends was transformed into lamellar bone.
In analyzing different sites of the skeleton we found that
restoration of a bone defect not only leads to a RAP but
also leads to a systemic acceleration of osteogenesis (SAP).
This systemic osteogenic stimulus is directed toward
weight-bearing (tibiae and femora) as well as so-called
non-weight-bearing structures (vertebrae). It affects only
the cancellous compartment of bone. The cortex does not
respond.
We stress that that systemic osteogenic stimulus is cou-
pled to the phase during which woven bone is formed lo-
cally and is subject to temporal limitations. Woven bone
formation occurring in fracture and defect repair has been
compared to fetal osteogenesis.(291"During fracture repair,
osteoblast differentiation is very similar to that occurring
at the epiphyseal plate during bone growth or in expenmen-
tal osteogenesis. The signals governing these processes could
be similar to those operating during embryogenesis."(")
We were able to relate the time course of woven bone
neoformation and periosteal proliferation during defect re-
pair with systemic cancellous osteogenesis. The pivotal role
of woven bone neoformation is obvious since an increase
in bone volume of the bridging lamellar bone in our exper-
iment had no osteogenic effect (unpublished data). Frost
stressed the importance of the woven bone and its reactiv-
ity, allowing a rapid adaptation of bone to new situa-
t i o n ~ . ' ~Thus
~ . ~ we
~ ) conclude that the reactivity of woven
bone neoformation not only leads to stabilization of the
local situation but also provides the stimulus for the sys-
temic enhancement of osteogentsis.
We suggest two models, which are in agreement with re-
cently published to explain the origin of this sys-
temic acceleration. Local injury to bone leads to local re-
pair, in the sense of a RAP.
Hypothesis 1: The injury to the bone per se leads to re-
lease of growth factors needed to initiate local repair and
simultaneously evokes systemic Iosteogenesis.
Hjpothesis 2: Local repair, either of bone or of bone
marrow, leads to the release of humoral factors that act
systemically (via the circulation).
Previous results showed that the inhibition of marrow
regeneration by injection of a silicone-based dental impres-
sion material after marrow ablaltion does not lead to an
elevation in bone formation parameters in the rat condylar
subchondral bone.[271If this mechanical inhibition of mar-
row regeneration does not affect the release of growth fac-
tors after injury per se, hypothesis 1 seems not to be rele-
vant. Thus the repair processes of bone or bone marrow
408 MUELLER ET AL.

mineralizing surface mineralizing surface MA A EfR

X I single label % 1 double label

4 sL l * i MS dl

30 I 8
1
20
1 0.6
10

I.r. I.r. 1.r. 1.r. ao5

a2

mineral apposition rate

a
MAR
I 1 rats with tibial defect
0 intact controls
0.6 T FIG. 5. MS sL, MS dL, MAR, and BFR of the fourth
lumbar vertebrae of rats with a tibial defect compared to
0.G intact controls, SD. (a) p < 0.005 experimental com-
pared to control group (Wilcoxon test).
1
0.2
TABLE2. VERTEBRALHISTOLOGIC
DATA^
I. r. I. r. I. r. I. r: Defect- bearing
rats Control rats
I left tibiae with defect Cortical bone volume, (70 18.2 (4.0) 19.2 (3.5)
1right tibiae of tibial defect group Endosteal MAR, @day 3.1 (0.2) 3.1 (0.3)

0 left and right

~

control tibiae aCortical bone volume and endocortical mineral apposition rate
of the fourth lumbar vertebrae of rats with tibial defect compared
to control rats.
FIG. 4. MS sL, MS dL, MAR, and BFR of the secon-
dary spongiosa (area 4 in Fig. 1) of both tibiae; rats with
left tibial defect compared to intact controls, + SD. (a)
p < 0.0015 experimental compared to control group (be- Prostaglandins (PGs) have also been shown to be present
tween-subject effect). (b) p < O.ooOo3 experimental at the site of bone injury.(4o)Several in vivo observations
compared to control group (between-subject effect). (c) p add support to the hypothesis that PGs have a stimulatory
< 0.015 left compared to right tibia (within-subject effect). effect on bone formation.'41) Shih and N~rridin"~)de-
scribed in their study on the effects of prostaglandins on
regional remodeling changes during tibial healing in bea-
gles an enhancement of periosteal bone proliferation in
may be responsible for the release of systemically acting PGE-treated groups. In children with congenital heart dis-
humoral factors. ease infusion of PGE,, used to maintain patency of the
A number of factors could play a role as mediators. Bab ductus arteriosus, evoked striking periosteal new bone for-
et al. partially purified a growth factor derived from heal- mation. (43.44) Thus the periosteal bone proliferation ob-
ing marrow of rat tibia and showed its in vitro growth-pro- served in our experiment could be due to the action of
moting activity to osteogenetic cells. ( 2 5 ) prostaglandins. Because the activation of periosteal bone
Transforming growth factor (TGF-P) may be another formation occurs exclusively on the lateral side of the de-
candidate in the mediation of the systemic e f f e ~ t . ' ~ ~It- ~is' ) fect-bearing tibia, we assume an increased biomechanical
well known that TGF-P and other factors, including fibro- stress on this side. Biomechanical stress is known to stimu-
blast growth factor (FGF), platelet-derived growth factor late prostaglandin production and cell proliferation.'4s'4')
(PDGF), and insulinlike growth factor (IGF), modulate We do not know if systemic growth factors or locally
various aspects of bone formation.'39) synthesized factors are responsible for the mediation of the
SYSTEMIC ACCELERATORY PHENOMENON
systemic enhancement in osteogenesis. However, we think
that the characterization of SAP gives evidence of the pos-
sible in vivo role of various local and systemic growth fac-
tors currently being investigated in vitro.
ACKNOWLEDGMENTS
We thank E.H. Burger (Free University, Amsterdam)
for useful discussion and helpful advice. A.M. Parfitt
(Bone and Mineral Research Laboratory, Henry Ford
Hospital, Detroit) is acknowledged for helpful ideas, excel-
lent advice, and critical reading of our manuscript. We
with to thank Opfermann Inc., Wiehl, Germany, which
kindly supported the publication of the color plates. Pre-
liminary data were presented at the 10th annual meeting of
the American Society for Bone and Mineral Research in
New Orleans, 1988.
REFERENCES
1. Raisz LG, 1987 Interactions among local and systemic regu-
lators of bone metabolism. In: Christiansen C, Johansen JS,
Riis BJ (eds) Osteoporosis 1987, Vol. 2. Osteopress, Copen-
hagen, pp. 751-755.
2. Canalis E 1985 Effect of growth factors on bone cell replica-
tion and differentiation. Clin Orthop 193:246-263.
3. Centrella M, Canalis E 1985 Local regulators of skeletal
growth: A perspective. Endoc Rev 65544-551.
4. Hefley TJ, Stern PH, Brand JS 1983 Enzymatic isolation of
cells from neonatal calvaria using two purified enzymes from
Clostridium histolyticum. Exp Cell Res 149:227-234.
5. Majeska RJ, Rodan SB, Rodan GA 1980 Parathyroid hor-
mone-responsive clonal cell lines from rat osteosarcoma.
Endocrinology 107:1494-1502.
6. Peck WA, Birge SJ, Fedak SA 1964 Bone cells biochemical
and biological studies after enzymatic isolation. Science 146:
1476- 1501.
7. Pallasch TJ 1968 The healing pattern of an experimentally-
induced defect in the rat femur studied with tetracycline la-
beling. Calcif Tissue Res 2:334-342.
8. Johner R 1972 Zur Knochenheilung in Abhangigkeit der
Defektgrosse. Helv Chir Acta 39:409-418.
9. Ziegler R, Delling G 1972 Effect of calcitonin on the regener-
ation of a circumscribed bone defect. Acta Endocrinol
(Copenh) 69:497-506.
10. Frost HM 1983 The regional acceleratory phenomenon: A re-
view. Henry Ford Hosp Med J 31:3-9.
11. Mueller M, Schilling T, Minne HW, Ziegler R 1988 An in
vivo model to study bone formation and bone resorption
under defined conditions. ASBMR 10th annual meeting,
New Orleans, abstract 476. J Bone Miner Res 3, Supp l:S187.
12. Bab 1, Gazit D, Massarawa J, Sela J 1985 Removal of tibia1
marrow induces increased formation of bone and cartilage in
rat mandibular condyle. Calcif Tissue Int 37551-555.
13. Saackel LR 1974 Elektronische Auswertung von Mikroradio-
grammen. Biomed Technol 19:lO-14.
14. Sterz H, Hebold G 1980 Analysis of radiographs in repro-
duction toxicology by means of densitometry and planime-
try. In: Merker HJ, Nau H, Neubert D (eds) Teratology of
the Limbs. De Gruyter, Berlin, pp. 373-382.
15. Archimedes 1963 (3rd century BC) Oxymenon A3, sect 3, 4,
5.
16. Frost HM 1969 Tetracycline-based histological analysis of
bone remodeling. Calcif Tissue Res 3:211-237.
409
17. Rahn BA 1971 Xylenol 0rang.e. a fluorochrome useful in
polychrome sequential labeling of calcifying tissues. Stain
Technol 46:125-129.
18. Tam CS, Anderson W 1980 Tetracycline labeling of bone in
vivo. Calcif Tissue Int 30121-125.
19. Rahn BA, Bacellar FC, Trapp L, Perren SM, 1980 Methode
zur Fluoreszenz-Morphometric: des Knochenanbaus. Akt
Traumatol 10:109-115.
20. Schenk RK 1965 Zur histologis,:hen Verarbeitung von unent-
kalktem Knochen. Acta Anat (Basel) 60:3-19.
21. Merz WA, Schenk RK 1970 A quantitative histological study
on bone formation in human cancellous bone. Acta Anat
(,Basel)7 6 1- 15 .
22. Schenk RK, Merz WA, Muller J 1969 A quantitative histo-
logical study on bone resorpticm in human cancellous bone.
Acta Anat (Basel) 74:44-53.
23. Von Kossa 1901 Nachweis vor Kalk. Uber die im Organis-
mus kiinstlich erzeugbaren Verkalkungen. Zieglers Beitr 29:
163-202.
24. Parfitt AM et al. 1987 Bone histomorphornetry: Standardiza-
tion of nomenclature, symbols and units. J Bone Miner Res
2:595-610.
25. Tapp E 1966 Tetracycline labelling methods of measuring the
growth of bones in the rat. J I3one Joint Surg [Br] 48517-
525.
26. O'Brien RG, Kaiser MK, 1985 b4ANOVA method for analyz-
ing repeated measures designs: An extensive primer. Psycho1
Bull 97:316-333.
27. Gazit D, Karmish M, Holzmari L, Bab I 1990 Regenerating
marrow induces systemic increase in osteo- and chondrogene-
sis. Endocrinology 126:2607-2(113.
28. Foldes J, Naparstek E, Statter M, Menczel J, Bab I 1989
Osteogenic response to marrow aspiration: Increased serum
osteocalcin and alkaline phosptatase in human bone marrow
donors. J Bone Miner Res 4:643-646.
29. Schenk RK 1978 Die Histologie der primaren Knochenheilung
im Lichte neuer Konzeptionen iiber den Knochenumbau. Un-
fallheilkunde 81:2 19-227.
30, Melcher AH, Irving JT 1964 Healing in a circumscribed de-
tect in the rat femur. In: Blackvvood HJJ (ed) Proceedings of
the First European Symposiurn on Bone and Tooth. Per-
gamon Press, New York, pp. 337-341.
31. McKibbin B 1978 The biology of fracture healing in long
bones. J Bone Joint Surg [Br] 60:150-162.
32. Rodan GA, Rodan SB 1983 Expression of the osteoblastic
phenotype. In: Peck WA (ed) Bone and Mineral Research,
Annual 2. Elsevier, Amsterdam, pp. 244-285.
33. Frost HM 1963 Basic science. 111: Lam CR (ed) Bone Remod-
eling Dynamics. Charles C. Thomas, Springfield, IL, pp.
27-36.
34. Frost HM 1972 Biochemical rc:sponse characteristics of fi-
brous tissue. In: Frost HM (ed) The Physiology of Cartilagi-
nous, Fibrous and Bony Tissur:, Orthopaedic Lectures Vol-
ume 11. Charles C. Thomas, Springfield, IL, pp. 43-83,
130-165.
35. Bab 1, Gazit D, Muhlrad A, Shteyer A 1988 Regenerating
bone marrow produces a potent growth-promoting activity
to osteogenic cells. Endocrinology 123:345-352.
36. Seyedin SM, Thomas TC, Thompson A, Rosen DM, Piez
KA 1985 Purification and characterization of two cartilage
inducing factors from bovine dimineralized bone. Proc Natl
Acad Sci USA 82:2267.
37. Centrella M, McCarthy PL, Canalis E 1987 TGF-6 is a bi-
functional regulator of replicat: on and collagen synthesis in
osteoblast-enriched cell culture!, from fetal rat bone. J Biol
Chem 262:2869-2874.
410
38. Pfeilschifter J, Mundy GR 1987 Modulation of type &trans-
forming growth factor activity in bone cultures by osteo-
tropic hormones. Proc Natl Acad Sci USA 84:2024-2028.
39. Canalis E, MR Carthy T, Centrella M 1988 Isolation of
growth factors from adult bovine bone. Calcif Tissue Int 43:
346-351.
40. Dekel S, Lenthall G, Francis MJO 1981 Release of prosta-
glandins from bone and muscle after tibial fracture. J Bone
Joint Surg [Br] 63:185-189.
41. Raisz LF, Martin TJ 1983 Prostaglandins in bone and min-
eral metabolism. In: Peck WA (ed) Bone and Mineral Re-
search, Annual 2. Elsevier, Amsterdam, pp. 286-310.
42. Shih M-S, Norridin RW 1986 Effects of prostaglandins on
regional remodeling changes during tibial healing in beagles:
A histomorphometric study. Calcif Tissue Int 39191-197.
43. Ueda K, Saito A, Nakano H, Aoshima M, Yokota M,
Muraoka R, Iwaya T 1980 Cortical hyperostosis following
long-term administration of prostaglandin E, in infants with
cyanotic congenital heart disease. J Pediatr 972334436.
44. Ringel RE, Brenner JI, Haney PJ, Burns JE, Moulton AL,
Berman MA 1982 Prostaglandin-induced periostitis: A com-
plication of longterm PGE, infusion in an infant with con-
MUELLER ET AL.
genital heart disease. Radiology 142:657-658.
45. Davidovitch Z, Shanfeld JL, Montgomery PC, Lally E,
Laster L, Furst L, Korostoff E 1984 Biochemical mediators
of the effects of mechanical forces and electric currents on
mineralizaed tissues. Calcif Tissue Int 36:86-97.
46. Bindermann I, Shimshini Z, Somjen D 1984 Biochemical
pathway involved in the translation of physical stimulus into
biological message. Calcif Tissue Int M82-85.
47. Thompson DD, Rodan GA 1988 Indomethacin inhibition of
tenotomy-induced bone resorption in rats. J Bone Miner Res
3:409-4 14.
Address reprint requests to:
Helmut W. Minne
Klinik DER FURSTENHOF
Center of Endocrinology, Metabolic Bone Disease,
and Gynecology
P.O. Box 1660
0-3280 Bad Pyrmont, Germany
Received for publication June 1, 1990;in revised form November
14, 1990;accepted November 19, 1990.