Current Osteoporosis Reports

https://doi.org/10.1007/s11914-018-0416-1

MUSCLE AND BONE (L BONEWALD AND M HAMRICK, SECTION EDITORS)

The Role of Bone Secreted Factors in Burn-Induced Muscle Cachexia

Gordon L. Klein 1

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
Purpose of Review Burn injury results in resorptive bone loss, failure to make new bone, and muscle protein breakdown resulting
in cachexia. The purpose of this review is to examine the relationship between bone loss and muscle atrophy in burn injury with a
view to understanding the process at work and how it may apply to other conditions that have similar features.
Recent Findings We present data suggesting that the use of bisphosphonates in the first 10 days following the burn prevents not
only the resorptive bone loss but also the muscle wasting. While an extra-osseous effect of bisphosphonates remains possible,
existing evidence points to a paracrine effect of bone on maintenance of muscle mass and strength. Proposed paracrine factors
produced by bone include prostaglandin E2 and components of the Wnt signaling pathway. TGFβ may be a bone paracrine factor
that causes oxidative damage to muscle.
Summary In the light of the pattern of evidence, burn patients suffer acute resorptive bone loss and muscle wasting. This is likely
due to the effects of inflammatory cytokines and endogenous glucocorticoid production in exacerbating oxidative stress. Early
use of bisphosphonates can maintain bone mass leading to a paracrine effect of bone in the maintenance of muscle mass, although
one cannot completely discount a direct effect of bisphosphonate on muscle. Because investigators report this relationship in a
variety of conditions in addition to burns, physicians should seriously consider the early use of bisphosphonates to maintain bone
and muscle mass in a variety of neuromuscular and skeletal diseases.

Keywords Bisphosphonates . Bone and muscle crosstalk . Muscle wasting . Bone paracrine factors

Introduction What Happens Following a Burn Injury?

Humans did not evolve with unique defenses against burns or
other trauma. Burns are relatively rare occurrences to which
the body responds by non-specific adaptive mechanisms. The
study of these responses in the unique setting of burn injury
allows us to understand a bit more how these adaptive re-
sponses function and perhaps how we might use this under-
standing to manage other unrelated conditions. In this paper,
we examine what burns can teach us about bone and muscle
crosstalk, especially bone to muscle communication.
This article is part of the Topical Collection on Muscle and Bone
* Gordon L. Klein
gordonklein@ymail.com
1
Department of Orthopaedic Surgery and Rehabilitation, University of
Texas Medical Branch and Shriners Burns Hospital, 301 University
Boulevard, Galveston, TX 77555-0165, USA
The destruction of the skin as a barrier to microorganisms sets
off a systemic inflammatory response that combats presump-
tive sepsis, and a stress response marked by the increased
endogenous production of glucocorticoids and catechol-
amines [1]. These responses lead to an increase in resting
energy expenditure to nearly twice normal and to a catabolic
response including muscle wasting and negative nitrogen bal-
ance [2]. What has become clear more recently is that burn
victims also lose bone acutely due to increased resorption [3]
and more chronically due to the failure to make new bone to
replace what was lost [4]. Long-term follow-up of pediatric
burn patients suggests that burn injury may affect trabecular
bone more profoundly than cortical bone given that whole
body bone mineral content in pediatric burn patients given a
placebo returns to parity with those given a bisphosphonate
more rapidly than does lumbar spine bone mineral content
under the same conditions [5]. Moreover, the risk of post-
burn fractures in pediatric patients is approximately 15% [6].
The incidence of post-burn fractures in adults has not yet been

reported, although data obtained by Muschitz et al. in adults
[7] indicate long-lasting changes as determined by
microcomputed tomography showing reduced trabecular
width and increased spacing as well as increased cortical po-
rosity potentially increasing the risk of post-burn fragility frac-
ture. The persistence of low bone density in pediatric burn
patients would also appear to put them at increased risk to
have a lower peak bone mass and thus to enter adulthood with
a potentially greater risk for osteoporosis.
Contributing Factors to Post-Burn Bone Loss
Factors resulting from the burn that contribute to bone
loss would include inflammation and increased endoge-
nous glucocorticoid production as well as possibly sepsis.
All three are known to produce oxidative stress [8•, 9, 10]
leading to the nuclear migration of forkhead box O
(FOXO) transcription factors that affect both bone and
muscle. In bone, FOXO migrates to the nucleus of bone
marrow stem cells, binding to β catenin and preventing it
from being activated by the Wnt signaling system, thus
interfering with osteoblastic bone formation [11].
Additionally, oxidative stress stimulates the accumulation
of hydrogen peroxide in the mitochondria of osteoclast
precursors, inhibiting their differentiation into mature os-
teoclasts, and resulting in reduced bone formation and
breakdown [12]. Further, marrow adipogenesis, which
has been associated with reduced osteoblastogenesis in
other conditions [13], has not been studied in burns and
may play a role in bone loss as well.
Untreated, this catabolic response lasts for 2–3 years [2].
Treating this response considerably shortens the periods of
muscle wasting and bone loss and may provide insight into
how muscle and bone interact. In the sections that follow, we
will discuss the effects of treatment with the anabolic agents,
recombinant human growth hormone and oxandrolone, and
then the surprising effects of treatment with bisphosphonates,
anti-resorptive drugs not previously thought to be anabolic.
Treatment with Anabolic Agents
The use of recombinant human growth hormone (rhGH) as an
anabolic agent to treat severe burns began in the 1990s and
experience with the drug was published beginning in 2001
[14, 15, 16•]. With the use of dual-energy X-ray absorptiom-
etry in pediatric burn patients, we found a significant increase
in lean body mass from discharge to 6 months post-burn
starting with a standard daily subcutaneous dose of 0.05 μg/
kg body weight. An increase in bone density was not seen
during the first year following burn, although a proportional
increase in bone mineral content and bone area was observed
Curr Osteoporos Rep
at 1 year following the burn injury. These increases resulted in
a larger and biomechanically stronger bone but without in-
crease in density. Therefore, an increase in lean body mass,
presumably muscle, preceded any change in bone [14]. This
finding was substantiated by the concomitant observation that
the injection of rhGH was immediately followed by an in-
crease in circulating insulin-like growth factor 1 (IGF-1) but
not by an increase in serum osteocalcin concentration [15]
again suggesting that there was a lag in the effect of rhGH
on bone compared to muscle. Finally, in a dose-response
study, Branski et al. [16•] reported that at a dose of 0.2 μg/
kg body weight/day rhGH produced a significant increase in
lean body/muscle mass but a reduction in bone density, sug-
gesting that at this high-dose rhGH directly promoted bone
resorption while still increasing muscle mass.
Given the expense of the drug and the route of administra-
tion as well as the potential difficulty in patient compliance
taking this drug, oral oxandrolone, a non-aromatizable andro-
gen, was successfully substituted for rhGH in the early 2000s.
Early studies indicated a similar pattern of primary increase in
muscle mass followed by an increase in bone mineral content
and bone area after a year of treatment [17, 18] with an in-
crease in bone density observed only after at least 12 consec-
utive months of oxandrolone treatment [19]. Oxandrolone is
the currently preferred drug for the management of the cata-
bolic changes brought about by burn injury. It is given in
conjunction with the β adrenergic blocker propranolol, but
there remains a lag of approximately 6 months from the be-
ginning of treatment to the increase in muscle mass and a
further delay till an improvement in bone mass is detected.
This lag may be due in part to the high levels of endogenous
glucocorticoids resulting from the stress response to burn
injury.
At this point, we introduce the studies done using the bis-
phosphonate pamidronate given to children as part of a dou-
ble-blind, randomized, placebo-controlled trial in which we
infused the drug over 12 h within the first 10 days following
burn injury. We found that doing so not only prevented resorp-
tive bone loss [20], the effects of which are known to normally
last for at least 2 years [5], but a retrospective review of the
data indicates that muscle breakdown was also reduced [21•,
22] and muscle strength is either preserved or restored
9 months after the burn [21•]. Bisphosphonates, as far as is
known, are taken up primarily by bone and any effect on
muscle is likely to be mediated by bone. We will now review
the existing data and examine the possible mechanisms by
which these effects can be achieved.
Bisphosphonates, Bone, and Muscle
Once we understood that the early post-burn administra-
tion of pamidronate to pediatric burn patients prevented
Curr Osteoporos Rep
the early resorption of bone, we undertook a retrospective
review of those patients who participated in the random-
ized, controlled double-blind study of pamidronate on
bone loss who also participated in studies involving mus-
cle protein kinetics. The aim of the study was to deter-
mine whether the prevention of bone loss had any effect
on muscle protein turnover or muscle strength. Seventeen
pediatric burn patients, ten receiving pamidronate and
seven receiving placebo, had also undergone muscle pro-
tein kinetic studies [21•]. We reviewed the data on muscle
protein kinetics from these patients by examining the ki-
netics of stable isotopes of phenylalanine as they related
to muscle protein synthesis, breakdown, and balance. We
found that in the burned patients receiving pamidronate,
muscle protein synthesis and breakdown were significant-
ly lower than in the group receiving placebo [21•].
Moreover, the difference between muscle protein synthe-
sis and breakdown resulted in a net muscle protein bal-
ance that was positive at 30 days post-burn in the
pamidronate group and negative in the placebo group
[21•, 22]. Furthermore, muscle fiber diameter at 30 days
post-burn was significantly greater in the subjects receiv-
ing pamidronate compared to controls [21•] and leg
strength at 9 months post-burn was not different between
subjects receiving pamidronate and normal, physically fit,
age-matched children. Burned children who had leg
strength determinations and who had received the placebo
tended to have lower limb strength that was less than
those in the pamidronate group although not quite signif-
icant, p = 0.052 [21•]. Importantly, urine free cortisol was
not different between groups receiving pamidronate or
placebo; thus, glucocorticoid antagonism can be eliminat-
ed as a significant factor affecting the outcome. This is so
despite the presence of a glucocorticoid response element
in the myostatin promoter [23] and that increased gluco-
corticoid production following burns increases myostatin
production, which may contribute to the suppression of
osteoblastogenesis [24].
How Did This Happen?
As stated above, we can eliminate any effect of
bisphosphonates on steroid biosynthesis given the lack of dif-
ference between pamidronate and placebo groups in urine free
cortisol. A second possibility is that pamidronate attenuated
the inflammatory response so as to reduce cytokine produc-
tion that might have contributed to muscle wasting. This was
not borne out by bioplex analysis of cytokine profiles (Jay JW,
Finnerty CC, unpublished data). Furthermore, inasmuch as we
have demonstrated that in pediatric burn patients the parathy-
roid Ca-sensing receptor is likely upregulated by pro-
inflammatory cytokines [25, 26], we should have seen a
difference in blood ionized Ca or parathyroid hormone con-
centrations in this patient group, which we did not [20].
Therefore, two possibilities remain (a) that bisphosphonates
are taken up by muscle and have a direct and persistent effect
on mass and strength or (b) that bisphosphonate-mediated
preservation of bone mass allows for normal paracrine com-
munication between bone and muscle, thus preserving muscle
mass.
With regard to the possibility of direct uptake of bis-
phosphonate by skeletal muscle, there is one paper by
Griffin et al. [27] indicating that some bisphosphonate
can be taken up by horse skeletal muscle as seen on scin-
tigraphy by 1 day post-injection, but that this quantity
rapidly diminishes over subsequent days and is almost
always gone by 3 days post-injection. A review of soft
tissue uptake of diphosphonates used in scintigraphy sug-
gests that soft tissue damage, inflammation, and poor re-
nal function may contribute to these unusual findings
[28]. Further, in those studies that examined the duration
of uptake the diphosphonate did not stay in the soft tissue.
Binding to calcium in the soft tissue may be an important
factor in soft tissue bisphosphonate uptake. While burn
injury can fit into this possible situation due to tissue
destruction and inflammation, no studies have examined
this possibility, although bisphosphonates given to any
subject with impaired renal function and extra-osseous
calcification is very uncommon, at least in pediatric
burns. Thus, while it may be possible for bisphosphonates
to be transiently taken up by muscle, the drug appears not
to stay in muscle. Moreover, this transient uptake is un-
likely to explain the effect of bisphosphonate on muscle
protein kinetics up to 3 weeks post-injection and the effect
on muscle strength at nearly 9 months post-injection, but
further studies are necessary to validate this explanation.
Eliminating the above possibilities leaves the chance that
the bone mediates the effect of bisphosphonates on muscle
and that preservation of bone mass is critical to the function
of bone regulation of muscle mass and strength.
While there is no direct evidence that osteocyte-derived
proteins support muscle metabolism in vivo, several re-
ports of in vitro studies indicate that these bone-derived
proteins have anabolic effects on muscle tissue. Thus, a
report by Mo et al. [29•] suggests that prostaglandin E2
from the osteocyte-like cell line MLOY4 increases myo-
genic differentiation of skeletal muscle while sclerostin
does not. Furthermore, recently published work by
Huang et al. [30•] shows that the Wnt/β catenin system
components may also affect crosstalk between these
osteocyte-like cells and C2C12 myoblasts, stimulating
myogenic differentiation, as well as increasing soleus con-
tractile strength. Moreover, sclerostin inhibited both ef-
fects suggesting that the increasing sclerostin concentra-
tions with time that we observed in the serum of pediatric

Table 1 Putative bone paracrine factors affecting muscle
Prostaglandin E2
Wnt 3a
Wnt 1
Tumor necrosis factor α (osteokine?)
Transforming growth factor β
Osteocalcin
burns patients given a bisphosphonate [29•] may reflect a
response to a putative increase in Wnt signaling resulting
from the prevention of bone resorption by the bisphospho-
nate administration to this group of patients. It should also
be noted that serum concentrations of FGF23 in these
same patients was undetectable regardless of whether they
were given bisphosphonates, thus eliminating FGF23 as
an osteocyte-derived mediator of muscle mass and
strength [31].
Other studies support these results. Watanabe et al.
[32•] reported that in a mouse model of muscular atrophy
the bisphosphonate ibandronate inhibits reduction of mus-
cle volume and induction of the muscle specific ubiquitin
ligases atrogin-1 and MuRF-1 in vivo. These factors are
involved in the ubiquitination of muscle proteins targeting
them for breakdown. Furthermore, they found that
ibandronate blocked immobilization-induced bone loss
in vivo and inhibited C2C12 myogenic cell expression
of atrogin-1 and MuRF-1. Moreover, they noted that the
in vitro effects of ibandronate were blunted by MG132, a
ubiquitin/proteasome inhibitor suggesting that ibandronate
works via the ubiquitin-proteasome system, the same one
that is affected by FOXO during oxidative stress [10].
Also of note, Regan et al. [33•] recently reviewed their
work on bone metastases liberating bone-produced TGFβ,
Fig. 1 Hypothetical description of the role of bisphosphonates in the
production of post-burn bone secretory factors that have a paracrine
action on muscle
Curr Osteoporos Rep
which can migrate to skeletal muscle and induce oxidative
changes causing disruption of the ryanodine receptor and
calcium leakage with consequent weakness. The use of
bisphosphonates can prevent this sequence of events. In
addition, Yoon et al. [34] in a mouse model of Duchenne
muscular dystrophy demonstrated that pamidronate treat-
ment decreased serum and muscle creatine kinase and
improved grip strength. Table 1 lists putative paracrine
factors secreted by bone.
Conclusions
Thus, a small but growing body of evidence supports the
idea that the use of bisphosphonates either improves mus-
cle mass and strength or prevents muscle breakdown and
consequent weakness. Whether this bisphosphonate effect
is direct or indirect is uncertain. The paper on horses by
Griffin et al. [27] suggests that muscle uptake of bisphos-
phonate is limited and transient and would support the
idea that maintenance of bone mass is necessary for the
maintenance of muscle mass under normal conditions.
If maintenance of muscle mass is dependent on bone
paracrine factors, then the current candidates are prosta-
glandin E2, Wnt 3a, Wnt 1, and osteocalcin [35, 36]. We
assayed Transforming growth factor (TGF) β in the serum
of our patients receiving pamidronate in the randomized
double-blind controlled study and there was no difference
between those receiving pamidronate and those given pla-
cebo (Jay JW, unpublished data 2016). However,
inasmuch as TGFβ in serum is not restricted to bony
origin, this finding is not specific enough to discount a
role of bone-derived TGFβ in our burn patients.
Certainly, the results of these studies warrant increased
scrutiny of these factors and others yet to be identified
as biomarkers in clinical trials. Figure 1 shows a hypo-
thetical illustration of the effects of bisphosphonates on
bone paracrine factors following burn injury.
The final point to make is the relative rates of improve-
ment in muscle and bone mass when one administers
pharmaceutical agents that stimulate muscle in order to
maintain bone mass, as with the use of either recombinant
human growth hormone or oxandrolone, or when one pri-
marily preserves bone mass by the use of
bisphosphonates. The early use of bisphosphonates to pre-
vent bone and muscle loss appears to result in a more
rapid recovery of, or a failure to lose, muscle mass, at
least in burn injury. Therefore, as newer treatment options
are available not only for burn injury but for any condi-
tion in which muscle and bone are lost perhaps more
attention should be paid to the early use of agents to
conserve bone mass.