Gap junctions in skeletal development and function
Gap junctions play a critical role in the coordinated function and activity of nearly all of the skeletal cells. This is not surprising, given the elaborate orchestration of skeletal patterning, bone modeling and subsequent remodeling, as well as the mechanical stresses, strains and adaptive responses that the skeleton must accommodate. Much remains to be learned regarding the role of gap junctions and hemichannels in these processes. A common theme is that without connexins none of the cells of bone function properly. Thus, connexins play an important role in skeletal form and function.
Gap junctions in skeletal development and function
Joseph P. Stains
, Roberto Civitelli
University of Maryland School of Medicine, Department of Orthopaedics, Baltimore, MD 21201, USA
Washington University, Division of Bone and Mineral Diseases, St. Louis, MO 63110, USA
Received 1 September 2005; received in revised form 26 October 2005; accepted 28 October 2005
Available online 14 November 2005
Gap junctions play a critical role in the coordinated function and activity of nearly all of the skeletal cells. This is not surprising, given the elaborate
orchestration of skeletal patterning, bone modeling and subsequent remodeling, as well as the mechanical stresses, strains and adaptive responses that
the skeleton must accommodate. Much remains to be learned regarding the role of gap junctions and hemichannels in these processes. A common
theme is that without connexins none of the cells of bone function properly. Thus, connexins play an important role in skeletal form and function.
D2005 Elsevier B.V. All rights reserved.
Keywords: Connexin; Bone; Cartilage; Osteoblast; Chondrocyte
1. Introduction .............................................................. 69
2. Connexins, gap junctions and hemichannels ............................................. 69
3. Gap junctions and skeletal development ............................................... 71
4. Gap junctions and skeletal cells ................................................... 73
4.1. The chondrocytes ....................................................... 73
4.2. Osteoblasts and osteocytes................................................... 75
4.3. Osteoclasts ........................................................... 77
4.4. Additional cells present in bone ................................................ 77
References ................................................................. 78
Bone is a dynamic tissue that is constantly modeled and
remodeled in response to a large number of stimuli, including
calcitropic hormones, growth factors and mechanical load. The
precise and coordinated control of bone remodeling requires a
tightly orchestrated interplay among, osteoblasts [the bone
forming cells], osteocytes [the bone embedded cells that are
the putative mechanosensory cells of bone] and osteoclasts [the
bone macrophage-like bone resorbing cells]. Gap junctional
communication has been hypothesized to play a critical role in
the coordination of bone remodeling. Osteoblasts and osteocytes
have been shown to express three major gap junction proteins,
connexin43 (Cx43), connexin45 (Cx45) and connexin46
(Cx46); and form functional gap junctions. Chondrocytes, the
cells that form cartilage, have also been shown to express Cx43;
as do the bone resorbing osteoclasts. In this review, we will
summarize recent findings which have elucidated some of the
roles of gap junctions in bone development and maintenance.
2. Connexins, gap junctions and hemichannels
Gap junctions are aqueous conduits that are formed by the
docking of two hemichannels on juxtaposed cells. The gap
junctional hemichannel, or connexon, is composed of a
0005-2736/$ - see front matter D2005 Elsevier B.V. All rights reserved.
* Corresponding author. Division of Bone and Mineral Diseases, Department
of Internal Medicine, 660 S. Euclid Avenue-PO Box 8301, St. Louis, MO
63110, USA. Tel.: +1 314 454 7765; fax: +1 314 454 5047.
E-mail address: email@example.com (R. Civitelli).
Biochimica et Biophysica Acta 1719 (2005) 69 – 81
hexameric array of connexin monomers. Each connexin
monomer is a single polypeptide, composed of four transmem-
brane domains, two small extracellular loops, an intracellular
loop and intracellular amino- and carboxyl-ends (Fig. 1). More
than 17 connexin genes have been identified in the mouse
genome and 20 in the human genome . Additionally, a
family of gap junction-like proteins, termed pannexins, have
been identified in the brain . Orthologs of mammalian
connexins, termed innexins, have even been identified and
characterized in insect cells . Connexins are almost
ubiquitously expressed in cells, implicating that they serve a
critical role in the function of multicellular organisms. But
what is the precise function of gap junctions? Why are there so
many connexin genes?
Gap junctions permit diffusion of ions, metabolites and
small signaling molecules (e.g., cyclic nucleotides and inositol
derivatives). Depending upon the expressed connexin genes,
the resultant gap junction channels will exhibit specific charge
and size permeability. For example, Cx43 permits the diffusion
of relatively large signal molecules <1.2 kDa molecular mass,
with a preference for negatively charged molecules. Inositol
derivatives [4– 9] and cADP-ribose [5,10] are capable of
diffusion through gap junctions and can elicit a Ca
in coupled cells. In fact, Cx43-transfected HeLa cells micro-
injected with inositol 1,4,5-triphosphate (IP
) have been shown
to propagate Ca
waves to neighboring cells 2.5-fold more
efficiently than untransfected cells . In contrast, Cx45 forms
a smaller pore, permitting diffusion of molecules <0.3 kDa,
with a preference for positively charged molecules. Further,
connexins can assemble as a homomeric or heteromeric
hemichannel, and the connexin isotypes that forms the gap
junction hemichannels dictate the molecular size and perme-
ability of the resulting gap junction channel [11 – 15]. Cx43 and
Cx45 are two such connexins that can assemble into a single
hemichannel composed of both monomeric units. In the
resultant Cx43/Cx45 heteromeric channel, the biochemical
properties of Cx45 dominate and chemical and electrical
coupling among cells is markedly reduced [16– 18].In
addition, some connexon (hemichannel) pairs can form
heterotypic interactions dependent upon the compatibility of
the extracellular loops of the opposing hemichannels (e.g., one
cell expressing monomeric Cx43 hemichannels may dock with
an adjacent cell expressing monomeric Cx45 hemichannels).
These properties provide the gap junction great plasticity in
dictating the size permeability and selectivity of the resultant
communicative channel, restricting or allowing signaling only
to coupled cells. Further, gap junction channels are regulated in
a similar fashion as other membrane channel, with open/closed
states sensitive to transmembrane voltage and post-translational
modification of the connexin subunits. Activation of extracel-
lular signal regulated kinase (ERK), src and protein kinase C
have been shown to dynamically regulate Cx43 channel open/
closed state by phosphorylation of the C-terminal tail of the
connexin monomers [19– 21].
Accumulating evidence from many model systems consis-
tently suggests that the unique profile of connexins expressed
by a particular cell type can dictate the types of signals, second
messengers, and metabolites that are propagated among cells.
In this way, the cells can form a ‘‘functional syncytium’’ within
which the cells communicate, with the advantage that the type
of signals that can be diffused can be regulated. Thus, not all
cells in the network share every signal; while some signals that
diffuse through the gap junctions are rapidly distributed,
propagation of others may be limited to serve specific
Connexins have also been shown to serve as docking
platforms for signaling complexes. Cx43 has been found to co-
localize and co-immunoprecipitate with h-catenin [22,23],c-
src [24– 26], protein kinase C (PKC) aand ([27,28]. Co-
localization has been reported between Cx43 and p38 MAPK
; and Cx43 is a target for extracellular signal regulated
kinases (ERK), PKC, phosphatidylinositol-3-kinase (PI3K) and
src phosphorylation, indicating at least a transient physical
association [20,30,31]. Accumulated data support a model in
Fig. 1. Structure of the human Cx43 monomer. Connexin43 is a transmembrane
protein composed of an amino- (NT) and carboxyl- (CT) terminal cytoplasmic
tail, a large cytoplasmic loop (CL), two extracellular loops (EC1– 2) and 4
transmembrane domains (TM1 – 4). The three cysteine re sidues in each
extracellular loop (indicated in red) are required for docking of two
hemichannels or connexons to form a gap junction channel. The residues
indicated in blue are mutated in the autosomal dominant disorder ODDD. The
residue indicated in green indicates a frame shift mutation that produces ODDD
as well as an additional skin phenotype. The residue indicated in orange
indicates an amino acid duplication that leads to ODDD [44 – 47,135].
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 8170
which connexins are capable of recruiting a unique profile of
signal proteins to the resultant channel that is specific for the
biologically relevant signal molecule transmitted among cells.
Thus, it is not only the permeability of the gap junction, but
also the repertoire of signal molecules at the gap junctional
plaque that determines the functional consequence of signals
passed among coupled cells.
Much recent work has also elucidated the function of Cx43
hemichannel activity [32,33]. Unlike the classic paradigm of
the formation of a gap junctional channel between two cells,
unpaired hemichannels can open to the extracellular milieu,
thus providing an alternative mechanism for connexin function.
Hemichannels have been shown to regulate the release of
and ATP [32,33], and recently, prostaglandin E2 (PGE
. The role of hemichannels in bone cell function will be
3. Gap junctions and skeletal development
Gap junctions are involved in many phases of embryonic
development and patterning, including heart morphogenesis,
left– right asymmetry and limb patterning [35– 37]. For the
purposes of this review, we will discuss the role of gap
junctions in skeletal and limb development.
Bone is formed by two mechanisms, endochondral ossifi-
cation and intramembranous ossification. The majority of the
bones in the body are formed by endochondral ossification, in
which a cartilaginous anlage is formed by chondrocytes.
Subsequently the anlage becomes vascularized, then infiltrated
and mineralized by osteoblastic cells. Conversely, during
intramembranous ossification, osteoblastic cells condense and
mineralize in the absence of a cartilaginous scaffold. Osteo-
blastic mineralization occurs through the synthesis and
deposition of a complex extracellular matrix, which includes
type I collagen, and other non-collagenous proteins, such as
osteopontin, bone sialoprotein, and osteocalcin, among others.
During the deposition of extracellular matrix, some osteoblasts
become embedded in the non-mineralized matrix (osteoid) and
eventually become encased within the mineralized bone. These
cells, osteocytes, form long cellular processes, which run
through canaliculi within the bone tissue, thus coming in
contact among themselves and with cells on the bone surface.
At the interface between cytoplasmic processes and at contact
sites at the bone surface, gap junctions form, thus permitting
efficient signal exchanges among distally located cells,
embedded in the ossified matrix. Importantly, osteoblasts on
the bone surface are also abundantly connected via gap
junctions. Indeed, all of the cells of the skeleton express gap
junction proteins, including chondrocytes (Cx43), osteoblasts
(Cx43, Cx45, and Cx46), osteocytes (Cx43) and even
osteoclasts (Cx43) (Fig. 2).
In the developing limb bud, signals from the apical
ectodermal ridge, including fibroblast growth factor 4, affect
Cx43 expression in the underlying mesenchymal limb bud
[38,39]. In fact, in chick limb buds gap junctional communi-
cation exhibits a gradient, being most intense in the zone of
polarizing activity and almost absent in the opposing side of
the limb bud [40,41]. In the developing mouse limb bud, Cx43
gene expression was similarly detected in the zone of
polarizing activity, and upon limb outgrowth, Cx43 was
detected in the presumptive progress zone. Cx43 message
was observed in the condensing limb mesenchyme, but in the
more proximal regions where differentiation has been initiated
Cx43 expression was more restricted to the presumptive
perichondrium . These data suggest that Cx43 expression
is regulated by a morphogen gradient that may contribute to
limb patterning. Demonstrating the importance of Cx43 for
limb development, antisense oligonucleotides inhibition of
Cx43 expression in the chick embryo resulted in limb
malformations, including truncation of the limb bud, fragmen-
tation into two or more domains, or complete splitting of the
limb bud into two or three branches [42,43].
Interestingly, skeletal malformations have been documented
in humans with an autosomal dominant disorder termed
oculodentodigital dysplasia, or ODDD. Human genetic studies
have reported at least 24 separate mutations in the Cx43 gene
in association with oculodentodigital dysplasia [44 – 46] (Fig.
1). The limb phenotype of the affected patients includes
syndactyly of the hands and foot, hypoplasia or aplasia of the
middle phalanges, and abnormalities in craniofacial elements
[44– 46]. Affected patients frequently have cranial hyperosto-
Fig. 2. Gap Junctions in the Skeleton. Chondrocytes of the growth plate form a
columnar structure of progressive differentiation from resting chondrocytes >
proliferative chondrocytes (PRO) > pre-hypertrophic chondrocytes > hypertro-
phic chondrocytes (HYP) > terminal differentiated chondrocytes followed by
the osteo-chondral border. These cells are known to express Cx43, however
whether they function as gap junction channels (GJC) or hemichannels (HC)
has yet to be determined. Bone is covered by surface osteoblasts (OB) and
osteoclasts (OC). Embedded in the bony matrix is the osteocytes (OCY).
Osteocytes form long caniliculae, which interconnect the osteocytes and surface
osteoblasts via gap junctions. The interaction among osteoblasts and osteocytes
is magnified in the red box to the right to highlight the presence of gap
junctions at the intersection of caniliculae, and among adjacent osteoblasts.
Note the presence of connexin hemichannels on the surface of osteoblasts and
osteocytes, which have been shown to play an important gap junction-
independent role in bone. Osteocytes are pictured with putative connexin
hemichannels, though the precise distribution and function of these connexins
is not clear at this time.
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 81 71
sis, a widened alveolar ridge of the mandible, and broad tubular
bones. There are also dentition abnormalities including
microdontia, adontia, and enamel hypoplasia [44 –46]. The
Cx43 point mutations found in ODDD patients span much of
the Cx43 gene, including the N-terminal tail, the transmem-
brane domains, and the cytoplasmic loop. Recently, a
frameshift mutation in the C-terminal tail of the Cx43 gene
has been associated with ODDD as well as an additional skin
phenotype (Fig. 1). The phenotype of these patients
suggests that the Cx43 mutants might be hypomorphs of the
wild type gene. Indeed, a recent report functionally character-
ized eight of the Cx43 mutations found in patients with ODDD
by overexpressing them in HeLa cells . Two of the
mutants, F52dup and R202H, failed to generate gap junctional
plaques with the majority of protein found in the intracellular
space, mostly associated with the endoplasmic reticulum.
However, when co-expressed with the wild type Cx43, the
plaque forming defect was rescued, but gap junctional
communication was impaired. The other six mutants generated
in this study (K134E, I130T, L90V, Y17S, G21R and A40V)
all formed gap junctional plaques, yet had markedly reduced or
no electrical coupling. Similarly, two additional ODDD Cx43
mutants (G21R and G138R) were screened by Roscoe et al.
, and found to act in a dominant negative fashion with
respect to wild type Cx43. More recently, a new mouse model
originated from a mutagenesis screen was identified with a
phenotype closely resembling that of ODDD, including
syndactyly, enamel hypoplasia, craniofacial abnormalities and
cardiac dysfunction . These mice (Gja1
) carry a new
point mutation in the Cx43 gene causing a G60S substitution,
and the resulting mutant protein has dominant negative
properties, just like most of the other ODDD mutants.
Interestingly, these mice also have severely decreased bone
mass and mechanical strength, and exhibited bone marrow
abnormalities indicative of defects in hematopoietic stem cells,
feature not present or not reported in patients with ODDD.
The developmental defects observed in human ODDD, and
partly reproduced by this new mouse model are remindful of
skeletal abnormalities observed in a model of Cx43 ‘‘knock-
down’’ obtained using antisense oligonucleotides in chicken
[42,51]. When the antisense Cx43 oligonucleotides were
applied to the developing chick face primordia, significant
facial defects were observed, including aberrant maxillary and
mandibular primordium development and nasal pit defects.
Intriguingly, ‘‘knockdown’’ of Cx43 in the chick face, resulted
in downregulation of the homeobox transcription factor Msx1,
particularly in the affected areas of the developing chick
primordia . Msx1 and Msx2 serve critical roles in skeletal
development and patterning [52,53]. Double knockouts of
Msx1 and Msx2 lack anterior skeletal elements in the
developing limb, and single gene deletion results in craniofa-
cial developmental abnormalities, which include cleft palate
(Msx1) and craniosynostosis (Msx2) [53,54].
A role for Cx43 in osteogenesis emerges also from other
animal models. Recent data demonstrate that development of
the short fin (sof) phenotype in zebrafish is caused by a
mutation in the Cx43 gene . The first identified allele,
, expressed markedly reduced amounts of Cx43, while
three ENU induced alleles causing sof encoded missense
mutations of Cx43 (F30V0, F209I, and P191S). The develop-
ing zebrafish fin is composed of bony segments, formed by
intramembranous ossifications that determine the length of the
fin skeleton . Zebrafish homozygous for the mutant sof
alleles have tail segments that are approximately 1/3 the length
of wild type bony segments. Though the molecular details are
still unclear, there is an apparent defect in cell proliferation and
possibly osteogenic differentiation caused by reduced expres-
sion or hypomorphic Cx43 function .
Given all of the previous data regarding the role of Cx43 in
skeletal growth and development, it is quite surprising that
genetic ablation of Cx43 in mice did not overtly affect
patterning of the skeleton. Further, only a few of the defects
observed in ODDD were recapitulated in Cx43 null mice.
These mice die perinatally due to a severe defect in the heart,
leading to swelling and blockage of the right ventricle outflow
tract [57,58]. Despite no overt impact on skeletal morphogen-
esis, the Cx43 null mice exhibit profound mineralization
defects in the shape and mineralization of skeletal elements
derived from both intramembranous and endochondral ossifi-
cation (Fig. 3). While most skeletal elements originate from the
mesoderm; many of the skeletal elements in the head are
derived from migratory neural crest cells . Previous studies
had revealed a major role for Cx43 in the migration of neural
crest cells that likely account for the cardiac defect observed in
Cx43 null mice . Not surprisingly, many of the bony
elements of the skull derived from the neural crest are affected
by loss of Cx43, including a lack of ossification centers in the
bones of the cranial vault and delayed ossification of the
premaxilla, maxilla and mandibula at E16.5 . Furthermore,
the developmental delay in the parietal and frontal bones
resulted in a smaller calvarium, a flattened skull and open
Fig. 3. Delayed Ossification of the Cx43 null mice. Alcian Blue/Alizarin red
staining of E15.5 mouse embryos shows that Cx43 deficient mice have severe
delay in ossification of many of the craniofacial elements, including the cranial
vault and bones of the jaw, as well as the axial and appendicular skeleton. Blue
staining indicates cartilaginous scaffold. Red staining is indicative of
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 8172
parietal foramen. This is likely caused by defective migration
of the neural crest cells to these skeletal elements. Yet, the
hypoplastic and hypomineralized cranial phenotype of Cx43
null and Gja1
mice is in striking contrast to human subjects
with ODDD whose cranial vault is thickened and hyperminer-
alized . Clearly, other factors are at play in the development
of the adult human phenotype that modify the effect of the
mutant Cx43 in ODDD.
In addition to the defects observed in the neural crest
derived skeleton, there is also considerable delay in the
ossification of the mesoderm derived skeleton in the Cx43
null mice. Affected skeletal elements in mesoderm derived
skull elements included basiocciptal and exoccipital bones,
which were hypomineralized even at birth. The axial skeleton
also was affected by loss of Cx43. The ribs of Cx43 null mice
are dysmorphic and hypomineralized at E15.5. Similarly, the
axial skeleton was hypomineralized at E15.5 (Fig. 3). At birth,
however, many of the skeletal elements appeared ossified,
indicating approximately a 2-day delay in skeletal ossification.
Given the data demonstrating a role in bone growth and
patterning, it was quite surprising that ablation of Cx43 would
not yield a more severe skeletal phenotype, with respect to size
and morphology. Also as noted, surprisingly, the Cx43 null
mouse is not a phenocopy of human ODDD. One possible
explanation is compensation by other connexins, and the best
candidate may be Cx45, also expressed by bone cells. Recent
findings on another genetic mutation may offer an additional,
potential compensatory mechanism for lack of Cx43 attributed
to this lack of a massive morphological defect. A recent paper
by Pizard et al. , using a mouse model of Holt – Oram
syndrome, a human dominants disorder characterized by limb
malformations and heart disease caused by haploinsufficiency
of T-box transcription factor, Tbx5 , found that Tbx5
regulates expression of connexin40 (Cx40) . Indeed, many
of the skeletal abnormalities present in Tbx5
shared by Cx40
mice. In situ hybridization
revealed co-expression of Tbx5 and Cx40 in the developing
forelimbs and carpal bones as well as the sternal perichondri-
um. These sites of co-expression are consistent with the
phenotype of Holt – Oram syndrome, which includes dys-
morphic or amorphic thumbs, shortened arms, misshapen and
fused bones of the wrist and an abnormal sternum morphology
. Mice with only a single allele of Tbx5 exhibited reduced
expression of Cx40 at these sites, as well as the heart. The
occurrence of skeletal fusions in the wrist bones was identical
mice though the frequency of fusion
was markedly increased in Cx40
animals. Similar observa-
tions of skeletal dysmorphogenesis were observed in the
phalanges, carpal bones and sternum of both Tbx5
Cx40 mutant mice, implicating Cx40 as the downstream target
of Tbx5 that contributes to limb patterning in the upper limbs
and sternum. Interestingly, additional skeletal defects were
observed in the Cx40
mice in tissues where Cx40, but not
Tbx5, is expressed (e.g., the ribs and hindlimbs), suggesting
that Cx40 plays a large scale role in skeletal patterning.
mutants were unable to
generate live pups. Accordingly, the analyses of Tbx5
mice revealed that the mice die shortly after birth, and
demonstrate nearly identical features of the individual mutants
or in some cases a mild exacerbation of the phenotype. These
data indicate that Cx40 plays a critical role in the formation of
endochondral derived elements of the axial and appendicular
In summary, data accumulated thus far clearly demonstrate
that connexins play a critical role in endochondral (Cx43 and
Cx40) and intramembranous (Cx43) skeletal development. The
role of Cx45 in ossification has yet to be studied as these mice die
very early during embryogenesis (¨E10) before ossification has
begun and will thus require conditional deletion in the
developing skeleton . However, the functional redundancy
and distinct control of varying skeletal elements by gap junctions
certainly indicates a role for gap junctions in skeletal develop-
ment, patterning, growth and, as will be discussed below,
cellular function. Indeed, the striking phenotypic differences
between humans with ODDD, chickens with Cx43 knockdown,
Cx43 null mice or zebrafish with Cx43 tailfin growth mutation
may indicate that the degree of redundancy and the connexin
isoforms that provide overlapping functions may vary in
different species, thus, explaining the different phenotypic
expression of Cx43 deficiency so far observed in different
models. If the field is to progress in the understanding of the role
of gap junctions in skeletal growth and patterning, these
differences must be systematically addressed.
4. Gap junctions and skeletal cells
We will now review the current literature on the role of gap
junctions and connexin hemichannels on the function of
skeletal cells: the chondrocytes, the osteoblast, the osteocyte,
the osteoclast and the ‘‘other cells’’ of bone.
4.1. The chondrocytes
Chondrocytes are the cells of bone that form the initial
cartilaginous scaffold upon which bone will eventually form.
These cells build two critical skeletal structures: the growth
plate and the articular cartilage. The growth plate is comprised
of chondrocytic cells arranged in columnar fashion in stages of
progressive differentiation. The most distal portion of the
growth plate with respect to the metaphysis contains the resting
chondrocytes, a precursor cell that will enter the next stage of
differentiation, the proliferative chondrocyte. As the cells
divide and differentiate, they extend the length of the bone.
Moving towards the proximal side of the growth plate, the cells
undergo hypertrophy, eventually forming a mineralized carti-
lage that is eventually replaced by bone (Fig. 2). Conversely,
the articular cartilage is essentially a biomechanical shock
absorber at the epiphysis of long bones that is produced by
articular chondrocytes. These cells produce massive amounts
of collagens, proteoglycans and absorbed water, cushioning the
load placed upon the bones.
Chondrocytes express Cx43 by immunohistochemistry in
vivo in mice and rats . In this study, Cx43 was detected in
both the outer layer of the knee joint articular chondrocytes and
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 81 73
in growth plate chondrocytes. However, the functionality of
Cx43 in growth plate chondrocytes has not yet been tested, and
it is unclear whether gap junctional communication is
established between the columnar cells of the growth plate or
whether Cx43 is functioning as a hemichannel to release
paracrine signals, such as ATP or NAD
. In contrast, in vitro
assays have revealed that articular chondrocytes form func-
tional gap junctions in culture by dye transfer experiments
[66– 68]. Recently, functional Cx43 gap junctions in the
superficial zone of articular chondrocytes have been documen-
ted in vivo . The role of Cx43 in chondrocytes has not been
extensively studied in vivo, although it is fair to say that Cx43
is not absolutely required in the growth plate for bone growth
during embryogenesis, as the long bones of Cx43 null mice are
indistinguishable in size from their wild-type littermates .
In vitro micromass cultures of chondrocytes from the chick
limb bud have demonstrated that gap junctions are required for
the differentiation of chondrocytes, as inhibition of gap
junctional communication with 18a-glycyrrhetinic acid
reduces production of proteoglycans and type II collagen.
Furthermore, inhibition of gap junctional communication was
demonstrated to be important for the full chondro-anabolic
affects of BMP2 on these micromass cultures . A likely
explanation for the discrepancy between Cx43 null mice and in
vitro studies of inhibition of gap junctional communication
may be compensation by other connexins. The effects of 18a-
glycyrrhetinic acid on gap junctional communication are not
limited to inhibition of only Cx43. As discussed above, the
high levels of expression of Cx40 in the perichondrium (and
particular abundance in the periarticular perichondrium)
coupled to the defects in bone length in the Cx40 mutant mice
suggests that other connexins certainly contribute to chondro-
cyte function and long bone growth . Again these are open
questions that, with the sequencing of the mouse genome and
identification of all the putative members of the connexin gene
family, are ripe to be addressed.
In vitro studies have elaborated on the role of connexins in
articular cartilage. It has been shown that gap junctions
propagate intercellular Ca
waves when articular chondro-
cytes are stimulated by perturbation of a single cell with a
micropipette [67,71,72]. Interestingly, the generated Ca
wave propagation to adjacent cells was attenuated when cells
were treated with an inhibitor of gap junctions. Data have
suggested that articular chondrocytes accumulate intracellular
inositol 1,4,5-trisphosphate (IP
) following mechanical stimu-
lation and that IP
diffuses through gap junctions into adjacent
cells to amplify the response . Furthermore, it has been
shown that gap junctions form between the articular chon-
drocytes and synovial fibroblasts in co-culture and that these
cells elicit a coordinated Ca
response to mechanical
perturbation . Whether gap junctions form between
articular chondrocytes and synovial fibroblasts in vivo remains
The gap junctional communication between and/or among
articular chondrocytes and synovial fibroblasts may have
some interesting extrapolations into the field of arthritis.
Cx43 has been implicated in the etiology of osteoarthritis.
Osteoarthritis is mediated by a dynamic interplay between the
articular chondrocytes and the synovial cells. Articular
chondrocytes are the single cell type present in articular
cartilage. These cells are responsible for producing and
maintaining the articular cartilage extracellular matrix. Oste-
oarthritis changes are accompanied by extracellular matrix
degradation and production of a non-functional extracellular
matrix, which eventually results in cartilage destruction
[73,74]. Synovial fibroblasts are mesenchymal-derived cells,
which along with synovial macrophages, form a thin lining of
synovial tissue surrounding the fibrous capsule of the joint.
The physiologic role of synovial tissue is to produce a
synovial fluid that lubricates the joints and supplies nutrients
to the articular chondrocytes. However, the pathological
changes that occur in the synovium during osteoarthritis shift
the balance from an anabolic to a catabolic role, leading to
slow progressive destruction of articular cartilage. Despite the
fact that osteoarthritis is considered a non-inflammatory form
of arthritis, there are changes within the joint that are
associated with chronic low-grade inflammation. It has been
speculated that this inflammation is a result of the release of
cartilage breakdown products into the synovium . These
products induce the production of the inflammatory cytokine,
interleukin 1h(IL-1h), by the synovial fibroblasts. Indeed,
IL-1 is considered one of the most prevalent catabolic factors
in osteoarthritic joints, and it has been suggested to be
driving cartilage destruction during OA [76,77].Three
important findings suggest that Cx43 plays a role in both
articular chondrocytes and synovial fibroblasts during the
onset and progression of osteoarthritis: (1) IL-1hupregulates
Cx43 expression in cultured chondrocytes [78,79]. (2) In
osteoarthritis, there is a pathologic increase of the expression
of the gap junction protein, Cx43 (Cx43), in both synovial
fibroblasts and articular chondrocytes caused by IL-1h
[78,80,81]. (3) Transmission electron microscopic analysis
of synovial tissues from healthy and osteoarthritis affected
human patients demonstrated an increase in the size and
number of Cx43 gap junctional plaques in osteoarthritic
synovium relative to non-osteoarthritic synovium .
Critically important, in vitro experiments revealed that
blocking gap junctional communication with the inhibitors,
18a-glycyrrhetinic acid or octanol, attenuates the production
of matrix metalloproteinases, which degrade the cartilaginous
extracellular matrix, by synovial fibroblasts following stim-
ulation with IL-1h[80,82]. Thus, it seems apparent that gap
junctional communication may amplify catabolic signals
among synovial fibroblasts affecting articular cartilage
(Fig. 4). As mentioned previously, gap junction-dependent
intercellular calcium signaling occurs between articular
chondrocytes and synovial cells in co-culture, and these
gap junction-dependent calcium signals are amplified by two
factors, IL-1htreatment and mechanical perturbation; two
major influences on the onset and progression of OA
[71,78,79]. Changes in cytosolic Ca
modulate the articular
chondrocyte phenotype, pushing them towards hypertrophy
and terminal differentiation, a fate not normally ascribed to
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 8174
4.2. Osteoblasts and osteocytes
The most studied skeletal cells with regard to gap junctions
and connexins are the bone forming osteoblast and the
osteocyte. Numerous in vitro experiments have defined the
role of gap junctions in regulating osteoblast and osteocyte
function. Early work demonstrated the existence of gap
junctions among osteoblasts and osteocytes by electron
microscopy [83– 85]. The data were later confirmed by
ultrastructural analyses of histological bone sections [86,87],
and subsequently the presence of functional gap junctional
communication was established in murine  and human
[89,90] osteoblasts. Osteoblasts and osteocytes express pri-
marily Cx43 and to a lesser extent Cx45 [90,91]. Cx46 is also
expressed, however this protein does not traffic correctly to the
plasma membrane and remains in the trans Golgi . The
functional consequences or role for Cx46 are thus unknown.
Osteoblasts on the bone surface and the bone embedded
osteocytes have long been postulated to regulate bone anabolic
function via coordinated signaling among the cells via long
cellular processes originating from the osteocyte that intercon-
nect osteocytes and osteoblasts by gap junctions.
During osteoblast differentiation in vitro, the expression of
Cx43 increases, as does gap junctional communication [93,94].
In contrast, the expression level of Cx45 is unaltered during
osteogenic differentiation . In seminal work on the role of
Cx43 on osteoblast function, it was shown that inhibition of
gap junctional communication retarded the differentiation of
these cells, resulting in a reduced ability to form a mineralized
extracellular matrix and an attendant reduction in the expres-
sion of osteoblastic genes associated with differentiation [93 –
95]. In fact, it was shown that treatment of osteoblastic cells
with pharmacologic inhibitors of gap junctions, 18a-glycyr-
rhetinic acid and oleamide, not only prevented differentiation
into mature osteoblasts, but caused transdifferentiation into
adipocyte-like cells . The defective osteogenic differenti-
ation was recapitulated in primary osteoblasts isolated from the
calvaria of Cx43
mice . These cells failed to express
markers of terminal differentiation, including alkaline phos-
phatase, bone sialoprotein and osteocalcin, but also exhibited a
delay in the ability to form mineralized nodules in vitro. These
data implicate that Cx43 plays a very important role in
Interestingly, numerous bone anabolic factors have been
shown to upregulate Cx43 protein and gap junctional
communication, including bone morphogenetic protein 2
and parathyroid hormone (PTH) [97– 102].
Signaling activated by PTH results in a feed forward
progression that upregulates Cx43 expression, which in turn
amplifies the ability of the cell to respond to PTH. The cAMP
response in osteoblast-like cell lines treated with PTH was
shown to be attenuated when Cx43 expression was disrupted
using antisense RNA . Similarly, the ability of PTH to
induce matrix mineralization in mature osteoblasts is markedly
reduced when gap junctions are inhibited . A similar
observation has been made for PGE
, which is produced in a
gap junction-dependent manner during fluid flow shear stress
by osteocyte-like cells in vitro [34,105 – 107]. The increased
production feeds forward to further upregulate Cx43
A role of gap junctions in mechano-sensing by bone has
longed been hypothesized. The gap junctional plaques formed
at cell– cell borders of osteocytic processes have been
postulated to play a critical role in signaling among osteoblasts
and osteocytes in response to mechanical strain. In addition to
the studies on PGE
and mechanical strain or fluid flow
mentioned above, both osteoblasts (MC3T3-E1 cells) and
osteocytes (MLO-Y4 cells) have been shown to remodel their
gap junctions (Cx43 and Cx45) in response to fluid shear stress
using immunofluorescence detection of Cx43 and Cx45 in
these cells . This remodeling of gap junctional plaques is
dependent upon the mechanical shear stress applied; but in
both cell types the fluid shear at 5 dyn/cm
or 20 dyn/cm
resulted in diminished staining of Cx43 and Cx45 at
appositional membranes, and reduced gap junctional commu-
nication as assessed by dye coupling. Interestingly, despite the
loss in gap junctional coupling, both osteoblasts and osteocytes
were shown to increase Cx43 transcription at low fluid shear (5
), but not at high shear (20 dyn/cm
). Cx45 mRNA
Fig. 4. A putative model of the role of connexin43 in osteoarthritis. Wear-and-
tear on the articular cartilage has been shown to induce an inflammatory
response by the synovial tissue. The resultant production of IL-1 by the
synovium feeds back to upregulate Cx43 expression in both synovial
fibroblasts and articular chondrocytes. In turn, Cx43 potentiates signaling
further enhancing the production of catabolic factors such as IL-1h, MMPs and
additional cytokines. Subsequently, the surface articular chondrocytes produce
a response that triggers further degradation of the articular cartilage.
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 81 75
showed reciprocal regulation, being increased only at high
shear stress .
In contrast, recent work in the same osteocyte like cell line
has shown that fluid shear stress at 16 dyn/cm
amount of Cx43 on the plasma membrane . Importantly, this
discrepancy was shown by detection of Cx43 that was capable of
being biotinylated in intact cells, thus labeling only surface
bound Cx43. Further, the accessibility of biotin to the Cx43 is
limited to connexins that have not yet formed a coupled gap
junction, and thus represent hemichannels. The authors also
show that the levels of PGE
released into the extracellular fluid
is inversely proportional to the density of plated cells when
normalized to cell number, i.e., low density cultures produce
per cell than higher density cultures . In contrast,
intracellular levels of PGE
remain unchanged. Further, treat-
ment with gap junction inhibitors, 18h-glycyrrhetinic acid or
carbenoxolone, or antisense Cx43 RNA ablates the fluid flow
induced release of PGE
in these cells. The authors argue that
low-density cultures implicate the activity of Cx43 hemichan-
nels rather than Cx43 gap junction channels in the resultant
effect on PGE
release. Indeed, it has been shown that fluid flow
or mechanical perturbation can induce the opening of Cx43
hemichannels in osteoblast and osteocytes [34,110].
Similar to osteocytes and chondrocytes, osteoblasts have
been shown to respond to mechanical perturbation. Osteo-
blasts, like chondrocytes, produce synchronized Ca
among cells following mechanical manipulation. These Ca
waves occur via both gap junction dependent and gap junction
independent mechanisms. The gap junction-independent Ca
waves are a result of the autocrine activity of secreted
extracellular ATP on P2X purinergic receptors, and the gap
junction-dependent propagation of Ca
-waves are caused by
influx of Ca
though L-type voltage operated calcium
channels [111– 113]. These signals converge upon cell function
in a connexin43 dependent manner.
Several groups have undertaken detailed molecular analyses
to investigate the mechanisms of how alteration in gap
junctions and Cx43 in bone cells affects cell phenotype. Work
from our group and others have focused on examining how
modulating Cx43 function at the plasma membrane can
influence osteogenic gene expression. In primary osteoblasts
isolated from Cx43
mice, transcription of many markers of
osteogenic differentiation are reduced, including bone sialo-
protein, osteocalcin, and type I collagen .Itwas
subsequently shown that inhibition of Cx43 function in
osteoblast-like cell lines could reproduce the attenuation of
transcription [94,96,114,115]. Alternately, overexpression of
Cx43 in moderately coupled cells could increase transcription
of osteoblast genes [95,116]. Gap junction dependent gene
expression regulation has also been reported in other cell
contexts [117–122], but the molecular mechanisms attending
to this novel function of gap junctions remain unclear.
Recently, we have shown that disruption of Cx43 function in
osteoblasts results in an attenuation in the activation of the
extracellular signal regulated kinase (ERK) response to serum
stimulation . Further, we have shown that the Cx43-
dependent alteration of ERK signaling modulates gene
transcription from the promoters of several osteoblast gene
promoters . In fact, the transcriptional deficiency ob-
served in cells with disrupted Cx43 gap junctions can be
mimicked by inhibition of ERK signaling and rescued by over
expression of constitutively active members of the ERK
signaling cascade . The consequence of modulation of
gap junction dependent ERK signal transduction converges
upon an Sp1/Sp3 binding element in the promoters of the
osteocalcin and collagen Ia1 genes, two genes downregulated
by loss of Cx43. By chromatin immunoprecipitation, we
observed that under conditions of robust Cx43-mediated gap
junctional communication, the activator Sp1 and the repressor
Sp3 both can occupy the promoter at a nearly 1:1 ratio, slightly
favoring occupancy by Sp1. As a result transcription from
these promoters is high. In contrast, when Cx43-mediated gap
junctional communication is inhibited, the occupancy of this
promoter element is maintained almost exclusively by the
repressor, Sp3, markedly reducing transcription [123,124].
Accordingly, we have shown that ERK cascade dependent
phosphorylation of Sp1 mediates the preferential recruitment of
Sp1 over Sp3 in well-coupled cells, and loss of Sp1
phosphorylation results in the preferential recruitment of Sp3
. Thus, we termed this Sp1/Sp3 binding element a
connexin response element, or CxRE, due to its exquisite
responsiveness to the degree of gap junctional coupling in
osteoblastic cells. These data led us to propose a model in
which cells facilitate a ‘‘primary response’’ to an extracellular
cue by ligand– receptor mediated activation of cellular signals.
The magnitude of this ‘‘primary’’ response is regulated in part
by the bioavailability of ligand and the surface abundance of
the receptor in a cell, leading to modulation of signal
transduction cascades. In contrast, gap junctional communica-
tion permits a ‘‘secondary’’ response, which potentiates the
‘‘primary’’ response of the cells. Signals, in the form of second
messengers like IP
or cADP ribose, generated by the
‘‘primary’’ response are propagated to adjacent cells via gap
junctions initiating this ‘‘secondary’’ response in coupled cells
that potentiates and coordinates signaling among a local
population of cells (Fig. 5).
Another interesting role for Cx43 and downstream signaling
in osteoblasts and osteocytes has been demonstrated. A series
of papers have documented the role of Cx43 hemichannels in
the anti-apoptotic response of osteoblasts and osteocytes
caused by the administration of the bisphosphonate, alendro-
nate. In the first paper, Plotkin et al., show that the bone
anabolic therapeutic drug, alendronate, can act on osteoblasts
to prevent etoposide and dexamethasone-induced apoptosis,
and that this effect required Cx43 .Theyfurther
demonstrated that this effect was mediated by the src-ERK
signaling cascade. Importantly, the authors demonstrated that
treatment of MLO-Y4 osteocyte-like cells with alendronate
could open Cx43 hemichannels, as assessed by dye uptake.
The argument for Cx43 hemichannel function in preventing
induced apoptosis of cells was enhanced by showing that cells
cultured at low density or in suspension, thus minimizing the
likelihood of gap junctional communication, still maintained
the activation of ERK and anti-apoptotic effects initiated by
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 8176
alendronate treatment. These data suggested that Cx43 hemi-
channels may serve as a ‘‘receptor’’ for bisphosphonates, and
that the resultant Cx43-dependent ERK activation is essential
for the anti-apoptotic effects of alendronate on osteoblasts.
More recently, this same group elaborated upon their earlier
findings, reporting that the anti-apoptotic effects of alendronate
on osteocytes was mediated by cytoplasmic ERK action, rather
than the canonical nuclear translocation pathway .
Further, they demonstrated that a permeability impaired Cx43
mutant lacking 7 amino acids in the cytoplasmic loop (amino
acids, 130– 136), failed to protect osteocytes from etoposide
induced apoptosis by alendronate. Similarly, they revealed that
the C-terminal tail of Cx43 is likewise required for the
hemichannel mediated anti-apoptotic effects of alendronate.
In summary, the role of gap junctions in osteogenic function
is critically involved in lineage progression and gene transcrip-
tion. Signals propagated through Cx43 gap junction channels (or
hemichannels) regulate osteoblastic and osteocytic phenotype.
In the absence of Cx43, osteoblast function is impaired as the
responsiveness to hormonal and physical cues are attenuated.
The least studied bone cell type with respect to gap
junctions is the hematopoietic derived osteoclast. Osteoclasts
have been shown to express Cx43 [87,127]. These bone-
resorbing cells are generated from the fusion of monocyte like
precursor cells to form multinucleated cells. A role for Cx43
has been implicated in the fusion of these precursor cells into
osteoclasts. Treatment of osteoclasts with the gap junction
inhibitor heptanol markedly reduced the number of osteoclast
like cells . The number of unfused, mononuclear
precursor cells was increased, suggesting a defect in fusion.
Further, the activity of the osteoclasts that were formed had
reduced ability to resorb bone slices. Fewer multinucleated
osteoclasts were present and fewer of them were active as
determined by bone pit assays. Surprisingly, it was found that
the minority of osteoclasts that were active in the heptanol-
treated cultures produced larger resorption pits in the bone
slices than vehicle-treated samples. Similar results were
obtained using a peptide inhibitor of Cx43, known as
Gap27 . However, one notable difference is that not
only did Gap27 prevent osteoclast fusion and activity, there
appeared to be a marked effect on apoptosis in these cells,
though the effect was not rigorously analyzed. Analogous
results for the resorption pit data were also obtained by
another group using different chemical inhibitors of gap
junctions . In this study, they found that not only did
inhibition of gap junctions results in fewer resorption pits by
osteoclasts, but that the ability of PTH and vitamin D
stimulate osteoclast activity was markedly inhibited by
blockage of gap junctions.
4.4. Additional cells present in bone
In addition to the ‘‘primary’’ cells of bone listed above,
numerous additional cell types are present within the skeletal
network. The marrow cavity serves as a critical domain for
progenitor cells, and interactions between bone and stem cell
populations has been shown to serve a critical niche [130,131].
In addition, vascular invasion of the cartilaginous anlage is
critical for endochondral ossification; and the mineralized bone
is a highly vascularized tissue. Gap junctions have been shown
to form among bone lining cells and marrow stromal cells .
In vitro work has demonstrated the formation of gap junctional
communication among human bone marrow stromal cells and
endothelial cells . Further, in this study, co-culture of
human umbilical vein endothelial cells with human bone
marrow stromal cells increased the expression of alkaline
phosphatase and type I collagen, two markers of osteoblast
differentiation, by the stromal cells when placed in direct
contact with the endothelial cells . These data suggest that
the physical interaction between endothelial cells and stromal
cells supports osteoblastogenesis. The authors of this study
revealed heterotypic gap junctional communication among the
co-cultured cells, and demonstrated that inhibition of gap
junctional communication could attenuate the osteogenic
effects of co-culture. In subsequent work, it was shown that
the contribution of endothelial cell-stromal cell interactions to
osteoblast commitment and/or differentiation could be mediat-
ed by several additional endothelial cell populations, including
human primary vascular endothelial cells, endothelial cells
Fig. 5. Model of the role of gap junctions in potentiation of signals from an
extracellular cue. For example, we show our model for transcription from the
osteocalcin CxRE. Extracellular cues such as growth factors and mechanical
loading induce second messengers (e.g., IP
, cAMP, Ca
waves), leading to
what we have termed a primary cellular response. This response is independent
of gap junction. In the presence of gap junctional communication the
intercellular propagation of second messengers that activate the ERK cascade,
leads to a secondary response in the adjacent, gap junctionally coupled cell. In
the example of the osteocalcin Sp1/Sp3-binding CxRE shown here, these
signals converge on nuclear function by regulating the recruitment of the
transactivator Sp1 to the CxRE in the promoter, leading to robust transcription
(Txn). When gap junctional communication is disrupted the cellular response is
attenuated, due to the failure to propagate signals among cells (i.e., only the
primary response occurs).
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 81 77
isolated from cord blood and endothelial cells isolated from the
saphen vein . As was observed for the human umbilical
vein endothelial cells, these effects were mediated by direct
contact and gap junctional communication. The authors of
these two studies implicate Cx43 in the heterotypic gap
junctional communication among endothelial cells and stromal
cells [132,133]; however, in light of the data from the Cx40
mutant mice, it may be interesting to examine whether
heterotypic interactions between endothelial cells, which
express Cx40 as well as Cx43 , and stromal cells
contribute to the skeletal phenotype observed in these animals.
 K. Willecke, J. Eiberger, J. Degen, D. Eckardt, A. Romualdi, M.
Guldenagel, U. Deutsch, G. Sohl, Structural and functional diversity of
connexin genes in the mouse and human genome, Biol. Chem. 383
 R. Bruzzone, S.G. Hormuzdi, M.T. Barbe, A. Herb, H. Monyer,
Pannexins, a family of gap junction proteins expressed in brain, Proc.
Natl. Acad. Sci. U. S. A. 100 (2003) 13644.
 P. Phelan, J.P. Bacon, J.A. Davies, L.A. Stebbings, M.G. Todman, L.
Avery, R.A. Baines, T.M. Barnes, C. Ford, S. Hekimi, R. Lee, J.E. Shaw,
T.A. Starich, K.D. Curtin, Y.A. Sun, R.J. Wyman, Innexins: a family of
invertebrate gap-junction proteins, Trends Genet. 14 (1998) 348.
 H. Niessen, H. Harz, P. Bedner, K. Kramer, K. Willecke, Selective
permeability of different connexin channels to the second messenger
inositol 1,4,5-trisphosphate, J. Cell Sci. 113 (Pt. 8) (2000) 1365.
 G.C. Churchill, C.F. Louis, Roles of Ca2+, inositol trisphosphate and
cyclic ADP-ribose in mediating intercellular Ca2+ signaling in sheep
lens cells, J. Cell Sci. 111 (Pt. 9) (1998) 1217.
 T. Fry, J.H. Evans, M.J. Sanderson, Propagation of intercellular calcium
waves in C6 glioma cells transfected with connexins 43 or 32, Microsc.
Res. Tech. 52 (2001) 289.
 S. Boitano, E.R. Dirksen, M.J. Sanderson, Intercellular propagation of
calcium waves mediated by inositol trisphosphate, Science 258 (1992)
 K. Braet, K. Paemeleire, K. D’Herde, M.J. Sanderson, L. Leybaert,
Astrocyte-endothelial cell calcium signals conveyed by two signalling
pathways, Eur. J. Neurosci. 13 (2001) 79.
 M.J. Sanderson, A.C. Charles, E.R. Dirksen, Mechanical stimulation and
intercellular communication increases intracellular Ca
cells, Cell Regul. 1 (1990) 585.
 G.S. Goldberg, P.D. Lampe, D. Sheedy, C.C. Stewart, B.J. Nicholson,
C.C. Naus, Direct isolation and analysis of endogenous transjunctional
ADP from Cx43 transfected C6 glioma cells, Exp. Cell Res. 239 (1998)
 A.D. Martinez, V. Hayrapetyan, A.P. Moreno, E.C. Beyer, Connexin43
and connexin45 form heteromeric gap junction channels in which
individual components determine permeability and regulation, Circ.
Res. 90 (2002) 1100.
 G.T. Cottrell, Y. Wu, J.M. Burt, Cx40 and Cx43 expression ratio
influences hetromeric/heterotypic gap junction channel properties, Am.
J. Physiol.: Cell Physiol. 282 (2002) C1469 – C1482.
 D.S. He, J.X. Jiang, S.M. Taffet, J.M. Burt, Formation of heteromeric
gap junction channels by connexin 40 and 43 in vascular smooth muscle
cells, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 6495.
 J.X. Jiang, D.A. Goodenough, Heteromeric connexons in lens gap
junction channels, Proc. Natl. Acad. Sci. U. S. A. 93 (1996) 1287.
 P.A. Weber, H.C. Chang, K.E. Spaeth, J.M. Nitsche, B.J. Nicholson, The
permeability of gap junction channels to probes of different size is
dependent on connexin composition and permeant-pore affinities,
Biophys. J. 87 (2004) 958.
 A.P. Moreno, G.I. Fishman, E.C. Beyer, D.C. Spray, Voltage Dependent
Gating and Single Channel Analysis of Heterotypic Gap Junction
Channels Formed of Cx45 and Cx43, Progress in Cell Research, Elsevier
Science, BV, 1995, pp. 405 – 408.
 M. Koval, S.T. Geist, E.M. Westphale, A.E. Kemendy, R. Civitelli, E.C.
Beyer, T.H. Steinberg, Transfected connexin45 alters gap junction
permeability in cells expressing endogenous connexin43, J. Cell Biol.
130 (1995) 987.
 A.D. Martinez, V. Hayrapetyan, A.P. Moreno, E.C. Beyer, Connexin43
and connexin45 form heteromeric gap junction channels in which
individual components determine permeability and regulation, Circ.
Res. 90 (2002) 1100.
 B.J. Warn-Cramer, A.F. Lau, Regulation of gap junctions by tyrosine
protein kinases, Biochim. Biophys. Acta 1662 (2004) 81.
 V. Cruciani, S.O. Mikalsen, Connexins, gap junctional intercellular
communication and kinases, Biol. Cell 94 (2002) 433.
 M.Z. Hossain, A.L. Boynton, Regulation of Cx43 gap junctions: the
gatekeeper and the password, Sci. STKE 2000 (2000) E1.
 Z. Ai, A. Fischer, D.C. Spray, A.M. Brown, G.I. Fishman, Wnt-1
regulation of connexin43 in cardiac myocytes, J. Clin. Invest. 105 (2000)
 X. Xu, W.E. Li, G.Y. Huang, R. Meyer, T. Chen, Y. Luo, M.P. Thomas,
G.L. Radice, C.W. Lo, Modulation of mouse neural crest cell motility by
N-cadherin and connexin 43 gap junctions, J. Cell Biol. 154 (2001) 217.
 T. Toyofuku, Y. Akamatsu, H. Zhang, T. Kuzuya, M. Tada, M. Hori,
c-Src regulates the interaction between connexin-43 and ZO-1 in cardiac
myocytes, J. Biol. Chem. 276 (2001) 1780.
 M.Y. Kanemitsu, L.W. Loo, S. Simon, A.F. Lau, W. Eckhart, Tyrosine
phosphorylation of connexin 43 by v-Src is mediated by SH2 and SH3
domain interactions, J. Biol. Chem. 272 (1997) 22824.
 B.N. Giepmans, T. Hengeveld, F.R. Postma, W.H. Moolenaar, Interaction
of c-Src with gap junction protein connexin-43. Role in the regulation of
cell – cell communication, J. Biol. Chem. 276 (2001) 8544.
 B.W. Doble, P. Ping, E. Kardami, The epsilon subtype of protein kinase
C is required for cardiomyocyte connexin-43 phosphorylation, Circ. Res.
86 (2000) 293.
 N. Bowling, X. Huang, G.E. Sandusky, R.L. Fouts, K. Mintze, M.
Esterman, P.D. Allen, R. Maddi, E. McCall, C.J. Vlahos, Protein kinase
C-alpha and -epsilon modulate connexin-43 phosphorylation in human
heart, J. Mol. Cell Cardiol. 33 (2001) 789.
 R. Schulz, P. Gres, A. Skyschally, A. Duschin, S. Belosjorow, I.
Konietzka, G. Heusch, Ischemic preconditioning preserves connexin 43
phosphorylation during sustained ischemia in pig hearts in vivo, FASEB
J. 17 (2003) 1355.
 P.D. Lampe, A.F. Lau, The effects of connexin phosphorylation on gap
junctional communication, Int. J. Biochem. Cell Biol. 36 (2004) 1171.
 P.D. Lampe, A.F. Lau, Regulation of gap junctions by phosphorylation of
connexins, Arch. Biochem. Biophys. 384 (2000) 205.
 L. Ebihara, New roles for connexons, News Physiol. Sci. 18 (2003) 100.
 D.A. Goodenough, D.L. Paul, Beyond the gap: functions of unpaired
connexon channels, Nat. Rev., Mol. Cell Biol. 4 (2003) 285.
 P.P. Cherian, A.J. Siller-Jackson, S. Gu, X. Wang, L.F. Bonewald, E.
Sprague, J.X. Jiang, Mechanical strain opens connexin 43 hemichannels
in osteocytes: a novel mechanism for the release of prostaglandin, Mol.
Biol. Cell 16 (2005) 3100.
 M. Levin, Isolation and community: a review of the role of gap
junctional communication in embryonic patterning, J. Membr. Biol. 185
 F.D. Houghton, Role of gap junctions during early embryo development,
Reproduction 129 (2005) 129.
 C.J. Wei, X. Xu, C.W. Lo, Connexins and cell signaling in development
and disease, Annu. Rev. Cell Dev. Biol. 20 (2004) 811.
 R.A. Meyer, M.F. Cohen, S. Recalde, J. Zakany, S.M. Bell, W.J.J. Scott,
C.W. Lo, Developmental regulation and asymmetric expression of the
gene encoding Cx43 gap junctions in the mouse limb bud, Dev. Genet.
21 (1997) 290.
 H. Makarenkova, D.L. Becker, C. Tickle, A.E. Warner, Fibroblast growth
factor 4 directs gap junction expression in the mesenchyme of the
vertebrate limb Bud, J. Cell Biol. 138 (1997) 1125.
 C.N. Coelho, R.A. Kosher, A gradient of gap junctional communication
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 8178
along the anterior – posterior axis of the developing chick limb bud, Dev.
Biol. 148 (1991) 529.
 C.N. Dealy, E.C. Beyer, R.A. Kosher, Expression patterns of mRNAs for
the gap junction proteins connexin43 and connexin42 suggest their
involvement in chick limb morphogenesis and specification of the
arterial vasculature, Dev. Dyn. 199 (1994) 156.
 D.L. Becker, I. McGonnell, H.P. Makarenkova, K. Patel, C. Tickle, J.
Lorimer, C.R. Green, Roles for alpha 1 connexin in morphogenesis of
chick embryos revealed using a novel antisense approach, Dev. Genet. 24
 C.R. Green, L. Bowles, A. Crawley, C. Tickle, Expression of the
connexin43 gap junctional protein in tissues at the tip of the chick limb
bud is related to the epithelial – mesenchymal interactions that mediate
morphogenesis, Dev. Biol. 161 (1994) 12.
 R. Richardson, D. Donnai, F. Meire, M.J. Dixon, Expression of Gja1
correlates with the phenotype observed in oculodentodigital syndrome/
type III syndactyly, J. Med. Genet. 41 (2004) 60.
 K.W. Kjaer, L. Hansen, H. Eiberg, P. Leicht, J.M. Opitz, N. Tommerup,
Novel Connexin 43 (GJA1) mutation causes oculo-dento-digital dyspla-
sia with curly hair, Am. J. Med. Genet. 127A (2004) 152.
 W.A. Paznekas, S.A. Boyadjiev, R.E. Shapiro, O. Daniels, B. Wollnik,
C.E. Keegan, J.W. Innis, M.B. Dinulos, C. Christian, M.C. Hannibal,
E.W. Jabs, Connexin 43 (GJA1) mutations cause the pleiotropic
phenotype of oculodentodigital dysplasia, Am. J. Hum. Genet. 72
 M.A. van Steensel, L. Spruijt, D.B. van, R.S. Bladergroen, M. Vermeer,
P.M. Steijlen, G.M. van, A 2-bp deletion in the GJA1 gene is associated
with oculo-dento-digital dysplasia with palmoplantar keratoderma, Am.
J. Med. Genet., A 132 (2005) 171.
 J. Shibayama, W. Paznekas, A. Seki, S. Taffet, E.W. Jabs, M. Delmar, H.
Musa, Functional characterization of connexin43 mutations found in
patients with oculodentodigital dysplasia, Circ. Res. 96 (2005) e83 – e91.
 W. Roscoe, G.I. Veitch, X.Q. Gong, E. Pellegrino, D. Bai, E. McLachlan,
Q. Shao, G.M. Kidder, D.W. Laird, Oculodentodigital dysplasia-causing
connexin43 mutants are non-functional and exhibit dominant effects on
wild-type connexin43, J. Biol. Chem. 280 (2005) 11458.
 A.M. Flenniken, L.R. Osborne, N. Anderson, N. Ciliberti, C. Fleming,
J.E. Gittens, X.Q. Gong, L.B. Kelsey, C. Lounsbury, L. Moreno, B.J.
Nieman, K. Peterson, D. Qu, W. Roscoe, Q. Shao, D. Tong, G.I.
Veitch, I. Voronina, I. Vukobradovic, G.A. Wood, Y. Zhu, R.A.
Zirngibl, J.E. Aubin, D. Bai, B.G. Bruneau, M. Grynpas, J.E.
Henderson, R.M. Henkelman, C. McKerlie, J.G. Sled, W.L. Stanford,
D.W. Laird, G.M. Kidder, S.L. Adamson, J. Rossant, A Gja1 missense
mutation in a mouse model of oculodentodigital dysplasia, Develop-
ment 132 (2005) 4375.
 I.M. McGonnell, C.R. Green, C. Tickle, D.L. Becker, Connexin43 gap
junction protein plays an essential role in morphogenesis of the
embryonic chick face, Dev. Dyn. 222 (2001) 420.
 S. Alappat, Z.L.Y. Zhang, Y.P. Chen, Analysis of Msx1; Msx2 double
mutants reveals multiple roles for Msx homeobox gene family and
craniofacialgenes in limb development, Cell Res. 13 (2003) 429.
 M.M. Cohen Jr., Craniofacial disorders caused by mutations in homeo-
box genes MSX1 and MSX2, J. Craniofacial Genet. Dev. Biol. 20 (2000)
 S. Alappat, Z.Y. Zhang, Y.P. Chen, Msx homeobox gene family and
craniofacial development, Cell Res. 13 (2003) 429.
 M.K. Iovine, E.P. Higgins, A. Hindes, B. Coblitz, S.L. Johnson,
Mutations in connexin43 (GJA1) perturb bone growth in zebrafish fins,
Dev. Biol. 278 (2005) 208.
 M.K. Iovine, S.L. Johnson, Genetic analysis of isometric growth control
mechanisms in the zebrafish caudal Fin, Genetics 155 (2000) 1321.
 A.G. Reaume, P.A. De Sousa, S. Kulkarmi, B.L. Langille, D. Zhu, T.C.
Davies, S.C. Juneja, G.M. Kidder, J. Rossant, Cardiac malformation in
neonatal mice lacking connexin43, Science 267 (1995) 1831.
 J. Ya, E.B. Erdtsieck-Ernste, P.A. de Boer, M.J. van Kempen, H.
Jongsma, D. Gros, A.F. Moorman, W.H. Lamers, Heart defects in
connexin43-deficient mice, Circ. Res. 82 (1998) 360.
 G.F. Couly, P.M. Coltey, N.M. Le Douarin, The triple origin of skull in
higher vertebrates: a study in quail-chick chimeras, Development 117
 G.Y. Huang, E.S. Cooper, K. Waldo, M.L. Kirby, N.B. Gilula, C.W. Lo,
Gap junction-mediated cell – cell communication modulates mouse
neural crest migration, J. Cell Biol. 143 (1998) 1725.
 F. Lecanda, P.M. Warlow, S. Sheikh, F. Furlan, T.H. Steinberg, R.
Civitelli, Connexin43 deficiency causes delayed ossification, craniofa-
cial abnormalities, and osteoblast dysfunction, J. Cell Biol. 151 (2000)
 A. Pizard, P.G. Burgon, D.L. Paul, B.G. Bruneau, C.E. Seidman, J.G.
Seidman, Connexin 40, a target of transcription factor Tbx5, patterns
wrist, digits, and sternum, Mol. Cell. Biol. 25 (2005) 5073.
 C.T. Basson, D.R. Bachinsky, R.C. Lin, T. Levi, J.A. Elkins, J. Soults, D.
Grayzel, E. Kroumpouzou, T.A. Traill, J. Leblanc-Straceski, B. Renault,
R. Kucherlapati, J.G. Seidman, C.E. Seidman, Mutations in human
TBX5 [corrected] cause limb and cardiac malformation in Holt – Oram
syndrome, Nat. Genet. 15 (1997) 30.
 C.T. Basson, G.S. Cowley, S.D. Solomon, B. Weissman, A.K.
Poznanski, T.A. Traill, J.G. Seidman, C.E. Seidman, The clinical and
genetic spectrum of the Holt – Oram syndrome (heart – hand syndrome),
N. Engl. J. Med. 330 (1994) 885.
 M. Kumai, K. Nishii, K. Nakamura, N. Takeda, M. Suzuki, Y. Shibata,
Loss of connexin45 causes a cushion defect in early cardiogenesis,
Development 127 (2000) 3501.
 W. Schwab, A. Hofer, M. Kasper, Immunohistochemical distribution of
connexin 43 in the cartilage of rats and mice, Histochem. J. 30 (1998)
 P. D’Andrea, A. Calabrese, I. Capozzi, M. Grandolfo, R. Tonon, F.
Vittur, Intercellular Ca2+ waves in mechanically stimulated articular
chondrocytes, Biorheology 37 (2000) 75.
 H.J. Donahue, F. Guilak, M.A. Van der Molen, K.J. McLeod, C.T.
Rubin, D.A. Grande, P.R. Brink, Chondrocytes isolated from mature
articular cartilage retain the capacity to from functional gap junctions,
J. Bone Miner. Res. 10 (1995) 1359.
 S.S. Chi, J.B. Rattner, J.R. Matyas, Communication between paired
chondrocytes in the superficial zone of articular cartilage, J. Anat. 205
 W. Zhang, C. Green, N.S. Stott, Bone morphogenetic protein-2
modulation of chondrogenic differentiation in vitro involves gap
junction-mediated intercellular communication, J. Cell. Physiol. 193
 P. D’Andrea, A. Calabrese, M. Grandolfo, Intercellular calcium
signalling between chondrocytes and synovial cells in co-culture,
Biochem. J. 329 (Pt. 3) (1998) 681.
 I. Capozzi, R. Tonon, P. D’Andrea, Ca2+-sensitive phosphoinositide
hydrolysis is activated in synovial cells but not in articular chondrocytes,
Biochem. J. 344 (Pt. 2) (1999) 545.
 K.E. Kuettner, M.B. Aydelotte, E.J.M.A. Thonar, Articular cartilage
matrix and structure: a minireview, J. Rheumatol., Suppl. 27 (1991) 46.
 K.E. Kuettner, V. Goldberg, Osteoarthritic disorders, in: K.E. Kuettner,
V. Goldberg (Eds.), American Academy of Orthopaedic Surgeons
Symposium Series, Rosemont, 1995, pp. 27 – 45.
 M.J. Benito, D.J. Veale, O. Fitzgerald, W.B. van den Berg, B. Bresnihan,
Synovial tissue inflammation in early and late osteoarthritis, Ann.
Rheum. Dis. (2005) 1263.
 W.B. van den Berg, The role of cytokines and growth factors in cartilage
destruction in osteoarthritis and rheumatoid arthritis, Z. Rheumatol. 58
 X. Chevalier, Upregulation of enzymatic activity by interleukin-1 in
osteoarthritis, Biomed. Pharmacother. 51 (1997) 58.
 R. Tonon, P. D’Andrea, Interleukin-1beta increases the functional
expression of connexin 43 in articular chondrocytes: evidence for a
Ca2+-dependent mechanism, J. Bone Miner. Res. 15 (2000) 1669.
 P. R.Tonon, P. D’Andrea, The functional expression of connexin 43 in
articular chondrocytes is increased by interleukin 1beta: evidence for a
Ca2+-dependent mechanism, Biorheology 39 (2002) 153.
 A.A. Marino, D.D. Waddell, O.V. Kolomytkin, W.D. Meek, R. Wolf,
K.K. Sadasivan, J.A. Albright, Increased intercellular communication
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 81 79
through gap junctions may contribute to progression of osteoarthritis,
Clin. Orthop. (2004) 224.
 O.V. Kolomytkin, A.A. Marino, K.K. Sadasivan, R.E. Wolf, J.A.
Albright, Intracellular signaling mechanisms of interleukin-1beta in
synovial fibroblasts, Am. J. Physiol. 276 (1999) C9 – C15.
 O.V. Kolomytkin, A.A. Marino, D.D. Waddell, J.M. Mathis, R.E. Wolf,
K.K. Sadasivan, J.A. Albright, IL-1beta-induced production of metallo-
proteinases by synovial cells depends on gap junction conductance, Am.
J. Physiol.: Cell Physiol. 282 (2002) C1254 – C1260.
 S.B. Doty, B.H. Schofield, Metabolic and structural changes within
osteocytes of rat bone, in: R.V. Talmage, P.L. Munson (Eds.), Calcium,
Parathyroid Hormone and the Calcitonins, Excerpta Medica, Amsterdam,
The Netherlands, 1972, pp. 353 – 364.
 S.B. Doty, Morphological evidence of gap junctions between bone cells,
Calcif. Tissue Int. 33 (1981) 509.
 P. Stanka, Occurrence of cell junctions and microfilaments in osteoblasts,
Cell Tissue Res. 159 (1975) 413.
 C. Palumbo, S. Palazzini, G. Marotti, Morphological study of intercel-
lular junctions during osteocyte differentiation, Bone 11 (1990) 401.
 S.J. Jones, C. Gray, H. Sakamaki, M. Arora, A. Boyde, R. Gourdie, C.R.
Green, The incidence and size of gap junctions between the bone cells in
rat calvaria, Anat. Embryol. 187 (1993) 343.
 B.G. Jeasonne, F.A. Faegin, R.W. McMinn, R.L. Shoemaker, W.S.
Rehm, Cell-to-cell communication of osteoblasts, J. Dent. Res. 58 (1979)
 R. Civitelli, P.M. Warlow, A.J. Robertson, L.V. Avioli, E.C. Beyer, T.H.
Steinberg, Cell Coupling and Expression of Connexin43 in Bone Cell
Networks, 1991, pp. 191a.
 R. Civitelli, E.C. Beyer, P.M. Warlow, A.J. Robertson, S.T. Geist, T.H.
Steinberg, Connexin43 mediates direct intercellular communication in
human osteoblastic cell networks, J. Clin. Invest. 91 (1993) 1888.
 H.J. Donahue, Gap junctions and biophysical regulation of bone cell
differentiation, Bone 26 (2000) 417.
 M. Koval, J.E. Harley, E. Hick, T.H. Steinberg, Connexin46 is retained
as monomers in a trans-Golgi compartment of osteoblastic cells, J. Cell
Biol. 137 (1997) 847.
 H.J. Donahue, Z. Li, Z. Zhou, C.E. Yellowley, Differentiation of human
fetal osteoblastic cells and gap junctional intercellular communication,
Am. J. Physiol.: Cell Physiol. 278 (2000) C315 – C322.
 P.C. Schiller, G. D’Ippolito, W. Balkan, B.A. Roos, G.A. Howard, Gap
junctional communication is required for the maturation process of
osteoblastic cells in culture, Bone 28 (2001) 362.
 F. Lecanda, D.A. Towler, K. Ziambaras, S.-L. Cheng, M. Koval, T.H.
Steinberg, R. Civitelli, Gap junctional communication modulates gene
expression in osteoblastic cells, Mol. Biol. Cell 9 (1998) 2249.
 P.C. Schiller, G. D’Ippolito, R. Brambilla, B.A. Roos, G.A. Howard,
Inhibition of gap-junctional communication induces the trans-differenti-
ation of osteoblasts to an adipocytic phenotype in vitro, J. Biol. Chem.
276 (2001) 14133.
 P.C. Schiller, P.P. Mehta, B.A. Roos, G.A. Howard, Hormonal regulation
of intercellular communication: parathyroid hormone increases con-
nexin43 gene expression and gap-junctional communication in osteo-
blastic cells, Mol. Endocrinol. 6 (1992) 1433.
 H.J. Donahue, K.J. McLeod, C.T. Rubin, J. Andersen, E.A. Grine, E.L.
Hertzberg, P.R. Brink, Cell-to-cell communication in osteoblastic net-
works: cell line-dependent hormonal regulation of gap junction function,
J. Bone Miner. Res. 10 (1995) 881.
 R. Civitelli, K. Ziambaras, P.M. Warlow, F. Lecanda, T. Nelson, J.
Harley, N. Atal, E.C. Beyer, T.H. Steinberg, Regulation of connexin43
expression and function by prostaglandin E2 (PGE2) and parathyroid
hormone (PTH) in osteoblastic cells, J. Cell. Biochem. 68 (1998) 8.
 V. Shen, L. Rifas, G. Kohler, W.A. Peck, Prostaglandins change cell
shape and increase intercellular gap junctions in osteoblasts cultured
from rat fetal calvaria, J. Bone Miner. Res. 1 (1986) 243.
 G.H. Rudkin, D.T. Yamaguchi, K. Ishida, W.J. Peterson, F. Bahadosingh,
D. Thye, T.A. Miller, Transforming growth factor-beta, osteogenin, and
bone morphogenetic protein-2 inhibit intercellular communication and
alter cell proliferation in MC3T3-E1 cells, J. Cell. Physiol. 168 (1996) 433.
 P.P. Cherian, B. Cheng, S. Gu, E. Sprague, L.F. Bonewald, J.X. Jiang,
Effects of mechanical strain on the function of Gap junctions in
osteocytes are mediated through the prostaglandin EP2 receptor, J. Biol.
Chem. 278 (2003) 43146.
 M.A. Van der Molen, C.T. Rubin, K.J. McLeod, L.K. McCauley, H.J.
Donahue, Gap junctional intercellular communication contributes to
hormonal responsiveness in osteoblastic networks, J. Biol. Chem. 271
 P.C. Schiller, G. D’Ippolito, W. Balkan, B.A. Roos, G.A. Howard, Gap-
junctional communication mediates parathyroid hormone stimulation of
mineralization in osteoblastic cultures, Bone 28 (2001) 38.
 M.M. Saunders, J. You, Z. Zhou, Z. Li, C.E. Yellowley, E.L. Kunze, C.R.
Jacobs, H.J. Donahue, Fluid flow-induced prostaglandin E2 response of
osteoblastic ROS 17/2.8 cells is gap junction-mediated and independent
of cytosolic calcium, Bone 32 (2003) 350.
 D.C. Genetos, D.J. Geist, D. Liu, H.J. Donahue, R.L. Duncan, Fluid
shear-induced ATP secretion mediates prostaglandin release in MC3T3-
E1 osteoblasts, J. Bone Miner. Res. 20 (2005) 41.
 M.M. Saunders, J. You, J.E. Trosko, H. Yamasaki, Z. Li, H.J. Donahue,
C.R. Jacobs, Gap junctions and fluid flow response in MC3T3-E1 cells,
Am. J. Physiol.: Cell Physiol. 281 (2001) C1917 – C1925.
 B. Cheng, S. Zhao, J. Luo, E. Sprague, L.F. Bonewald, J.X. Jiang,
Expression of functional gap junctions and regulation by fluid flow in
osteocyte-like MLO-Y4 cells, J. Bone Miner. Res. 16 (2001) 249.
 M.M. Thi, T. Kojima, S.C. Cowin, S. Weinbaum, D.C. Spray, Fluid shear
stress remodels expression and function of junctional proteins in cultured
bone cells, Am. J. Physiol.: Cell Physiol. 284 (2003) C389 – C403.
 M. Romanello, P. D’Andrea, Dual mechanism of intercellular commu-
nication in HOBIT osteoblastic cells: a role for gap-junctional hemi-
channels, J. Bone Miner. Res. 16 (2001) 1465.
 N.R. Jørgensen, S.T. Geist, R. Civitelli, T.H. Steinberg, ATP- and gap
junction-dependent intercellular calcium signaling in osteoblastic cells, J.
Cell Biol. 139 (1997) 497.
 N.R. Jørgensen, Z. Henriksen, C. Brot, E.F. Eriksen, O.H. Sorensen, R.
Civitelli, T.H. Steinberg, Human osteoblastic cells propagate intercellular
calcium signals by two different mechanisms, J. Bone Miner. Res. 15
 N.R. Jorgensen, S.C. Teilmann, Z. Henriksen, R. Civitelli, O.H.
Sorensen, T.H. Steinberg, Activation of L-type calcium channels is
required for gap junction-mediated intercellular calcium signaling in
osteoblastic cells, J. Biol. Chem. 278 (2003) 4082.
 Z. Li, Z. Zhou, C.E. Yellowley, H.J. Donahue, Inhibiting gap junctional
intercellular communication alters expression of differentiation markers
in osteoblastic cells, Bone 25 (1999) 661.
 B.L. Upham, J. Suzuki, G. Chen, Y. Wang, L.R. McCabe, C.C. Chang,
V.A. Krutovskikh, H. Yamasaki, J.E. Trosko, Reduced gap junctional
intercellular communication and altered biological effects in mouse
osteoblast and rat liver oval cell lines transfected with dominant-negative
connexin 43, Mol. Carcinog. 37 (2003) 192.
 B. Gramsch, H.D. Gabriel, M. Wiemann, R. Grummer, E. Winterhager,
D. Bingmann, K. Schirrmacher, Enhancement of connexin 43 expression
increases proliferation and differentiation of an osteoblast-like cell line,
Exp. Cell Res. 264 (2001) 397.
 T. Kojima, D.C. Spray, Y. Kokai, H. Chiba, Y. Mochizuki, N. Sawada,
Cx32 formation and/or Cx32-mediated intercellular communication
induces expression and function of tight junctions in hepatocytic cell
line, Exp. Cell Res. 276 (2002) 40.
 R. Huang, Y. Lin, C.C. Wang, J. Gano, B. Lin, Q. Shi, A. Boynton, J.
Burke, R.P. Huang, Connexin 43 suppresses human glioblastoma cell
growth by down-regulation of monocyte chemotactic protein 1, as
discovered using protein array technology, Cancer Res. 62 (2002) 2806.
 S. Fritz, L. Kunz, N. Dimitrijevic, R. Grunert, C. Heiss, A. Mayerhofer,
Muscarinic receptors in human luteinized granulosa cells: activation
blocks gap junctions and induces the transcription factor early growth
response factor-1, J. Clin. Endocrinol. Metab. 87 (2002) 1362.
 S.M. Nicholson, D. Gomes, B. de Nechaud, R. Bruzzone, Altered gene
expression in Schwann cells of connexin32 knockout animals, J. Nucl.
Med. 66 (2001) 23.
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 8180
 C. Vozzi, S. Ulrich, A. Charollais, J. Philippe, L. Orci, P. Meda,
Adequate connexin-mediated coupling is required for proper insulin
production, J. Cell Biol. 131 (1995) 1561.
 D.A. Iacobas, E. Scemes, D.C. Spray, Gene expression alterations in
connexin null mice extend beyond the gap junction, Neurochem. Int. 45
 J.P. Stains, R. Civitelli, Gap junctions regulate extracellular signal-
regulated kinase signaling to affect gene transcription, Mol. Biol. Cell 16
 J.P. Stains, F. Lecanda, J. Screen, D.A. Towler, R. Civitelli, Gap
junctional communication modulates gene transcription by altering the
recruitment of Sp1 and Sp3 to connexin-response elements in osteoblast
promoters, J. Biol. Chem. 278 (2003) 24377.
 L.I. Plotkin, S.C. Manolagas, T. Bellido, Transduction of cell survival
signals by connexin-43 hemichannels, J. Biol. Chem. 277 (2002) 8648.
 L.I. Plotkin, J.I. Aguirre, S. Kousteni, S.C. Manolagas, T. Bellido,
Bisphosphonates and estrogens inhibit osteocyte apoptosis via distinct
molecular mechanisms downstream of extracellular signal-regulated
kinase activation, J. Biol. Chem. 280 (2005) 7317.
 J. Ilvesaro, K. Vaananen, J. Tuukkanen, Bone-resorbing osteoclasts
contain gap-junctional connexin-43, J. Bone Miner. Res. 15 (2000) 919.
 J. Ilvesaro, P. Tavi, J. Tuukkanen, Connexin-mimetic peptide Gap 27
decreases osteoclastic activity, BMC, Musculoskelet. Disord. 2 (2001) 10.
 M. Ransjo, J. Sahli, A. Lie, Expression of connexin 43 mRNA in
microisolated murine osteoclasts and regulation of bone resorption in
vitro by gap junction inhibitors, Biochem. Biophys. Res. Commun. 303
 J. Zhang, C. Niu, L. Ye, H. Huang, X. He, W.G. Tong, J. Ross, J. Haug,
T. Johnson, J.Q. Feng, S. Harris, L.M. Wiedemann, Y. Mishina, L. Li,
Identification of the haematopoietic stem cell niche and control of the
niche size, Nature 425 (2003) 836.
 L.M. Calvi, G.B. Adams, K.W. Weibrecht, J.M. Weber, D.P. Olson, M.C.
Knight, R.P. Martin, E. Schipani, P. Divieti, F.R. Bringhurst, L.A. Milner,
H.M. Kronenberg, D.T. Scadden, Osteoblastic cells regulate the
haematopoietic stem cell niche, Nature 425 (2003) 841.
 F. Villars, B. Guillotin, T. Amedee, S. Dutoya, L. Bordenave, R. Bareille,
J. Amedee, Effect of HUVEC on human osteoprogenitor cell differen-
tiation needs heterotypic gap junction communication, Am. J. Physiol.:
Cell Physiol. 282 (2002) C775 – C785.
 B. Guillotin, C. Bourget, M. Remy-Zolgadri, R. Bareille, P. Fernandez,
V. Conrad, J. Medee-Vilamitjana, Human primary endothelial cells
stimulate human osteoprogenitor cell differentiation, Cell. Physiol.
Biochem. 14 (2004) 325.
 H.V. van Rijen, M.J. van Kempen, S. Postma, H.J. Jongsma, Tumour
necrosis factor alpha alters the expression of connexin43, connexin40,
and connexin37 in human umbilical vein endothelial cells, Cytokine 10
 C. Vitiello, P. D’Adamo, F. Gentile, E.M. Vingolo, P. Gasparini, S. Banfi,
A novel GJA1 mutation causes oculodentodigital dysplasia without
syndactyly, Am. J. Med. Genet., A 133 (2005) 58.
J.P. Stains, R. Civitelli / Biochimica et Biophysica Acta 1719 (2005) 69 – 81 81