Strain uses gap junctions to reverse stimulation of osteoblast proliferation by osteocytes

Identifying mechanisms by which cells of the osteoblastic lineage communicate in vivo is complicated by the mineralised matrix that encases osteocytes, and thus, vital mechanoadaptive processes used to achieve load‐bearing integrity remain unresolved. We have used the coculture of immunomagnetically purified osteocytes and primary osteoblasts from both embryonic chick long bone and calvariae to examine these mechanisms. We exploited the fact that purified osteocytes are postmitotic to examine both their effect on proliferation of primary osteoblasts and the role of gap junctions in such communication. We found that chick long bone osteocytes significantly increased basal proliferation of primary osteoblasts derived from an identical source (tibiotarsi). Using a gap junction inhibitor, 18β‐glycyrrhetinic acid, we also demonstrated that this osteocyte‐related increase in osteoblast proliferation was not reliant on functional gap junctions. In contrast, osteocytes purified from calvarial bone failed to modify basal proliferation of primary osteoblast, but long bone osteocytes preserved their proproliferative action upon calvarial‐derived primary osteoblasts. We also showed that coincubated purified osteocytes exerted a marked inhibitory action on mechanical strain–related increases in proliferation of primary osteoblasts and that this action was abrogated in the presence of a gap junction inhibitor. These data reveal regulatory differences between purified osteocytes derived from functionally distinct bones and provide evidence for 2 mechanisms by which purified osteocytes communicate with primary osteoblasts to coordinate their activity.

Osteocytes are confined to lacunae, however, and can make little if any direct contribution to the architectural adaptive bone (re)modelling activities that load-related strains might stimulate. It is assumed therefore that their influence is achieved via their control of the remodelling activity of osteoclasts (via osteoblasts and lining cells) and osteoblasts on the bone surface. A potential route by which osteocytes could influence the behaviour of overlying osteoblasts in response to external mechanical stimuli is via the passage of small molecules through the osteoblast: osteocyte network of gap junctions or via molecules secreted into the intralacunar fluid. This fluid bathes osteocytes and the bone-facing processes of osteoblasts and lining cells, and its movement through canaliculae results from the pressure differentials induced by dynamic loads. The repetitive bending of the bone matrix is thought to generate a "pumping" action forcing fluid to the bone surface and subsequent dynamic shear strains on osteocytic processes. 10,[23][24][25][26][27][28] Previous studies have shown that gap junctions are expressed in all different types of bone cells [23][24][25][26][27][28][29][30][31][32][33][34][35] and are likely candidates for chemical information transfer between bone cells, 36,37 providing evidence that gap junction communications are potentially important in mechanotransduction. 35,[37][38][39][40][41][42] Recent studies have examined the effects of fluid shear applied to an osteocytic cell line derived from long bone (MLO-Y4) 35,37,41-47 and shown that at least some consequences of this stimulation can be transmitted via gap junctions to otherwise unstimulated osteoblasts. 48 Indeed, connexin 43 hemichannels have been postulated to serve a central function in fluid shear-induced PGE 2 release from the MLO-Y4 osteocytic cell line. 49 Moreover, other studies have demonstrated that fluid flow increases the gap junction expression and function in the osteocytic MLO-Y4 cells. 35,41,45 In addition, mechanical stimulation also results in the opening of connexin 43 hemichannels and release of PGE 2 50 and adenosine triphosphate 46 from the osteocytes.
Despite the attractive characteristics of these proposed models, there is, to our knowledge, little if any data that have demonstrated that purified osteocytes purified directly from different bones can exert any regulatory influence upon the behaviour of osteoblast.
Furthermore, the mechanisms coordinating these interactions either under basal conditions or in response to mechanical strain also remain the subject of some speculation.   Experimental strategy used. Eighteen-day-old Alizarin Red S-and Alcian Blue GX-stained chick embryo depicting the harvest sites for both tibiotarsal and calvarial primary osteoblasts and osteocytes. Primary osteoblasts from both tibiotarsal long bone (LOBs) and calvariae (COBs) were either cultured alone or cocultured with osteocytes. The latter were also derived from both bone sites and were cocultured in either "homotypic" (LOCs + LOBs or COCs + COBs) or "heterotypic" (LOCs + COBs or COCs + LOBs) conditions minute were counted using a 1214 Rackbeta liquid scintillation coun-

| Statistical analysis
Statistical analyses were performed using either Microsoft Excel or GraphPad Prism 6 (GraphPad Software, Inc., San Diego, California).
Data are presented as mean ± SEM and were considered statistically significant when P ≤ .05. A 2-sample, unpaired t test was used to compare means between control and treated groups.

| Morphologically characteristic phenotypes retained in vitro in purified osteocytes
Using scanning electron microscopy, we found that Ob7.3(5) + cells from both calvarial and tibiotarsal bones were generally smaller, exhibited a lower cytoplasmic area, had a distinct stellate appearance, and contained many more long slender processes radiating from a central cell body (Figure 2A (Table 1). Henceforth, Ob7.3(5) + osteocytes derived from calvarial or tibiotarsal long bones will be referred to as COC and LOC, respectively, and osteoblast-like cells as COB (calvarial) and LOB (long bone). 3.3 | Mechanical strain use of gap junctions to reverse proliferative influence of purified osteocytes on primary osteoblasts; gap junction-independent osteocyte-related osteoblast proliferation

| Stimulation of primary osteoblast proliferation through coculture with long bone purified osteocytes
We found, consistent with previous studies, that mechanical strain application increases proliferation of primary osteoblasts cultured alone ( Figure 4A). Surprisingly, however, we found that mechanical strain

FIGURE 3
Long bone osteocytes enhance calvarial osteoblast proliferation. To assess whether COBs are unable to respond to osteocyte-derived proliferative stimuli, COBs were maintained in heterotypic cultures with LOCs. LOCs induced proliferation of COBs, whilst COCs were unable to influence the proliferation of LOBs. This suggests the potential for fundamental differences signalling methods/signals derived from LOCs and COCs. Data are presented to show incorporation of 3 H-thymidine over an 18-h period and all 6 wells of a 6-well plate were used for each variable culture condition (n = 3 experiments in total). The asterisk denotes significance vs osteoblast monocultures (P < .05) behaviour to control remodelling activity to ensure mechanical competence, but evidence for this contention is sparse and unequivocal proof is lacking. A recent study performed in chick and mouse calvarial parietal bone shows, through FRAP analysis, the existence of cell-to-cell communication via gap junctions in the 3D morphology of the osteocyte network. 70 Our use of primary osteoblast-osteocyte cocultures, however, reveals additional relationships that would otherwise not be apparent in monoculture.
Using coculture, we sought evidence that purified osteocytes contribute to regulating primary osteoblast behaviour and whether any such contribution was reliant upon functional gap junctions. We found that postmitotic purified osteocytes were indeed capable of stimulating enhanced rates of proliferation by primary osteoblasts. Our data showed that osteocytes purified from mechanically responsive long bone, but not those from nonresponsive skull bones, exhibit a capacity to promote proliferation of primary osteoblasts. Using a pharmacological blocker (18β-glycyrrhetinic acid) under these basal conditions, we also show that this osteocyte-mediated promotion of proliferation of primary osteoblast was not dependent on functional gap junctions.
Further to blocking gap junctions, 18β-glycyrrhetinic acid has been reported to exert additional pharmacological actions including on 11β-hydroxysteroid dehydrogenase 1, pannexin channel activity as well as high-mobility group box protein 1 action, and glucocorticoid metabolism. [71][72][73][74] This is a limitation of our studies, and future examination of gap junction requires a more specific method of blocking these junctions. In marked contrast, we found that application of physiological levels of dynamic mechanical strain to cocultured long bone cells efficiently abrogated proliferation of primary osteoblast, with gap junction blockade indicating that strain-related transfer of an inhibitory stimulus between purified osteocytes and primary osteoblasts involves functional communicating gap junctions.
The Ob7.3(5) antibody, which was used herein to isolate osteocytes, has been shown to be specific for the phosphate-regulating gene with homology to endopeptidases on the X chromosome protein that is abundant in osteocytes. 52 We find that cells isolated from chick bone (tibiotarsal or calvarial) using this antibody exhibit low, nonsignif-  cells [66][67][68]75 ; however, it is possible that cell survival is also affected.
Although no evidence of an increased rate of cell death was observed in our experimental protocols, further studies to assess apoptosis directly would determine whether lack of proliferation might be associated with increased cell death.
The proproliferative influence of postmitotic osteocytes on cocultured osteoblasts is perhaps most readily interpreted, owing to the lack of apparent gap junction involvement, to suggest the involvement of an osteocyte-derived soluble mediator. There are many candidate soluble mediators that are produced by osteocytes, which may account for this osteoblast stimulation. These include secretory phospholipase A 2 , which evokes increased PGE 2 and prostaglandin I 2 release from osteoblasts in nonloaded long bone organ explant cultures 5 and factors such as transforming growth factor-β and nitric oxide, which are produced by osteocytes. [76][77][78] It is pertinent to emphasise that these proproliferative effects were limited to osteocytes derived from long bones but that those derived from calvariae, with distinct origins and protective rather than load-bearing functions, had no effect on osteoblasts derived from either source. This is a fascinating  84 Our studies allow additions to the growing evidence that osteocytes can promote osteoblast proliferation, that this influence can be restricted by the application of strain, and that only the latter, osteocyte strain-related control of osteoblast proliferation is dependent upon gap junctions ( Figure 5). 69 If our findings were to be directly extrapolated to the in vivo scenario, they would imply that, at rest, osteocytes act via transcellular signalling to maintain an active osteoblast population on the bone surface. Unlike the situation in culture, even at rest, osteocytes are constantly subjected to mechanical inputs, amongst which the most continuous is fluid shear strain stimuli driven by the circulatory system.
In stark contrast, however, these proliferative signalling molecules emanating from osteocytes must rather be completely "overruled" in response to mechanical loading of bones, by information transferred to osteoblasts via gap junctions to promote an appropriate (re)modelling event. This implies that there is continual osteocyte-derived signalling to cells on the bone surface maintaining the delicate balance of formation and resorption.
It is worth emphasising that the application of the mechanical strain stimulus is only transient in our studies, and yet this is nevertheless sufficient to significantly restrict the proliferation of osteoblasts induced by osteocytes. We interpret our findings to reflect the mechanism by which osteocytes and mechanical inputs together act to regulate the proliferation of osteoblasts. Thus, with increased loading, inhibition of proliferation would have to precede osteoblast differentiation at locations where new bone is required to withstand increased mechanical demands. An alternative interpretation is that osteocytes exert some hitherto unresolved suppression of the increases in proliferation that normally ensue periods of loading. In conclusion, our studies suggest that purified osteocytes, derived from load-bearing long bones, exert a direct proproliferative influence upon primary osteoblasts and that mechanical strain may use gap junctions to reverse this osteocytederived stimulatory effect.