Little is known on how sensory nerves and osteoclasts affect degenerative processes in subchondral bone in osteoarthritis (OA). Substance P (SP) effects on bone are ambivalent but physiological levels are critical for proper bone quality whereas α-calcitonin gene-related peptide (αCGRP) has anabolic effects. Here, we aimed to analyse the influence of an altered sensory neuropeptide microenvironment on subchondral bone in murine OA. Transection of the medial meniscotibial ligament (DMM) of the right hind leg induced joint instability leading to development of OA. Subchondral bone of tibiae from wildtype (WT), alendronate-treated WT (ALN, osteoclast inhibition), αCGRP- and SP- (Tachykinin (Tac)1) knockout mice was analysed by micro-computed tomography 4 and 12 weeks after DMM or sham surgery. Bone resorption marker CTX-I was measured in serum.

We observed osteophytosis in all DMM groups and ALN sham mice 4 weeks after surgery but also in sham groups 12 weeks after surgery. In subchondral bone, bone volume density (BV/TV) increased from 4 to 12 weeks after surgery in DMM WT and Tac1-/− mice. DMM WT mice additionally had increased trabecular numbers (Tb.N.) and decreased trabecular space (Tb.Sp.) over time. Sham mice also showed time-dependent alterations in subchondral bone. In sham WT and αCGRP-/− mice specific bone surface (BS/BV) decreased and trabecular thickness (Tb.Th.) increased from 4 to 12 weeks after surgery while subchondral BV/TV of αCGRP-/− mice increased. Comparison of subchondral bone parameters at each time point showed elevated BV/TV in ALN DMM compared to WT DMM mice 4 weeks after surgery. In addition, both ALN sham and DMM mice showed a reduced BS/BV compared to WT. 4 weeks after sham surgery Tb.Th. was highest in ALN mice. In DMM WT mice Tb.Sp. was higher compared to ALN and αCGRP-/−. 12 weeks after surgery (late OA stage), BS/BV of ALN sham mice was significantly reduced in relation to ALN DMM, WT and Tac1-/− sham, while Tb.Th. increased compared to WT. DMM significantly decreased Tb.N. and increased Tb.Sp. in Tac1-/− compared to sham 12 weeks after surgery. CTX-I concentrations were significantly higher in ALN compared to Tac1-/− mice 4 weeks after sham surgery. 12 weeks after sham surgery CTX-I concentrations of WT mice were increased compared to αCGRP-/− and Tac1-/− mice.

Over time, DMM induced stronger changes in subchondral bone of WT mice compared to knockout strains. WT and αCGRP-/− sham mice also show alterations in bone parameters over time indicating age-related effects on bone structure. SP deficiency enhanced DMM-induced structural bone alterations in late stage OA emphasizing the importance of SP under pathophysiological conditions. Osteoclast inhibition with alendronate proved to be preservative for time-dependent changes of subchondral bone observed in both, DMM and sham mice. Interestingly, ALN treatment did not reduce bone turnover marker CTX-I, and additionally promoted early osteophyte formation in sham mice.

Previously, we have demonstrated reduced biomechanical bone strength and matrix quality in Tachykinin (Tac)1-deficient mice lacking the sensory neuropeptide substance P (SP). A similar distortion of bone microarchitecture was described for α-calcitonin gene-related pepide (α-CGRP)-deficient mice. In previous studies we observed alterations in cell survival and differentiation capacity of bone cells isolated from wildtype mice when stimulated with SP and α-CGRP. We assume that changes in sensory neurotransmitter balance modulate bone cell metabolism thereby possibly contributing to inferior bone quality. In order to explore this hypothesis, we investigated and compared metabolic parameters in osteoblasts and osteoclasts isolated from SP- and α-CGRP-deficient mice and wildtype (WT) controls.

Bone marrow-derived macrophages (BMMs) and osteoblast-like cells from female C57Bl/6J (WT-control), Tac1-deficient (Tac1-/−) and α-CGRP-deficient (α-CGRP-/−) mice were isolated and differentiated according to established protocols (Niedermair et al., 2014). Cell metabolism studies were performed for enzyme activity and cell survival.

We observed reduced numbers of BMM from Tac1-/− and α-CGRP-/− mice after initial seeding compared to WT but no changes in viability. Osteoblast-like cells from Tac1-/− mice tend to migrate out faster from bone chips compared to WT-controls whereas migration of osteoblast-like cells from α-CGRP-/− mice was not affected. Osteoblasts and osteoclast/BMM cultures from WT mice endogenously synthesize and secrete SP as well as α-CGRP at a picomolar range. We found no changes regarding BMM or osteoblast proliferation from both, Tac1-/− and α-CGRP-/− mice when compared to WT-controls. Caspase 3/7-activity was reduced by trend in osteoclast/BMM cultures of α-CGRP-/− mice and significantly reduced in osteoclast/BMM cultures of Tac1-/− mice compared to WT-controls. We found significantly higher Caspase 3/7-activity in osteoblasts of Tac1-/− mice after 14 days of osteogenic culture conditions when compared to WT-controls whereas osteoblasts of α-CGRP-/− mice were unaffected. Cathepsin K enzyme activity was significantly reduced in osteoclast/BMM cultures of Tac1-/− and α-CGRP-/− mice compared to WT-controls. ALP activity of Tac1-/− osteoblasts was higher after 7 days and reduced after 21 days of osteogenic culture compared to WT-controls whereas ALP activity of osteoblasts of α-CGRP-/− mice was unchanged.

Acccording to our in vitro observations, we suggest some reduction in bone resorption rate but concomitantly a reduction in bone formation rate in Tac1-/− mice compared to WT-controls resulting in a net bone loss in these mice as bone resorption is faster than bone formation. Furthermore, we assume that bone resorption rate is slightly reduced in α-CGRP-/− mice but bone formation rate seems to be unchanged. Therefore we hypothesize that additional conditions present in vivo might contribute to the inferior bone properties of α-CGRP-/− mice.