Disuse sufficient to cause bone loss increases osteocytes’ expression of sclerostin but has no effect on osteocyte RANKL

The recent report that, in mice at least, osteocytes are a major source of RANKL has led to the inference that it is by this means that these mechanically-responsive cells influence osteoclast recruitment, differentiation and activation in the processes of mechanically-related control of bone mass and architecture. This assumption appears to be supported by the finding that mice lacking RANKL do not show bone loss with tail suspension.
We attempted to relate changes in expression of RANKL and sclerostin in osteocytes with disuse-related bone loss associated in tibias of mice in limbs paralysed by unilateral sciatic neurectomy.
Seventeen week old female C57BL/6 mice underwent right sciatic neurectomy and left sham operation. They were sacrificed in groups; for qRT-PCR analysis for RANKL and SOST on day 3 (n=7) and day 14 (n=8), immunolocalisation of RANKL and sclerostin (n=4, day 4) and for microCT assessment of bone loss (n=8, day 21).
As expected, disuse as a consequence of sciatic neurectomy was associated with a significant loss of bone both in the proximal cancellous region (BV/TV -44.5%, p<0.001) and in the cortical compartment (-18.5% p<0.001 at 37% of their length from their proximal ends). Disuse had no effect on the periosteally enclosed area at this level (+2.7%, p=0.09), while cortical bone area was reduced (-11.6% p<0.001) and medullary area expanded (+29.3% p<0.001). The number of sclerostin immunopositive osteocytes was increased 4 days after neurectomy (+35.8% p<0.01) and osteocyte sclerostin mRNA expression elevated (+118% p<0.05) at day 14. In contrast, RANKL mRNA expression was unchanged between control and paralysed limbs at 3 and 14 days (+8.3%, p=0.76 and -21.2%, p=0.23). RANKL protein immunolocalisation was similarly unaffected (+4.2%, p=0.38 at 4 days).
These data support previous studies showing that osteocytes respond to reduced loading in their surrounding matrix by increasing secretion of sclerostin.
The absence of any changes in RANKL with disuse suggests that RANKL is not involved in this resorption response, an observation that is incompatible with the view that RANKL derived from osteocytes is a necessary or universal influence in the recruitment, differentiation or activity of osteoclasts.